write short answer

timer Asked: Jul 4th, 2018
account_balance_wallet $5

Question description

read the cases in the file attached then answer (each case separate) :

1 Case from Crawford, Chapter 3 page 88

What is the current PIC (product innovation charter)?

How did Kellogg's PIC seem to have evolved from 1999 to 2006?

2 Case from Crawford, chapter 3, page 90.

What might the PIC (product innovation charter) have been for the Element?

What benefits resulted from bringing in the Voice of the Customer?

3. “I saw the other day where fi lmmakers (large ones as well as small ones) are

fi nding profi ts in low-budget fi lms. It seems they aim for narrow, but very

reachable market segments (e.g., young kids), and they use standard fi lmmak-

ing technologies but use only what they call emerging actors and directors

(meaning cheap now). They try hard to capture the interests of their core target

group, and they mean it when they say low-budget. I also read where several

of them are trying to move out rapidly from the core when they have a win-

ner: little kids, bigger kids, and so on. They think this approach yields the best

return on investment even though it causes them to miss out on the occasional

blockbuster winner. Some of these low-budget specials included American Pie,

There’s Something About Mary, and The Wedding Singer. That last one focused

on boys and men, but they added a love story line with Drew Barrymore that

brought women in too. Now, can you fi t all that into what might be the PIC of

these fi lms? What are the negatives of this approach?” 49

4 From chapter 4, answer this question: Describe the Need, Form and Function of the NEST thermostat.

New Products Management This page intentionally left blank New Products Management Eleventh Edition Merle Crawford University of Michigan—Emeritus Anthony Di Benedetto Temple University NEW PRODUCTS MANAGEMENT, ELEVENTH EDITION Published by McGraw-Hill Education, 2 Penn Plaza, New York, NY 10121. Copyright © 2015 by McGraw-Hill Education. All rights reserved. Printed in the United States of America. Previous editions © 2011, 2008, and 2006. No part of this publication may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the prior written consent of McGraw-Hill Education, including, but not limited to, in any network or other electronic storage or transmission, or broadcast for distance learning. Some ancillaries, including electronic and print components, may not be available to customers outside the United States. This book is printed on acid-free paper. 1 2 3 4 5 6 7 8 9 0 DOC/DOC 1 0 9 8 7 6 5 4 ISBN 978-0-07-802904-2 MHID 0-07-802904-x Senior Vice President, Products & Markets: Kurt L. Strand Vice President, General Manager, Products & Markets: Michael Ryan Vice President, Content Production & Technology Services: Kimberly Meriwether David Brand Manager: Sankha Basu Development Editor: Kelly Delso Marketing Manager: Donielle Xu Director, Content Production: Terri Schiesl Content Project Manager: Melissa M. Leick Buyer: Nichole Birkenholz Cover Designer: Studio Montage, St. Louis, MO Front Cover Image: Digital Tablet: Blend Images/Jetta Productions/Getty Images, Male Designer: Aping Vision/ STS/Getty Images, Businesswoman: Stockbyte/Getty Images, African Female Scientist: ERproductions Ltd/Blend Images LLC, Two Businesspeople: © Sam Edwards/age fotostock, Business Meeting: © Chris Ryan/age fotostock Back Cover Image: Author owned Typeface: 10/12 Palatino Compositor: MPS Limited Printer: R.R. Donnelley All credits appearing on page or at the end of the book are considered to be an extension of the copyright page. Library of Congress Cataloging-in-Publication Data Crawford, C. Merle (Charles Merle), 1924New products management / Merle Crawford, University of Michigan-Emeritus, Anthony Di Benedetto, Temple University. — Eleventh edition. pages cm ISBN 978-0-07-802904-2 (alk. paper) 1. New products—Management. I. Di Benedetto, C. Anthony. II. Title. HF5415.153.C72 2015 658.5'75—dc23 www.mhhe.com 2013042977 Remembering Dr. C. Merle Crawford (1924–2012) PROFESSOR, UNIVERSITY OF MICHIGAN Merle Crawford created this pioneering textbook and was its sole author for the first five editions, published in 1983, 1987, 1991, 1994, and 1997. Beginning with the sixth edition, he invited C. Anthony Di Benedetto to join him as co-author. The frequency of publication of the book’s editions increased as the need for skills in new products management became recognized, and the book has become a foundation for this emerging profession, appealing to both managers and students alike. Merle joined the faculty of the University of Michigan as an associate professor of marketing in 1965 after rising to the position of marketing director at Mead Johnson and Company and serving in other roles in that company for a decade. He retired from Michigan in 1992, taking his final year as a sabbatical leave. The combination of work in industry and in academe gave him a unique and valuable perspective that ultimately led him to become the primary force in the creation of the Product Development & Management Association in 1976. Simply put, Merle felt that academic researchers could do their best work when addressing challenges faced by leading managers, and that those leading managers would benefit from dialogue with leading academics. Merle served the PDMA in many capacities, beginning as its charter president for two years (1977 and 1978) while doubling in the role of secretary-treasurer for both years, collecting dues and publishing occasional newsletters. He operated the association from his university office as a volunteer from PDMA’s founding until 1984 and also served in many other capacities including VP research (1984 and 1985), VP publications (1989), and as a director in seven years of the eight years that followed (1987–1988 and 1990–1994). During this time Merle was joined by many others, many of whom he enlisted, to help the PDMA develop into the world’s leading association of new product development professionals with membership exceeding 1,700 by the end of his final term as a director. Merle worked to create the first regularly scheduled executive development program in new products management offered by a university in the United States, recruiting colleagues from the PDMA and Michigan and other universities to join the program’s charter faculty. The course began as a three-day program, offered quarterly, beginning in 1979. Due to its popularity, it soon transformed into a five-day program that continued for over 20 years. This course, along with subsequent growth of the PDMA, encouraged Merle to write the first edition of this book, published in 1983. The book was used by managers and executives in the course, making that program a living laboratory for extending the ideas he included in its pages. Merle’s research documented actual new product failure rates following launch in 1979, helping to demolish the resilient myth that “most new v vi Remembering Dr. C. Merle Crawford (1924–2012) products fail.” He refined his original concept of the Product Innovation Charter through discussions with program participants and colleagues, and he created concise templates for product concept statements and his original depiction of the product protocol. These emerged as fully defined tools in his textbook and have become foundation skills for today’s professionals. Merle continued to refine them in subsequent editions of this book. Another of his enduring gifts to our profession was his development of a comprehensive glossary of new product terms, first published in the third edition of his book and continuing as the foundation for the glossary now presented in three languages on the PDMA’s website. The PDMA honored his role as founder and champion of innovation with the creation of the order of Crawford Fellows of Innovation, appointing Merle the first Fellow in 1991. Appointment remains a highly exclusive honor, with four Fellows subsequently appointed to receive this honor. Merle was skilled in building bridges to connect new products professionals who may have come to the profession from diverse backgrounds. In the first edition of this book’s introduction he wrote, “. . . the new products process is both an art and a science. That process demands creativity and emotional commitment, but it also allows and rewards thorough and sophisticated analysis. Both aspects of the new products process are emphasized here.” Those words were powerful and framed a foundation for the growth our profession enjoys today. Perhaps one thing would have disappointed him. He defined new products inclusively, including both tangible goods and intangible services. Today, common usage separates these forms with the word product restricted to tangible goods, separate from services. He might smile, though, at the irony that most of us realize that there are very few tangible products that lack a significant intangible service component (including customer support), and intangible services that lack any sort of tangible element, whether taking the form of a physical location, accessible Web site, or supportive elements of form and technology that help to frame their successful delivery. Merle passed away on November 11, 2012. He was a friend and champion of innovation. We owe him a great debt. Thomas P. Hustad August 7, 2013 About the Authors Merle Crawford was Professor of Marketing (Emeritus) at the University of Michigan, where he taught from 1965 until his retirement in 1992. Prior to his appointment at Michigan, he was Marketing Director at Mead Johnson & Co. Professor Crawford was an original member of the Product Development & Management Association from its founding in 1976, and he served as the charter president from 1977 to 1978 and on the Board of Directors until 1994. He authored the first edition of the groundbreaking textbook New Products Management, published in 1983 and still widely used by managers, executives, and business students. Anthony Di Benedetto is Professor of Marketing and Supply Chain Management and Senior Washburn Research Fellow at Temple University, Philadelphia, PA, and Professor of High-Tech Entrepreneurial Marketing, TU Eindhoven, The Netherlands. He has held visiting professorships worldwide, lecturing on product development and management. He was named one of the 50 leading research scholars worldwide in Innovation and Technology Management by the International Association of Management of Technology. Professor Di Benedetto served as Editor-in-Chief of the Journal of Product Innovation Management for nine years. vii Preface New products have always been of interest to both academics and practitioners, and organized, college-level instruction on the subject of new products management traces to the 1950s. By the 1990s, a new products management discipline had evolved. The Product Development & Management Association (PDMA) has flowered to close to 3,000 members in some 50 countries around the world, and there are over 20 local chapters in the United States alone, plus international affiliates in a dozen countries. Over 300 colleges have courses on the subject of new products, and the field’s journal, the Journal of Product Innovation Management, has a 30-year track record of publication. The job title of new products manager or director is becoming much more common and is offering much earlier entry than 15 or 20 years ago; we also see the emergence of higher level positions for careers to build to. The PDMA now offers a practitioner certification (New Product Development Professional, or NPDP), recognizes the best product developing firms (with its Outstanding Corporate Innovator award), and has been able to do what those in many fields have not, that is, merge the thinking and activity of professors and practitioners. Information on the PDMA can be found at www.pdma.org. How This Book Views the Field of New Products Management Such exploding growth means that we still take a variety of approaches to the teaching of the new products subject—marketing, technical, creative, design, and so on. This book provides the management approach with the perspective of marketing. In every organization (industry, retailing, government, churches, and any other kind of institution) there is a person or group of persons who, knowingly or unknowingly, are charged with getting new goods and services onto the market. More and more today, those people are new products managers, or project managers, or team leaders. They lead a multifunctional group of people, with the perspective of a general manager, operating as a company within a company. They must deal with the total task—strategy, organization, concept generation, evaluation, technical development, marketing, and so on. They are not finished with their work until the new product has achieved the goals assigned to the team—this usually means some form of sales or profit, and certainly means the task is not finished when the new product is put onto the shipping dock. We try to avoid a functional myopia, and it is rare today to hear that “Marketing tells everyone what to do” or “R&D runs our new products activity.” When a functional specialist is assigned leadership of a new products team, that person must learn the general manager viewpoint, but one usually has to succeed as a functional member of new products teams before getting a shot at being a team leader. Marketing people, working as team members or as team leaders, need the types of information in this book. viii Preface ix Some Basic Beliefs That Guided the Writing People who have used the first 10 editions of this book know its unique viewpoints on the subject. But for newcomers, and of course all students are newcomers, here are some of them. 1. Product innovation is one single operation in an organization. It has parts (strategy, teams, plans, etc.) but they are all just parts. Any operation that runs as separate pieces misses the strength of the whole. 2. The field is still new enough that it lacks a systematic language. This makes it very difficult for students, who are accustomed to studying subjects where a term means one thing, and only that one thing. We use all product terms consistently throughout the book, and we urge students to use them. Naturally, new terms come and go; some survive and some don’t. Because of the terminology problem in a rapidly growing field, every term that might require definition has been made bold the first time it is used, and the index directs the reader to that section. We don’t include a glossary, but a useful one is available at the Product Development & Management Association Web site. 3. Ideas learned without application are only temporary residents in your mind. To become yours, a concept must be applied, in little ways or in big ones. Thus, the book is peppered with applications, short cases, and other opportunities for using the concepts studied. Projects are encouraged in the Instructor’s Manual. There are many examples from the business world, and up-to-date references on all important topics. 4. As much as we would like them and have diligently tried to find them, we believe there is no standard set of procedures for product innovators, nor particular sets for makers of consumer packaged goods, or of consumer durables, industrial goods, services, and so on. Like a marketing plan, there is a best plan for any particular situation. A manager must look at a situation and then compile a set of tools and other operations appropriate to that situation. All large firms use scores of different approaches, not one. 5. Next, there is the halo effect, which is a problem in the field of new products. The halo effect shows in the statement, “It must be a good thing for us to do—Apple does it, or GE does it, or Honda does it.” Those are excellent companies, but one reason they’re good is they spend lots of time and money studying, learning from others. They have huge training programs in product innovation and bring in every expert who appears on the scene with what looks like a good new products management idea. They assume everything they do is wrong and can be improved. You should too. This book does. Citations of their actions are given as examples, not recommendations. These well-known firms have many divisions and hundreds of new products under development at any one time. Managers there can’t know what other managers are doing, nor do they care, in the prescriptive sense. Each group aims to optimize its situation, so they look around, see what others in comparable situations are doing (inside and outside their firm), and pick and x Preface choose to fit the situation. To the extent there are generalizations (e.g., there should be some form of strategy), these will stand out as you work your way through the course. But what strategy, and exactly how should one determine it—that is situational. 6. An example of this lies in rejection of the belief that new products strategy should rest on the base of either technology or market. This choice has been argued for many years. But most firms seek to optimize on both, a dual-drive strategy. Of course, true to the previous point, firms will build on one or the other if the situation seems to fit—for example, DuPont’s platform program to find applications for the superstrength fabric Kevlar, or auto components firms that rely on process development engineering to better meet the needs of original equipment manufacturers. And yet, DuPont works to advance that technology, and the components firms are evolving their own research and development operations. 7. We believe that students should be challenged to think about concepts they have been introduced to. This book contains lists of things from time to time, but such lists are just a resource for thinking. The above belief about the best approach being situational is based on the need to analyze, consider, discuss, apply. The great variety in approaches used by businesspeople is not a testimony to ignorance, but to thinking. On a majority of the issues facing us today, intelligent people can come down with different views. Decisions are the same—they are not necessarily right or wrong at the time they are made. Instead, the manager who makes a decision then has to work hard to make that decision turn out right. The quality of the work is more important than the quality of the decision. An example of this phenomenon is the sadness we feel when a manager says, “We’re looking for the really great idea.” Managers of product innovation make ideas great—they don’t come that way. 8. Last, we have tried to implement more clearly the view that two things are being developed—the product and the marketing plan. There are two development processes going on in tandem. Marketing strategy begins at the very start and runs alongside the technical work and beyond it. Changes in the Eleventh Edition Past adopters of New Products Management will notice major changes in this edition. While there are some changes in virtually every chapter, some of the most substantial changes are as follows: 1. We have made major additions and updates to the cases to provide more plentiful and more current examples. We retired several cases from the previous edition, wrote many new cases, and thoroughly updated many others. New cases for this edition include: corporate strategy at LEGO (Chapter 2), open innovation at Pillsbury (Chapter 4), positioning and competition in the smartphone industry (Chapters 6, 9, and 16), the turnaround at Domino’s Pizza (Chapter 9), product protocol development at Fisher & Paykel (Chapter 12), business product development at DuPont (Chapter 12), corporate Preface xi culture at Provo Craft (Chapter 14), and the development of the Ford Fusion (Chapter 14). Substantial upgrades were also made to the following cases: Rubbermaid (Chapter 7), WiLife (A) and (B), now listed as Logitech (Chapters 10 and 12), and Gillette (Chapter 13). We also kept the Palm Pilot case (Chapter 13); though this case describes an older product, it is such a great illustration of how to respond to competitive challenges and customer requirements through excellent design that we leave it in as a good basis of discussion. As always, we aim to offer a mix of high-tech products and consumer products and services in the set of cases. 2. In addition, we have substantially updated examples throughout the text wherever possible. We try to make use of illustrative examples that will resonate with today’s students wherever possible. For example, we use the Hapifork (a fork that monitors one’s eating speed and provides health benefits) as an example of how to bring a product concept to product form, and how to make trade-offs and decisions at product protocol development. (Also, when trying this case out in class, it seems to stimulate a discussion of whether the claims made by the manufacturer are believable, making it a good illustration of “pre-use sense reaction” in product use testing.) Of course, we welcome the reader’s comments and suggestions for improvement. 3. There continues to be much new research in new products, and we have tried to stay current on all of these topics. The newest Comparative Performance Assessment Study (CPAS), published by the PDMA, was released shortly before this revision and provided many new statistics reflecting the state of the art in product development. Readers will also notice new or expanded coverage of serial innovators, spiral development, portfolio management, voice of the customer, open innovation, user toolkits, social media, crowdsourcing, conjoint analysis, design-driven innovation, product forecasting, concept testing, launch planning, postlaunch management, product development for emerging markets, and sustainable design, among other topics. 4. We continue the practice of referencing Web sites of interest throughout the text, from the Product Development & Management Association and the Consumer Product Safety Commission, to Web sites referencing failed products or bad designs, and we have added several new ones. Rather than a collective list of sites, we chose to place each reference in a suitable context in the book. We have received positive feedback from users of the last edition, so we have not made changes to the basic 20-chapter format. We still use analytical models to integrate the stages of the new products process. As in previous editions, perceptual mapping is introduced early in the new products process, during concept generation, but its output may guide selection of attributes in a conjoint analysis task, and may later be used in benefit segmentation and product positioning. Conjoint analysis results may be used in concept generation or evaluation and may provide a set of desired customer attributes for house-of-quality development. The sequence of three smartphone end-of-chapter cases illustrates how the analytical models bind the new products process together. As in previous editions, many other concepts—Product Innovation Charter, A-T-A-R models, evaluation xii Preface techniques, the multifunctional nature of new products management—are also used to integrate topics horizontally throughout the text. Because this book takes a managerial focus and is updated extensively, it is useful to the practicing new products manager. It has been used in many executive education programs. Great pains have been taken to present the “best practices” of industry and offer footnote references to business literature. From the first edition, the ends of chapters do not have a list of questions. Rather, we have culled mainly from many conversations with students the questions and comments they received from business managers on their fly-backs. These comments are built into a conversation with the president of a conglomerate corporation. Explanation of how to use them is given at the end of Chapter 1. As always, effort has been aimed at making the book increasingly relevant to its users. We consider a text revision to be a “new product,” and thus an opportunity for us to become even more customer-oriented. Academic colleagues have made many thoughtful suggestions based on their experiences with previous editions and have provided much of the driving force behind the changes you see in this edition. We gratefully acknowledge all the reviewers who provided extensive comments and suggestions that were extremely helpful in this revision, as well as all the instructors and students who contacted us to make suggestions and correct errors. We are very excited about the changes to this new edition and sincerely hope they fit your needs. Online Resources The instructor will find plenty of online support for this text at the companion Web site, www.mhhe.com/crawford11e. Available on the Web site are an online Instructor’s Manual, a set of PowerPoint slides, a test bank, and exercises and cases that can be used to accompany the text materials. Some of these materials are also available to the students where appropriate. Dedication This edition is dedicated to Merle Crawford (1924–2012), Professor Emeritus (University of Michigan), sole author of the first five editions of this textbook, cofounder of the Product Development & Management Association, and influential marketing scholar and consultant in new product development. In the field of product development, the highest award conferred by the PDMA, in recognition of a lifetime of achievements and contributions to the field of product development, is the Crawford Fellow. Merle Crawford was the namesake of this prestigious award, and also its first recipient. The naming of this award in his honor speaks eloquently of the high regard and respect bestowed on Professor Crawford by his colleagues at the PDMA. With this dedication I would also like to personally express my thanks for the opportunity to be his writing partner on this textbook. A.D.B. Contents in Brief PART ONE PART FOUR Overview and Opportunity Identification/Selection 3 Development 319 13. Design 323 1. The Strategic Elements of Product Development 5 14. Development Team Management 351 2. The New Products Process 25 15. Product Use Testing 3. Opportunity Identification and Selection: Strategic Planning for New Products 60 PART FIVE 95 4. Creativity and the Product Concept 97 5. Finding and Solving Customers’ Problems 130 6. Analytical Attribute Approaches: Introduction and Perceptual Mapping 154 7. Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques 171 PART THREE Concept/Project Evaluation 191 8. The Concept Evaluation System 193 9. Concept Testing Launch 403 16. Strategic Launch Planning PART TWO Concept Generation 381 214 406 17. Implementation of the Strategic Plan 441 18. Market Testing 464 19. Launch Management 20. Public Policy Issues 492 515 APPENDIXES A Sources of Ideas Already Generated 543 B Other Techniques of Concept Generation 549 C The Marketing Plan D Guidelines for Evaluating a New Products Program 565 INDEX 559 569 10. The Full Screen 243 11. Sales Forecasting and Financial Analysis 262 12. Product Protocol 291 xiii Contents PART ONE The Phases in the New Products Process OVERVIEW AND OPPORTUNITY IDENTIFICATION/SELECTION 3 Chapter 1 The Strategic Elements of Product Development 5 Setting 5 The Importance of New Products 6 Globalization and New Product Development 9 How Product Development Is Different 11 What Is a New Product, and What Leads to Success? 14 Does This Field of Activity Have a Unique Vocabulary? 16 Does the Field of New Products Offer Careers? 17 The Strategic Elements of Product Development 18 The Basic New Products Process 19 The Other Strategic Elements 22 Product Development in Action 23 Summary 24 Applications 24 Chapter 2 The New Products Process 25 Setting 25 The Procter & Gamble Cosmetics Saga 25 The Product Innovation Charter (PIC) 26 The New Products Process 27 The New Product Portfolio 27 Supporting the Strategic Elements: Effective Team Management 28 What Happened in That Saga? 28 xiv 29 Phase 1: Opportunity Identification and Selection Phase 2: Concept Generation 31 Phase 3: Concept/Project Evaluation 31 Phase 4: Development 32 Phase 5: Launch 33 Evaluation Tasks Throughout the New Products Process 34 Speeding the Product to Market 37 Risks and Guidelines in Speeding to Market 40 What about New Services? 43 New-to-the-World Products 46 The Role of the Serial Innovator 48 Spiral Development and the Role of Prototypes 50 Closing Thoughts about the New Products Process 51 Summary 52 Applications 53 Case: Lego 53 Case: Tastykake Sensables 55 Case: The Levacor Heart Pump 57 Chapter 3 Opportunity Identification and Selection: Strategic Planning for New Products 60 Setting 60 A Product Strategy for a “Company within a Company” 61 New Product Strategy Inputs and Identifying Opportunities 61 Product Platform Planning 61 Opportunity Identification 66 Noncorporate Strategic Planning Miscellaneous Sources 70 The Product Innovation Charter Why Have a PIC? 72 68 70 30 Contents The Sections of the PIC 74 Background Section of the PIC 74 The Arena (Area of Focus) Section of the PIC Goals and Objectives Section of the PIC 77 Special Guidelines Section of the PIC 77 74 How to Prepare a Product Innovation Charter 80 Product Portfolio Analysis: The New Product’s Strategic Fit 82 Summary 87 Applications 87 Case: New Product Strategy at Kellogg 88 Case: The Honda Element 90 PART TWO CONCEPT GENERATION 95 Chapter 4 Creativity and the Product Concept 97 Setting 97 Preparation 97 The Product Innovation Charter 97 Finding the Right People 98 Management’s Role in Creativity 99 Activities to Encourage Creativity 101 Special Rewards 103 The Removal of Roadblocks 103 The Product Concept 104 The Designer Decaf Example 107 The Concept Statement 108 Two Basic Approaches 110 Important Sources of Ready-Made New Product Ideas 111 User Toolkits 111 Crowdsourcing 114 Lead Users 115 Open Innovation 118 Summary 123 Applications 123 Case: Pillsbury Grands! Biscuit Sandwiches 124 Case: P&G CarpetFlick 126 Case: Aquafresh White Trays 127 xv Chapter 5 Finding and Solving Customers’ Problems 130 Setting 130 The Overall System of Internal Concept Generation 130 Gathering the Problems 131 Internal Records 132 Direct Inputs from Technical and Marketing Departments 132 Problem Analysis 133 Scenario Analysis 141 Solving the Problems 145 Group Creativity 145 Brainstorming 145 Electronic Brainstorming and Computer-Assisted Creativity Techniques 146 Online Communities 147 Disciplines Panel 149 Concept Generation Techniques in Action 149 Summary 150 Applications 150 Case: Campbell’s IQ Meals 151 Case: Earning Organizational Respect 152 Chapter 6 Analytical Attribute Approaches: Introduction and Perceptual Mapping 154 Setting 154 Understanding Why Customers Buy a Product 154 Products Are Groups of Attributes 154 Analyzing Product Attributes for Concept Generation and Evaluation 156 Gap Analysis 156 Determinant Gap Maps 157 Perceptual Gap Maps Based on Attribute Ratings (AR) 158 Perceptual Gap Maps Based on Overall Similarities (OS) 164 Comments on Gap Analysis 167 Summary 168 Applications 168 Case: Comparing Smartphones (A) 169 xvi Contents Chapter 7 Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques 171 Setting 171 Trade-Off Analysis 171 Using Trade-Off Analysis to Generate Concepts 172 A Conjoint Analysis Application 173 Is Conjoint the Right Method? 177 Alternatives to Full-Profile Conjoint Analysis 178 Recent Modifications in Conjoint Analysis 179 Virtual Prototypes in Concept Testing 179 Qualitative Techniques 180 Dimensional Analysis 180 Checklists 182 Relationships Analysis 183 About the Dimensions Used in Relationships Analysis 183 Two-Dimensional Matrix 183 Morphological or Multidimensional Matrix 184 Analogy 186 Summary 187 Applications 187 Case: Rubbermaid 188 193 The Evaluation System for the Basic New Products Process 194 Product Line Considerations in Concept Evaluation 197 The Cumulative Expenditures Curve 198 Everything Is Tentative 201 Potholes 202 Summary 209 Applications 210 Case: Chipotle Mexican Grill 210 Case: Concept Development Corporation Chapter 9 Concept Testing Setting 214 The Importance of Up-Front Evaluations The Product Innovation Charter 215 Market Analysis 216 Initial Reaction 216 Concept Testing and Development 217 What Is a New Product Concept? 219 The Purposes of Concept Testing 219 Identifying Benefit Segments Joint Space Maps 230 Preference Regression 232 228 228 Conjoint Analysis in Concept Testing 233 Market Research to Support Concept Testing 235 Conclusions 238 Summary 238 Applications 239 Case: Domino’s 240 Case: Comparing Smartphones (B) 242 Chapter 10 The Full Screen 212 214 Analyzing Research Results Setting 193 What’s Going On in the New Products Process? 193 Planning the Evaluation System 201 205 Where Do We Get the Figures for the A-T-A-R Model? 209 Further Uses of the A-T-A-R Model 209 Prepare the Concept Statement 221 Define the Respondent Group 225 Select the Response Situation 226 Prepare the Interviewing Sequence 227 Variations 228 CONCEPT/PROJECT EVALUATION 191 The Risk/Payoff Matrix 199 The Decay Curve 200 The A-T-A-R Model 203 Considerations in Concept Testing Research 221 PART THREE Chapter 8 The Concept Evaluation System The People Dimension Surrogates 204 243 Setting 243 Purposes of the Full Screen 244 214 Contents The Scoring Model 246 Introduction to Scoring Models 246 The Screening Procedure 247 Profile Sheet 252 A Screening Model Based on Project NewProd 253 The Analytic Hierarchy Process 255 Special Aspects 258 Summary 258 Applications 259 Case: Logitech (A) 259 Chapter 11 Sales Forecasting and Financial Analysis 262 Setting 262 Sales Forecasting for New Products 263 Forecasting Sales Using Traditional Methods 264 Forecasting Sales Using Purchase Intentions 266 Forecasting Sales Using the A-T-A-R Model 267 Techniques for Forecasting Product Diffusion 269 Observations on Forecasting Models 271 Problems with Sales Forecasting 272 Summary of the Problems 273 Actions by Managers to Handle These Problems 274 Improve the New Product Process Currently in Use 274 Use the Life Cycle Concept of Financial Analysis 274 Reduce Dependence on Poor Forecasts 275 Return to the PIC 280 Summary 283 Applications 283 Case: Bay City Electronics Chapter 12 Product Protocol Setting 291 The Product Protocol 291 292 284 xvii Purposes of the Protocol 294 Protocol’s Specific Contents 296 Target Market 298 Positioning 298 Product Attributes 299 Competitive Comparisons and Augmentation Dimensions 301 Other Components of the Product Protocol 301 Protocol and the Voice of the Customer Hearing the Voice of the Customer 302 302 Protocol and Quality Function Deployment (QFD) 305 QFD and the House of Quality Outcomes of QFD 309 305 Some Warnings about the Difficulty of the Protocol Process 311 Summary 312 Applications 312 Case: Fisher & Paykel 313 Case: DuPont 315 Case: Logitech (B) 317 PART FOUR DEVELOPMENT 319 Chapter 13 Design 323 Setting 323 What Is Design? 324 Design-Driven Innovation 324 The Role of Design in the New Products Process 325 Contributions of Design to New Product Goals Product Architecture 325 331 A Process for Product Architecture 331 Product Architecture and Product Platforms Industrial Design and the Industrial Designer 333 Prototype Development 334 Managing the Interfaces in the Design Process 336 Improving the Interfaces in the Design Process 339 Computer-Aided Design and Design for Manufacturability 340 333 xviii Contents Continuous Improvement in Design 342 Summary 343 Applications 343 Case: The Mini 344 Case: Palm Pilot 345 Case: Gillette Mach3 and Fusion 348 Chapter 14 Development Team Management Is Product Use Testing Really Necessary? Are These Arguments Correct? Knowledge Gained from Product Use Testing 387 351 Setting 351 What Is a Team? 351 Structuring the Team 352 Another Look at Projectization Building a Team 355 356 Establishing a Culture of Collaboration 356 The Team Assignment and Ownership 357 Selecting the Leader 358 Selecting the Team Members 359 Roles and Participants 360 Network Building 363 Training the Teams 363 Managing the Team 364 Cross-Functional Interface Management 364 Overcoming Barriers to Market Orientation 367 Ongoing Management of the Team 368 Team Compensation and Motivation 368 Closing the Team Down 370 Virtual Teams 370 Managing Globally Dispersed Teams Summary 376 Applications 376 Case: Provo Craft 377 Case: Ford Fusion 378 Chapter 15 Product Use Testing 372 381 Setting 381 The Role of Marketing During Development 382 Pre-Use Sense Reactions 387 Early Use Experiences 388 Alpha and Beta Tests 388 Gamma Testing 390 Diagnostic Information 391 Decisions in Product Use Testing 391 Who Should Be in the User Group? 391 How Should We Reach the User Group? 392 Should We Disclose Our Identity? 393 How Much Explanation Should We Provide? 393 How Much Control over Product Use Should There Be? 393 How Should the Test Be Conducted? 394 Over What Time Period Should the Test Be Conducted? 395 What Should Be the Source of the Product Being Tested? 396 What Should Be the Form of the Product Being Tested? 396 How Should We Record Respondents’ Reactions? 397 How Should We Interpret the Figures We Get? 398 Who Should Do the Product Use Test? 398 Special Problems 399 Don’t Change the Data Just Because They Came Out Wrong 399 Be Alert to Strange Conditions 399 What If We Have to Go Ahead without Good Use Testing? 399 Summary 399 Applications 400 Case: Product Use Testing for New Consumer Nondurables 401 PART FIVE Marketing Is Involved from the Beginning of the Process 382 Marketing Ramp-Up, or the “I Think We’ve Got It” Phase 383 Why Do Product Use Testing? 384 384 385 LAUNCH 403 Chapter 16 Strategic Launch Planning Setting 406 406 Contents The Strategic Givens 407 Revisiting the Strategic Goals 408 Strategic Platform Decisions 409 Launch Tactics Alliances 450 A-T-A-R Requirements 451 Awareness 451 Stocking and Availability Trial 454 Repeat Purchase 457 413 Alternative Ways to Segment a Market Targeting May Also Use Diffusion of Innovation 418 Product Positioning 420 Creating Unique Value for the Chosen Target 422 Branding and Brand Management 414 424 435 The Role of Packaging 435 The Packaging Decision 435 Summary 436 Applications 437 Case: Wii 437 Case: Iridium 439 Case: Comparing Smartphones (C) 461 Setting 464 The Market Testing Decision Methods of Market Testing 471 473 Controlled Sale Methods Setting 441 The Launch Cycle 441 Prelaunch and Preannouncement 441 Announcement, Beachhead, and Early Growth 444 445 470 Pseudo Sale 470 Controlled Sale 470 Full Sale 471 Speculative Sale 472 Simulated Test Market 440 464 When Is the Decision Made? 465 Is This an Easy Decision to Make? 465 Market Tests Must Have Teeth 466 The Factors for Deciding Whether to Market Test 468 Pseudo Sale Methods Chapter 17 Implementation of the Strategic Plan 441 Lean Launch and Launch Timing Summary 458 Applications 458 CASE: Hulu 459 Case: Dodge Nitro 452 Chapter Eighteen Market Testing 464 Trademarks and Registration 424 What Is a Good Brand Name? 426 Managing Brand Equity 428 Brand Equity and Branding Strategies 432 Global Branding and Positioning: Standardize or Adapt? 433 Global Brand Leadership 434 Packaging 447 The Communications Plan 447 The Copy Strategy Statement 449 Personal Selling 449 Type of Demand Sought 409 Permanence 410 Aggressiveness 411 Competitive Advantage 411 Product Line Replacement 411 Competitive Relationship 413 Scope of Market Entry 413 Image 413 The Target Market Decision xix Informal Selling 476 Direct Marketing 477 Minimarkets 477 Scanner Market Testing Full Sale Methods 476 479 480 Test Marketing 480 The Rollout 484 Wrap-Up on Market Testing Methodologies 488 Summary 488 Applications 489 Case: PepsiCo—Pepsi-Kona and Pepsi One 489 xx Contents CHAPTER 19 Launch Management 492 Setting 492 What We Mean by Launch Management The Launch Management System 493 492 Step One: Spot Potential Problems 494 Step Two: Select the Control Events 498 Step Three: Develop Contingency Plans 499 Step Four: Design the Tracking System 500 Effective Innovation Metrics 504 A Sample Launch Management Plan 506 Launch Management and Knowledge Creation 506 Product Failure 509 Summary 511 Applications 512 Case: Levitra 512 CHAPTER 20 Public Policy Issues Strategy and Policy 534 Control Systems 534 Product Testing 534 Marketing and Market Testing 534 Customer Education and External Affairs 515 Phase I: Stirring 516 Phase II: Trial Support 517 Phase III: The Political Arena 517 Phase IV: Regulatory Adjustment 518 Appendix A Sources of Ideas Already Generated Business Attitudes toward Product Issues Current Problem Areas 518 Product Liability 519 Typology of Injury Sources 519 The Four Legal Bases for Product Liability Other Legislation 523 Planning for the Product Recall Prior to the Recall 524 During the Recall 524 After the Recall 524 535 Summary 535 Applications 536 Case: Clorox Green Works 536 Case: Hybrid or Hydrogen Vehicles at General Motors? 539 Case: Product (RED) 541 515 Setting 515 Bigger Picture: A Cycle of Concerns Attempts at Standardization and Clarification 525 Environmental Needs 525 Product Piracy 528 Worthy Products 529 Morality 530 Designing Products for Emerging Markets 530 Personal Ethics 532 The Underlying Residual Issues 533 What Are New Products Managers Doing about All This? 534 518 543 Appendix B Other Techniques of Concept Generation 549 Appendix C The Marketing Plan 559 521 Appendix D Guidelines for Evaluating a New Products Program 524 Index 569 565 New Products Management 2 Part One Overview and Opportunity Identification/Selection FIGURE I.1 Opportunity Identification and Selection Inputs from ongoing marketing planning Special opportunity audits Inputs from ongoing corporate planning Audit of relevant markets Audit of resources Sort Suggestions for Product Innovation Activity Exploit underutilized resources Exploit new resources Respond to external mandate Respond to internal mandate Technical Financial Product Market Discoveries Acquisitions Diversified markets Quality studies Customer needs Competitive threat Regulation Owners Top management plans Unit management plans Gaps Evaluate the resource and define it carefully Study the mandate, validate the threat Study the mandate, validate role for product innovation Cull to a Pool of Validated New Product Opportunities Reject those that conflict with ongoing product innovation strategy Reject those clearly not economically or technically viable Give Each Opportunity a Product Innovation Charter (PIC) Fits a PIC already in place Requires a new PIC To Figure II.1 (Concept Generation) P A R T O N E Overview and Opportunity Identification/Selection This book is divided into parts. They are (1) Overview and Opportunity Identification and Selection, (2) Concept Generation, (3) Concept/Project Evaluation, (4) Development, and (5) Launch. They follow the general flow of the new products process, which we will present in Chapter 1, Figure 1.5. We will see later, however, that the phases are not sequential, compartmentalized steps. They are quite fluid and overlap each other. At the beginning of each part is a short Part Introduction (noted with a Roman numeral) and a figure (see Figure I.1). The introduction describes briefly what aspects of the new products process will be covered in the upcoming chapters. The figure provides detailed information about what goes on at that phase in the new products process, and shows what phases come immediately before and after. Figure I.1, for example, details the opportunity identification and selection process, ending with the product innovation charter, a key topic of Chapter 3. Hence, the five part figures (Figures I.1, II.1, III.1, IV.1, and V.1) actually make up one long, detailed new products process, the essence of which is presented briefly in Figure 1.5. Before getting to opportunity identification and selection, we begin Part I with two introductory chapters. The first introduces the three strategic elements of product development: the new products process, the product innovation charter, and the product portfolio. It presents the first of these, the new products process, in relatively simplified form, as a kind of introduction to the rest of the book. Chapter 1 also attempts to answer the questions most often asked about such a course and helps to define some of the concepts we will be returning to throughout the text (such as, what exactly is a new product, how many new products really do succeed, and how do firms achieve globalization in product development). Chapter 2 goes much deeper into the new products process. Chapter 2 also introduces the key concepts of radical innovation, new service development, and speed to market, and how each of these may have an impact on the new products process as presented in the chapter. 4 Part One Overview and Opportunity Identification/Selection Chapter 3 completes the introductory part of the book, as it presents the second and third strategic elements. First, opportunity identification and selection are presented, which deal with the strategic planning lying at the very base of new products work that guides a new products team, just as corporate or SBU strategy guides the unit as a whole. Figure I.1 provides a flow model that describes the process of opportunity identification. Chapter 3 then discusses the product innovation charter (PIC). This can be thought of as a statement of strategy that will guide the new product development team: the arena in which they will operate, their goals and objectives, and other considerations. The last part of Chapter 3 discusses the product portfolio. Innovative ideas that can be converted into high-potential new product opportunities can come from many sources; but however the new product idea is arrived at, its fit with the firm’s product innovation strategies needs to be assessed. This is a portfolio issue: When assessing any potential new product, the firm needs to consider its technical viability (can we make it?) and its market viability (will customers buy it?). Most firms will have many other criteria, both financial and strategic, that they consider at this important step. As seen in Figure I.1, once the PIC has been determined, the next step is to generate product concepts. This will be taken up in Part II of this book. C H A P T E R O N E The Strategic Elements of Product Development Setting Mention new products and people think about technology—iPods, iPhones, YouTube, virtual realities, fiber optics, and the like. But most new products are far simpler—low-carb colas, new movies, new singing stars, fast foods, and new flavors of frozen yogurt. New products run the gamut from the cutting edge of technology to the latest version of the ballpoint pen. New products can be tangible goods or services. New products can be destined for the consumer market, the business-to-business market, or both. You have chosen to study how new products are developed and managed, so it would be nice to say they come from an orderly process, managed by experienced persons well versed in product innovation. Some do, but some don’t. Years ago, Art Fry became famous for an idea that became Post-it notes, when his hymnal page-marking slips kept falling out. He had a rough time persuading others at 3M that the idea was worth marketing, even though it soon became the second largest volume supply item in the office supply industry! Or consider James Dyson, an industrial designer by training who was dissatisfied with the performance of commercially available vacuum cleaners and set out to create a better one. After five years and about 5,000 prototypes, he created the Dual Cyclone bagless vacuum cleaner. Over the next eight years, he was unable to interest vacuum cleaner manufacturers or venture capitalists in the new product, frequently hearing that since he was a designer, he couldn’t possibly know anything about manufacturing or marketing! In 1985 and on the verge of bankruptcy, Dyson found an interested Japanese investor, and by 1993 he had set up Dyson Appliances in the United Kingdom (his home country). Since that time, Dyson Appliances has sold over $2 billion worth of vacuums worldwide.1 So you may be confused by the uncertainty you meet in this book. If so, welcome to the land of creative exploration. The activity we study in this book is sometimes called product innovation management; some call it product planning, and some (from a very biased perspective) call it research & development (R&D) or 1 Anonymous, “Dyson Fills a Vacuum,” @ Issue, 8(1), 2003. 6 Part One Overview and Opportunity Identification/Selection marketing. In this book, we use the most descriptive term we have—new products management—and we adopt the viewpoint of the marketing manager; that is, we are primarily concerned about the specific role for marketing in the overall task. The Importance of New Products New products are big business. Over a hundred billion dollars are spent yearly on the technical development phase alone. Untold thousands of new products are marketed every year, perhaps millions if we call each new Web site a new product. Hundreds of thousands of people make their living producing and marketing new products. Many managers realize that radical innovation is critical to future growth and even the survival of the firm. Here, we are defining radical innovation as innovation that displaces or makes obsolete current products and/or creates totally new product categories.2 The Industrial Research Institute identified “accelerating innovation” and “business growth through innovation” as the top challenges faced by technology leaders, and well-known business writer Gary Hamel has described the creation of radical innovation as “the most important business issue of our time.”3 The reason firms invest this much in new products is that they hold the answer to most firms’ biggest problems. Competitors do the most damage when (1) there is so little product differentiation that price-cutting takes everyone’s margins away or (2) when they have a desirable new item that we don’t. The fact is: A successful new product does more good for a firm than anything else. The very reason for a firm’s existence is the value its operations provide to others, and for which they pay. And in a competitive world this means that what we offer—be it a physical good or a service—must be better than what someone else offers, at least part of the time. This is true in all organizations, including hospitals, churches, colleges, and even political parties. Look at the winners in those arenas and ask yourself which ones are popular and growing. Another reason for studying about new products is that the new products process is exceedingly difficult. Hundreds of individuals are involved in the creation of a single product, but all are from separate departments (sales, engineering, manufacturing, and so on) where they may have their own agendas. When a product flops miserably, it often generates huge publicity, much to the chagrin of the producers: think of New Coke, Premier smokeless cigarettes, the movies Gigli and Catwoman, or countless others. Perhaps, as a result, we think failure rates are higher than they really are. New products do fail, of course, but at around a 40 percent rate, not 2 M. Rice, R. Liefer, and G. O’Connor, “Assessing Transition Readiness for Radical Innovations,” Research-Technology Management, 45(6), 2002, pp. 50–56; and Gina O’Connor, Joanne Hyland, and Mark P. Rice, “Bringing Radical and Other Innovations Successfully to Market: Bridging the Transition from R&D to Operations,” in The PDMA Toolbook 2 for New Product Development, ed. P. Belliveau, A. Griffin, and S. M. Somermeyer (Hoboken, NJ: Wiley, 2004), pp. 33–70. 3 Industrial Research Institute 2001/2002 Annual Reports, Washington, DC, Industrial Research Institute; and Gary Hamel, “Innovation Now! (It’s the Only Way to Win Today),” Fast Company, December 2002, pp. 114–124. Chapter One The Strategic Elements of Product Development 7 the 90 percent rate you often hear, and this percentage holds for both goods and services. The best product-developing firms can improve their odds further: They require only about four ideas to generate one winning product, as compared to over nine ideas for other firms. This is probably because the best firms are better at screening out bad ideas earlier.4 And after many years of research, we know many of the most important reasons why products fail. The firm doesn’t understand the customer, or underfunds the required research and development, or doesn’t do the required homework before beginning development (sometimes called the ready—fire—aim approach), or doesn’t pay enough attention to quality, or lacks senior management support, or chases a moving target (we will see moving-target issues such as unstable specifications and scope creep in Chapter 3).5 The goal at most firms is not necessarily to reduce failure rates to zero. Having too low a failure rate might mean that the firm is playing it too safe with close-tohome innovations, while missing out on the (risky) breakthroughs. The definition of “too low” probably depends on the industry and on how inherently risky product development is. The goal here is to minimize the dollar losses on the failures (don’t bankrupt the company!) and to learn from them. Regardless of the actual failure rate you encounter, the amount at stake and the risk of failure are high in new product development. Success rates have remained remarkably consistent over the years. The Comparative Performance Assessment Study (CPAS) is periodically conducted by the Product Development & Management Association (PDMA), most recently in 2012.6 In these studies, for every 100 ideas, a little under 70 make it through the initial screen; fewer than 50 pass concept evaluation and testing and are moved to the development phase; a little more than 30 make it through development; about 30 make it through testing; about 25 of them are commercialized; and about 15 are considered to be successes (about 60 percent of those that were commercialized). Interestingly, the percent success rate does not vary too much from one category to the next. The percent success rate ranges from 51 percent (frequently purchased consumer goods) to 65 percent (health care). If one splits the CPAS sample into two groups, the “Best” (the top-performing 25 percent of firms) and the “Rest,” a slightly different pattern emerges: In 2012, the Best firms attained a success rate of 4 Marjorie Adams, Competitive Performance Assessment (CPAS) Study Results, PDMA Foundation, 2004; and Stephen K. Markham and Hyunjung Lee, “Product Development and Management Association’s 2012 Comparative Performance Assessment Study,” Journal of Product Innovation Management, 30(3), 2013, pp. 408–429. Success rate has held steady at around 60 percent of products marketed since the 1995 CPAS study; the 2012 study suggests the success rates are slightly lower in Europe and Asia. 5 Robert Cooper, Winning at New Products: Accelerating the Process from Idea to Launch, 3rd ed. (New York: Perseus Books, 2001). 6 2003 CPAS results are found in Doug Boike and Marjorie Adams, “PDMA Foundation CPAS Study Reveals New Trends—While the ‘Best-Rest’ Gap in NPD Widens,” Visions, 28(3), July 2004, pp. 26–29; and Gloria Barczak, Abbie Griffin, and Kenneth B. Kahn, “Perspective: Trends and Drivers of Success in NPD Practices: Results of the 2003 PDMA Best Practices Study,” Journal of Product Innovation Management, 26(1), January 2009, pp. 3–23. The 2012 results are summarized in Markham and Lee (2012), op. cit. 8 Part One Overview and Opportunity Identification/Selection FIGURE 1.1 The Best Firms Achieve Superior NPD Results Percent Successes Percent of Sales from New Products Percent of Profits from New Products Number of Ideas Per Successful New Product The Best (top 25% of firms) The Rest (bottom 75% of firms) 82.2 47.9 48.5 4.5 52.9 25.4 25.0 11.4 Source: Adapted from Stephen K. Markham and Hyunjung Lee, “Product Development and Management Association’s 2012 Comparative Performance Assessment Study,” Journal of Product Innovation Management, 30(3), 2013, pp. 408–429. over 80 percent, while the Rest’s success rate was much lower at about 50 percent. The Best, therefore, have greater success with new product development!7 Figure 1.1 shows that the Best firms not only have a higher percentage rate of successes, but also derive almost twice as many sales and profits from new products (defined as five years old or younger) than do the Rest. Best firms are also more efficient in developing successful products: they require about 4.5 ideas to generate one success, while the Rest require almost three times as many ideas per success. In addition, the development cost per successful project for Best firms is roughly half the cost per successful project for the Rest.8 The 2012 CPAS study also reveals that the Best companies at product development manage their new products process differently than do the Rest. In sum, the Best companies are better at implementing many of the new products process concepts and principles that we discuss in upcoming chapters of this book. Relative to the Rest, the Best: • Are more likely to use market research tools like creativity sessions (which we explore in Chapter 5), trade-off analyses (Chapter 7), concept tests (Chapter 9), voice of the customer (Chapter 12), alpha and beta testing (Chapter 15), and test markets (Chapter 18). • Are more likely to have global market and operations strategies (Chapters 3 and 14). • Rely more on portfolio analysis for product selection (Chapter 3). • Tend to use social media and online communities more for information gathering (Chapter 5). • Employ formal processes for selecting which concepts to develop (Chapter 10). • Are more effective in using team support tools and team incentives (Chapter 14).9 7 The “Best” are defined in the CPAS study as those firms that are in the top 25 percent in their industry and above the mean in both program success and sales and profit success from new product development. 8 Stephen K. Markham and Hyunjung Lee, op. cit. 9 Stephen K. Markham and Hyunjung Lee, op. cit. Chapter One The Strategic Elements of Product Development 9 In sum, the concepts of new products management as presented throughout this book are used extensively, and well, by the top innovating companies, who achieve superior results from their new products! Globalization and New Product Development Like all aspects of modern business, product development has become more challenging due to increased globalization. To a greater extent than ever before, firms are seeing new product development as a global process in order to take advantage of worldwide opportunities and increase their efficiency and effectiveness of innovation. According to a 2007 study by consultants Booz & Company, the top global firms in terms of R&D spending deployed about 55 percent of their R&D spending in foreign countries. Among the 80 top U.S. R&D firms, $80.1 billion out of $146 billion was spent overseas, and similar percentages were found for top European and Japanese R&D firms.10 The Booz & Company study also showed that the firms with higher percentages of R&D spending deployed elsewhere did better than average on many important performance measures, such as return on investment and total shareholder return. This study found that firms have multiple reasons for increasing their global R&D efforts. In many foreign countries, R&D engineers are lower paid than in the United States, Western Europe, or Japan—but the salary gap is narrowing, especially for the most skilled engineers and scientists. Now, many firms look overseas not just to access a cheaper labor force, but to access the talent residing in these markets and the ideas generated by these skilled personnel. Huge markets such as India and China are obvious sources of talented engineers, and there is some evidence of specialization: India boasts strengths in automotive engineering, China in electronics. Another reason for increased global R&D is the increasing globalization of the innovating firms themselves. For example, as automakers seek to penetrate new markets such as China or India, it makes sense to conduct more of their design work in or near these markets than back in the home office located in Michigan or Bavaria. In addition, firms are under increased pressure to reduce product development times, or may be competing in increasingly turbulent market environments. These factors lead firms to leverage all the global resources they have at their disposal for product development.11 Many multinational firms seek to leverage their product development skills across their subsidiaries and gain competitive advantage by setting up global new 10 For a summary of the Booz & Company findings, see Barry Jaruzelski and Kevin Dehoff, “‘Beyond Borders: The Global Innovation 1000’ Study Reveals a Global Shift in R&D Spending,” Visions, 33(3), October 2009, pp. 27–30. 11 Elko J. Kleinschmidt, Ulrike de Brentani, and Sören Salomo, “Performance of Global New Product Development Programs: A Resource-Based View,” Journal of Product Innovation Management, 24(5), September 2007, pp. 419–441; see summary in K. Sivakumar, “Global Product Development,” in Jagdish N. Sheth and Naresh K. Malhotra, Wiley International Encyclopedia of Marketing, Volume 5, Product Innovation and Management (West Sussex, UK: John Wiley, 2011), pp. 68–74. 10 Part One Overview and Opportunity Identification/Selection product teams.12 A large firm may have R&D skills in its German subsidiary, its manufacturing in Asia, and its suppliers somewhere else again. A firm’s global presence, however, is no guarantee that it will automatically know how to efficiently manage its global operations. Effectively coordinating and marshaling the efforts across multiple countries to develop and to launch successful new products is a major challenge. There are many decisions to make that impact global product development effectiveness: how much autonomy should the subsidiaries have, how should they be rewarded, what work conditions should be imposed such that teamwork within and between subsidiaries is encouraged, and so forth. There is also the possibility of outsourcing some of the required new product capabilities, for example, through strategic alliances with global partners. Similarly, the global network of suppliers and distributors needs to be managed and coordinated so as to improve global product development as well as global launch. Selecting the best organizational structure for the global product team is more difficult than if only one culture is involved, as differences among team individuals as well as linguistic barriers and national culture differences must be taken into account. At the time of launch, even more decisions arise: Should a product be positioned the same way throughout the world, or should positioning, branding, or packaging decisions be localized? Many firms react to these challenges with well-defined, formal processes, while others leave the new products process relatively unstructured and adaptable to product or environmental considerations. The best research available on this topic finds that firms with a global innovation culture have the most effective global new product programs.13 Having a global innovation culture means that a firm is open to global markets, mindful of differences in customer needs and preferences, and respectful of different national cultural and business environments. Firms with such a corporate culture are able to recognize the specialized skills, resources, and ideas they possess in different subsidiaries around the world. In fact, at these firms, all operations and strategies (not just new product development) are defined in terms of the realities of the international market. A firm with a global innovation culture is better at integrating its global knowledge, can better manage the R&D tasks associated with the new products process, and has an advantage in implementing global launches.14 All of these factors contribute to improved global new product performance. Throughout this book, you will see examples of firms that practice innovation on a global basis, which includes managing virtual and highly diverse global product 12 Good references are: Roger J. Calantone and David A. Griffith, “From the Special Issue Editors: Challenges and Opportunities in the Field of Global Product Launch,” Journal of Product Innovation Management, 24(5), September 2007, pp. 414–418; and Ram Mudambi, Susan Mudambi, and Pietro Navarra, “Global Innovation in MNCs: The Effects of Subsidiary SelfDetermination and Teamwork,” Journal of Product Innovation Management, 24(5), September 2007, pp. 442–455. 13 Elko J. Kleinschmidt, Ulrike de Brentani, and Sören Salomo, op. cit. 14 Roger J. Calantone, S. T. Cavusgil, J. B. Schmidt, and G.-C. Shin, “Internationalization and the Dynamics of Product Adaptation: An Empirical Investigation,” Journal of Product Innovation Management, 22(2), March 2004, pp. 185–198. Chapter One The Strategic Elements of Product Development 11 FIGURE 1.2 Product Development as a Global Process Procter & Gamble: According to the P&G Web site, P&G products are developed as global R&D projects. P&G has 22 research centers in 13 countries from which they can draw expertise. As a good example of a global product, consider the Swiffer mop. P&G made use of its research centers in the United States and France to conduct market research and testing in support of this new product. Apple: In the development of the iPod, Apple worked with about ten different firms and independent contractors throughout the world, and did product design and customer requirement definition in both the United States and Japan. Ikea: The Swedish furniture retailer knows that its target market (middle-class strivers) crosses international and intercontinental lines, so it operates globally in a streamlined fashion. It identifies an unmet customer need (say a certain style of table at a given price point), commissions in-house and outsourced designers to compete for the best design, then its manufacturing partners worldwide compete for the rights to manufacture it. Excellent global logistics complete the value delivery to customers. Bungie Studios: This boutique software company, now owned by Microsoft, developed the MS Halo gaming software series in the United States, but product-tested it in Europe and Asia. Like Ikea customers in the prior example, gamers are much alike the world over. Source: Loida Rosario, “Borderless Innovation™: The Impact of Globalization on NPD Planning in Three Industries,” Visions, June 2006. development teams—no easy task! Figure 1.2 provides several samples of firms that take the global aspect of product development very seriously. Global new product teams are a way of life now for many firms, and we will see more about the challenges facing such teams in Chapter 14. There, we will focus on the issues facing the global new product development team, and how firms overcome these hurdles to take advantage of product knowledge residing in many corners of the world. We touch on some of the issues regarding global positioning and branding decisions in Chapter 16. How Product Development Is Different It is likely that this course is located in your university’s business school, within the marketing department. Or it might be part of your engineering training, or part of a specialized program in technology innovation management. In any case, this is a good time to note an underlying principle of product development: It’s all about teamwork. The new products team ideally is cross-functional, comprising personnel from marketing, R&D, engineering, manufacturing, production, design, and other functional areas as well. Unlike other courses you may be taking, we spend much time in this text on how you interact with people from other fields of study: discussing how team members work together, how they can improve communication, what they need to achieve when working together, and so on. So, whatever your background, and whatever course of study you are pursuing, remember that in product development you will spend a lot of your time coordinating and working closely with people from other functional areas. Above all else, product development is a joint effort. 12 Part One Overview and Opportunity Identification/Selection All members of a new products team make an important contribution to product development, so we must be aware of, and try to avoid, narrow functional viewpoints. Marketers have to learn to work with scientists, engineers, lawyers, production managers, and so on. We may come from marketing, and we will often return there when the project is finished, but, for now, we are all new products people, working with all functions, being biased to no one. A marketing type may not appreciate the thoroughness of a research scientist. And that scientist may not appreciate the marketer’s enthusiasm, which sometimes leads to what the scientist thinks are rash and unwarranted conclusions. Now is a good time to begin thinking like a general manager. This course of study calls for a strong creative contribution. Not only do we create new product concepts; in many firms, that’s easy. The tough part is how best to develop and market them—devising a concept-testing method that works, screening a totally new idea the firm has never faced, figuring out how to integrate engineers into a trade show booth effectively, how to position a product that creates its own new category, how to produce it on present equipment, how to name it in a way that communicates and is not confusing, and so on. No answers are found in the back of this book. We never will know whether any one decision was right, just whether the total package of decisions worked out. Being creative means we travel on unmarked roads. Most of our decisions are made on grossly inadequate facts. Not that we don’t know what facts we need or how to get good estimates of them—we usually do. But there’s never enough time or money. Worst of all, what seems to be a fact in January may not be a fact come June, when we actually introduce the new item. As a result, we often do things that make others nervous. For example, we use heuristics—rules of thumb that firms have found work for them: “On items such as this, about 30 percent of the people who hear of a new brand, try it,” or “When the product engineer from R&D disagrees with the process engineer from manufacturing, it’s better to go with manufacturing.” Heuristics sometimes leave us holding an empty bag; but without them, projects just won’t move forward fast enough. Another technique is to use simple intuition: hunch, or gut feel. This explains why most managers want new products people to have spent time in ongoing operations before moving on to new products work. This suggests another key difference between this course and many of your others. This course is about the activities of people working under intense pressure, making tough decisions under impossible conditions. Consider a now-classic example: a group of about 15 people sent by IBM from Armonk to Boca Raton during the dawn of the personal computer era, 1980. They were given one year to create and market a new product, which eventually became known as the IBM PC. Literally billions of dollars were at stake—the difference between becoming a major player in a new market or missing the boat completely. Virtually every day, someone on that team had to make a decision that could close the show. When studying how strategy guides teams throughout a project, or how firms telescope their market testing into simultaneous regional rollouts, remember that pressure. You may also be taking a course that deals with innovation in manufacturing or operations, and you may wonder how process innovation differs from product Chapter One The Strategic Elements of Product Development 13 FIGURE 1.3 Not All New Products Are Planned A Raytheon engineer working on experimental radar noticed that a chocolate bar in his shirt pocket melted. He then “cooked” some popcorn. The firm developed the first commercial microwave oven. A chemist at G. D. Searle licked his finger to turn a page of a book and got a sweet taste. Remembering that he had spilled some experimental fluid, he checked it out and produced aspartame (NutraSweet). A 3M researcher dropped a beaker of industrial compound and later noticed that where her sneakers had been splashed, they stayed clean. ScotchGard fabric protector resulted. A DuPont chemist was bothered by an experimental refrigerant that didn’t dissolve in conventional solvents or react to extreme temperatures. So the firm took the time to identify what later became Teflon. Another scientist couldn’t get plastic to mix evenly when cast into automobile parts. Disgusted, he threw a steel wool scouring pad into one batch as he quit for the night. Later, he noticed that the steel fibers conducted the heat out of the liquid quickly, letting it cool more evenly and stay mixed better. Bendix made many things from the new material, including brake linings. Others? Gore-Tex, dynamite, puffed wheat, Dextro-Maltose, LSD, penicillin, Dramamine, X rays, pulsars, and many more. In each case, a prepared mind. Sources: DuPont and Bendix cases, The Innovators (New York: Dow Jones, 1968); Raytheon, Searle, and 3M cases, Kenneth Labrich, “The Innovators,” Fortune, June 6, 1988, p. 56. innovation. The term process innovation usually applies to functions, especially the manufacturing or distribution process, and every new product benefits from this type of innovation. The term product innovation applies to the total operation by which a new product is created and marketed, and it includes innovation in all of the functional processes. The last difference worth noting here is in application. Sometimes the new product process is accidental, or serendipitous (see Figure 1.3). But remember the old adage that chance favors the prepared mind. At least two dozen scientists had observed mold killing their bacteria colonies before Alexander Fleming pursued the phenomenon into the discovery of penicillin. More recently, Pfizer researchers noticed that several of the men in a test study of a new angina medication reported that it was ineffective at treating their angina, but it did have an unexpected alternative effect on the body. Soon, Pfizer was marketing Viagra, one of their top products in recent years and showing enough market growth potential to have attracted several competitors.15 So, we must practice. You cannot learn how to develop a new product concept by reading about attribute analysis or gap analysis. You must do them. The same goes for product use testing, positioning, contingency planning, and many more. There are opportunities at the end of every chapter to think about the chapter’s material in a market setting. 15 Jenny Darroch and Morgan P. Miles, “Sources of Innovation,” in V. K. Narayanan and Gina C. O’Connor (eds.), Encyclopedia of Technology & Innovation Management (Chichester, UK: John Wiley, 2010), Chapter 14. 14 Part One Overview and Opportunity Identification/Selection What Is a New Product, and What Leads to Success? The term new product can mean different things to different people. Figure 1.4 shows that new products can include new-to-the-world (sometimes called really new) products, as well as minor repositionings and cost reductions. The list in Figure 1.4 may include things you would exclude. For example, can we have a new item just by repositioning an old one (telling customers it is something else)? Arm & Hammer did, several times, by coming up with a new refrigerator deodorant, a new carpet freshener, a new drain deodorant, and more, all in the same package of baking soda, even with the same brand name. These may be considered just new uses, but the firm still went through a process of discovery and development. And a new use (particularly in industrial firms) may occur in a completely separate division. DuPont, for example, uses basic fibers in many different ways, from technical to consumer. Financial firms use their common databases for different markets. Similarly, brand names have long been used as platforms for launching line extensions. The Dove soap name, for example, has been extended to almost two dozen box soaps and almost as many liquid body washes.16 FIGURE 1.4 What Is a New Product? New products can be categorized in terms of how new they really are to the world, or to the firm. One common set of categories is as follows: 1. New-to-the-world products, or really new products. These products are inventions that create a whole new market. Examples: Polaroid camera, the iPod and iPad, Hewlett-Packard’s laser printer, Rollerblade brand inline skates, P&G’s Febreze and Dryel. 2. New-to-the-firm products, or new product lines. Products that take a firm into a category new to it. The products are not new to the world, but are new to the firm. Examples: P&G’s first shampoo or coffee, Hallmark gift items, AT&T’s Universal credit card, Canon’s laser printer. 3. Additions to existing product lines. These are “flanker” brands, or line extensions, designed to flesh out the product line as offered to the firm’s current markets. Examples: P&G’s Tide Liquid detergent, Bud Light, Special K line extensions (drinks, snack bars, and crystals). 4. Improvements and revisions to existing products. Current products made better. Examples: P&G’s Ivory Soap and Tide powder laundry detergent have been revised numerous times throughout their history; countless other examples. 5. Repositionings. Products that are retargeted for a new use or application. Example: Arm & Hammer baking soda repositioned as a drain or refrigerator deodorant; aspirin repositioned as a safeguard against heart attacks. Also includes products retargeted to new users or new target markets; Marlboro cigarettes were repositioned from a woman’s cigarette to a man’s cigarette years ago. 6. Cost reductions. New products that simply replace existing products in the line, providing the customer similar performance but at a lower cost. May be more of a “new product” in terms of design or production than marketing. Sources: The categorization scheme was originally presented in Booz, Allen & Hamilton Inc., New Product Management for the 1980s (New York: Booz, Allen & Hamilton Inc., 1982) and is now standard in new product development. Some of the examples are from Robert G. Cooper, Winning at New Products: Accelerating the Process from Idea to Launch, 3rd ed. (Cambridge, MA: Perseus Publications, 2001). 16 Deborah L. Vence, “Just a Variation on a Theme,” Marketing News, February 2007, pp.18–20. Chapter One The Strategic Elements of Product Development 15 All the categories in Figure 1.4 are considered new products, but it is plain to see that the risks and uncertainties differ, and the categories need to be managed differently. Generally, if a product is new to the world or new to the firm (the first two categories), the risks and uncertainties faced by the firm are higher, as are the associated costs of development and launch. It cost Gillette far more, for example, to launch its newest shaving system (the Fusion) than to do upgrades to the earlier Mach 3 system (such as developing the women’s version, named Venus, which used the same blade technology). A greater commitment of human and financial resources is often required to bring the most innovative new products to market successfully. Note also that not all the new product categories in Figure 1.4 are necessarily innovations. Line extensions, like the Dove soap bars mentioned above, or new flavors of Oreo cookies, may have resulted from the company’s desire to increase display space and shelf space. As Bob Golden of Technomic, a food industry consultancy, notes, “Many of these companies [that launch line extensions] are cannibalizing existing brands in order to stimulate the [product] category.” Line extension shouldn’t be confused with “true” innovation—and management must recognize that true innovation that provides enhanced value to customers is where their long-term competitive advantage may lie.17 New-to-the-world products revolutionize existing product categories or define wholly new ones. They are the most likely to require consumer learning and/ or incorporate a very new technology. Desktop computers with word processing software defined a new product category that made electric and manual typewriters virtually obsolete, and consumer learning was required by those who type for a living. Hewlett-Packard LaserJet printers did much the same thing in the printer category. The launch of CDs required major differences at the retail level in terms of store layout and distribution of related components (such as CD players). Other familiar examples, such as hybrid cars, the iPod, and even the Swatch watch, illustrate the use of new technologies in new-to-the-world products. Manufacturers had to overcome perceived risks, perceived incompatibility with prior experience, or other barriers to customer adoption (more on this subject in Chapter 16). Of course, launching new-to-the-world products means risk—and the encouragement to take on the risk must permeate the whole firm and must start at the highest levels of management. At highly innovative firms like Intel and Gillette (the latter now a division of Procter & Gamble), top management may even abandon the use of quarterly earnings estimates in order to keep the business units focused on innovation and other long-term strategic goals.18 The new product line category in Figure 1.4 raises the issue of the imitation product, a strictly “me-too.” If a firm introduces a brand of light beer that is new to them but is identical to those already on the market, is it a new product? Yes, it is new to the firm, and it requires the new products process. Canon was not the first laser printer manufacturer, Coca-Cola was not the first orange-juice bottler, and P&G was not 17 Deborah L. Vence, op. cit.; the Bob Golden quote is from Karen Heller, “It’s in the Snack Aisle, But Is It Food?” Philadelphia Inquirier, March 14, 2007, pp. E1, E4. 18 Thomas D. Kuczmarski, “What Is Innovation? And Why Aren’t Companies Doing More of It?” Journal of Consumer Marketing, 20(6), 2003, pp. 536–541. 16 Part One Overview and Opportunity Identification/Selection the first competitor in the coffee business. These were new products to these firms, however, managerially speaking, and they are managed as such by the companies. Figure 1.4 shows that many new products can be considered additions to existing product lines or improvements and revisions to existing products. Many of these line extensions round out or add to existing product lines extremely well: Tide Liquid detergent, Bud Light, Special K snack bars or shakes. Nevertheless, studies suggest that the most innovative new product categories account for many more product successes. In one study, the two most innovative categories accounted for about 30 percent of new product launches, but about 60 percent of the most successful products. (Percentages, of course, will vary by industry: High-tech industries will produce proportionately more highly innovative new products.) In fact, a U shape between innovativeness and success was found: The most innovative new product categories and the least innovative categories (the repositionings and cost reductions) outperformed the middle categories in terms of meeting financial criteria, returns on investment, and resulting market shares!19 This is because new products in the “middle ground” are not new enough to really excite new customers, yet different enough from existing products that there are fewer synergies. The results suggest that many firms need to reconsider the importance and potential contribution of innovative new products when making project selection decisions. In Chapter 3 we shall look at building a strategic portfolio of products that strives for balance among the innovation categories. We have already seen that, even among the best firms, there are some product failures, and this entire book is devoted to developing new successful products, so there can be no easy answer to the question “What leads to new product success?” Nevertheless, several studies over the years on this question have yielded a consistent answer: The number one reason for success is a unique superior product. Additionally, common causes of failure include “no need for the product” and “there was a need but the new product did not meet that need.” In other words, it was not unique and superior.20 It did not offer the user sufficient value added relative to the costs of purchasing and use. Value added is a key concept to keep in mind as you travel the new product highway. Does This Field of Activity Have a Unique Vocabulary? Yes, it does, for two reasons. One, it is an expanding field, taking on new tasks and performing them in new ways. Second, it is a melting pot field, bringing in the language of scientists, lawyers, marketing people, accountants, production 19 Elko J. Kleinschmidt and Robert G. Cooper, “The Impact of Product Innovativeness on Performance,” Journal of Product Innovation Management, 8(4), December 1991, pp. 240–251; see also Abbie Griffin, Drivers of NPD Success: The 1997 PDMA Report (Chicago: Product Development & Management Association, 1997). 20 Discussions of product success and failure can be found in R. G. Cooper, “New Products: What Separates the Winners from the Losers?” in M. D. Rosenau, A. Griffin, G. Castellion, and N. Anscheutz (eds.), The PDMA Handbook of New Product Development (New York: John Wiley, 1996), pp. 3–18; and R.G. Cooper, “The Impact of Product Innovativeness on Performance,” Journal of Product Innovation Management, 16(2), April 1999, pp. 115–133. Chapter One The Strategic Elements of Product Development 17 people, corporate strategists, and many more. Because many of these people talk about the same event but using different terms, communication problems abound. For example, there is sometimes confusion over the terms invention and innovation. To managers invention refers to the dimension of uniqueness—the form, formulation, function of something. It is usually patentable. Innovation refers to the overall process whereby an invention is transformed into a commercial product that can be sold profitably. The invention may take but a few moments. We have far more inventions than we do innovations. Similarly, the average person might think that a product idea, a product concept, a product prototype, and maybe even a product are all about the same thing. As you will see in the pages of this book, we have specific, distinct definitions for each of these terms, and they are not interchangeable. The problem becomes much worse from a global perspective. Take, for example, the term design. In North American new product work, design means essentially industrial design or engineering (premanufacturing) design; in Europe, however, design means the entire technical creation function from initial specs to the shipping dock. To some design people, the term means the entire product innovation function. When in doubt, a complete glossary of new product terms is published online by the Product Development & Management Association (www.pdma.org; follow the link to the glossary). Does the Field of New Products Offer Careers? It does, though not many are entry positions for people right out of college. Generally, top managers want new products people to know the industry involved (for the customer understanding mentioned earlier) and the firm’s various operations (that multidimensional, orchestration task also mentioned). So, most new products managers get assigned to new products work from a position in a functional department. For example, a scientist finds working with marketing and manufacturing people interesting, a market researcher specializes in benefit segmentation, or a salesperson earns a reputation for good new product concepts. Each of these people is a candidate for full-time work on new products. The specific jobs in this field are three. First is functional representative on a team, sometimes full time, more often part time. An example is a marketing researcher or a production planner. These people may be representatives on several teams or just one. The second job is project manager or team leader. This role is leader of a team of people representing the functions that will be required. The third position is new products process manager, responsible for helping project managers develop and use good new product processes. Some of the career tips we hear are: 1. Be multifunctional, not functionally parochial. Have experience in more than one function (marketing, manufacturing, and so on). 18 Part One Overview and Opportunity Identification/Selection 2. Be a risk taker, willing to do whatever is necessary to bring a product to market, including facing the wrath of coworkers. 3. Think like a general manager. Scientists and sales managers can lead new products teams, but they must cease being scientists and sales managers. 4. Be a combination of optimist and realist, aggressor and team player, leader and follower. 5. Develop your creative skills, both for new product concepts and for new ways of doing things. 6. Be comfortable in chaos and confusion. Learn to work with depressives, euphorics, and those with no emotion at all. Fortunately, such managers do exist—and in increasing numbers. We hope you become one of them. The Strategic Elements of Product Development We cover a lot of product development material in this book, from opportunity identification right through to launch and postlaunch. Underlying all of this are three strategic elements, which will be a major focus in this book. These strategic elements provide a framework to guide management through product development and help them focus on what is most important. Top product development consultants, like Robert Cooper of the Product Development Institute, recommend a framework of this type to firms of all sizes to help guide product development.21 A key point here is that all three of the strategic elements must be in place, and each is coordinated with, and supports, all the others. The three elements are a new products process, a product innovation charter, and a well-managed product portfolio. The new products process is the procedure that takes the new product idea through concept evaluation, product development, launch, and postlaunch. This procedure is usually depicted as a phased process with evaluative steps between the phases, but as you will see in upcoming chapters, it is rarely so straightforward. The product innovation charter is essentially a strategy for new products. It ensures that the new product team develops products that are in line with firm objectives and strategies and that address marketplace opportunities. Product portfolio management helps the firm assess which new products would be the best additions to the existing product line, given both financial and strategic objectives. In this chapter, we introduce the first strategic element, the new products process, as it serves as a framework for everything that follows in this book, and explore it more deeply in Chapter 2. In Chapter 3, we discuss the last two strategic elements, the product innovation charter and product portfolio management. 21 Roger J. Calantone, S. T. Cavusgil, J. B. Schmidt, and G.-C. Shin, “Internationalization and the Dynamics of Product Adaptation: An Empirical Investigation,” Journal of Product Innovation Management, 22(2), March 2004, pp. 185–198. Chapter One The Strategic Elements of Product Development 19 The Basic New Products Process Figure 1.5 shows a simple new products process described in terms of phases and tasks. Research has shown that about 70 percent of firms use some kind of formal, cross-functional, phased new products process, and about 47 percent use clearly defined evaluation criteria after each phase. At least 40 percent of firms assign a process manager whose job it is to manage the phased new products process.22 The phased new products process is certainly well established among firms involved in new product development. The idea behind the new products process is that the phases represent activities that are conducted by the new product team; between the phases are evaluation tasks, or decision points.23 It is at these points that the hard Go/No Go decisions need to be made (that is, whether the project looks promising enough to go on to the next phase). Throughout this book, we will be looking at the kinds of tests (from concept tests, to product use tests, to market tests) that are used to gather information for project evaluation. The goal of a new products process is to manage down the amount of risk and uncertainty as one passes from idea generation to launch. There are periodic FIGURE 1.5 The Basic New Products Process Phase 1: Opportunity Identification and Selection Phase 2: Concept Generation Phase 3: Concept/Project Evaluation Phase 4: Development (includes both technical and marketing tasks) Phase 5: Launch 22 Markham and Lee (2012), op. cit.; Robert G. Cooper, Scott G. Edgett, and Elko J. Kleinschmidt, Improving New Product Development Performance and Practices: Benchmarking Study (Houston, TX: American Productivity and Quality Center, 2002); Marjorie Adams (2004), op. cit., and Kenneth B. Kahn, Gloria Barczak, and Roberta Moss (2002), op. cit. 23 Robert G. Cooper, Winning at New Products: Accelerating the Process from Idea to Launch, 3rd ed. (Cambridge, MA: Perseus Publishing, 2001). 20 Part One Overview and Opportunity Identification/Selection evaluations all the way through the process. A firm may have access to hundreds of ideas; weaker ones are immediately eliminated, and the better ones are refined into concepts. Later in the process, only the best concepts are approved and moved forward to the development phase. The product is continuously refined during the development phase and could still be halted before the launch phase if preliminary product use test results are not positive. By the time the product is launched, it has a much higher likelihood of succeeding (recall the roughly 60 percent success rate across many product categories cited earlier). Managing down the amount of uncertainty is important, because each additional phase means greater financial investment (possibly much greater), not to mention greater commitment of human resources. Firms using a new products process have reported improvements in product teamwork, less rework, greater success rates with new products, earlier identification of failures, improved launch, and up to 30 percent shorter cycle times.24 This is not to say, however, that all firms implement the process well. Other studies show that many firms that claim to have a new products process either designed it or implemented it poorly; thus, there is much room for improvement.25 One should note that the neat, linear sequencing of phases shown in Figure 1.5 is just not typical. The reality is that the activities are not sequential, but overlapping. It is not implied that one phase must be completed before work can begin on the next one, like a pass-the-baton relay race. In fact, overlapping is encouraged. There is much pressure for firms to accelerate time to market for new products, and a certain amount of phase overlapping is an important tool in speeding new products to market. To do this right, of course, requires that the product team members from different functional areas (marketing, R&D, manufacturing, design, engineering) communicate very effectively.26 Product development is truly multifunctional, where all functions (and, increasingly, the customer as well) work together on a cross-functional team to accomplish the required tasks. The whole of Chapter 14 investigates the organization and management of these crossfunctional teams in depth. But even though we discuss teams later in the text, keep in mind that the team must become involved as early as possible in the new products process. It is the responsibility of the team leader to bring together the right individuals with the right skill sets, and to encourage communication within the team, between the team and top management, and between the team and communities of customers. The effective team leader knows how to deal with power conflicts as well as technical complexity.27 24 Robert G. Cooper, “New Products: What Separates the Winners From the Losers and What Drives Success,” in K. B. Kahn, S. E. Kay, R. J. Slotegraaf, and S. Uban (Eds.), The PDMA Handbook of New Product Development (Hoboken, NJ: Wiley, 2013), Ch. 1, pp. 3–34. 25 Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, Best Practices in Product Innovation: What Distinguishes the Top Performers, Product Development Institute, 2003; Robert G. Cooper, “Perspective: The Stage-Gate® Idea-to-Launch Process—Update, What’s New, and NexGen Systems,” Journal of Product Innovation Management, 25(3), May 2008, pp. 213–232. 26 Preston G. Smith and D. G. Reinertsen, Developing Products in Half the Time (New York: Van Nostrand Reinhold, 1991). 27 Hans J. Thamhain, “Managing Product Development Project Teams,” in Kenneth B. Kahn, George Castellion, and Abbie Griffin (eds.), The PDMA Handbook of New Product Development (New York: John Wiley & Sons, 2005), pp. 127–143. Chapter One The Strategic Elements of Product Development 21 Another way that firms have been able to avoid delays and speed up time to market is to streamline the evaluation tasks. At Johnson & Johnson, the preparation for an evaluation task might have included preparing a 30- to 90-page review document. This was cut to a standardized presentation, with a one-page summary and a handful of slides—enough to inform senior management about the risks and commitments being decided upon. It was reported that weeks of preparation time were saved with the new format.28 Furthermore, we should clear up something about the evalulation tasks that occur after every phase in the new products process. Figure 1.5 implies that each phase is always followed by a Go/No Go decision. While this is often the case, it might be an oversimplification. If some key information is still missing or unavailable, a third option is possible, which we can call an “On decision.” This means that the project will move forward (a conditional “Go,” if you will), but the missing information must be gathered and the project could still be halted at a later phase. An evaluation task that includes conditional Go decisions is sometimes called a fuzzy gate. For example, a new packaged food product might do reasonably well at a concept test, but management might feel they don’t really have a read on the market until some product use testing (letting the customer actually taste the product) is conducted. An On decision would mean that the product is approved to move to development, but the product use test must yield positive results, otherwise the project would be halted at that point. Fuzzy gates, therefore, speed up the process because time is not wasted in obtaining complete information before the decision is made. They are relatively common; in the CPAS study, about 50 percent of projects move forward with some conditional decisions along the way. Nevertheless, the team must indeed make a firm decision once the necessary information is obtained; in other words, fuzzy gates still have teeth. A related problem occurs when teams actually make a full “Go” decision, but fail to commit any resources to the project. This is known as a hollow-gate problem and results in too many projects underway and, inevitably, cost overruns and launch delays. Similarly, a poor project may never be critically evaluated because it is the CEO’s pet project, or because a hidden personal or political agenda is influencing decision making. Gates without teeth, hollow gates, special treatment for executives, or hidden agendas can all hinder effectiveness of the new products process, but all are identifiable and avoidable.29 Another consideration is that the new products process might look very different for new-to-the-world, breakthrough products (more on these in Chapter 2) as compared to more incremental new products. A firm like P&G might use a simplified process for a low-risk project (such as a new detergent) in which some phases and evaluation tasks are combined or may even be omitted. The CPAS study showed that only about 40 percent of radical projects have phases that overlap or are skipped, while for incremental new products, about 59 percent have overlapping phases or skip some phases entirely. For a new-to-the-world product, such as 28 Robert G. Cooper, “What Leading Companies Are Doing to Reinvent Their NPD Processes,” Visions, 32(3), September 2008, pp. 6–10. 29 For more on all of these problem areas, see Cooper (2008), op. cit. 22 Part One Overview and Opportunity Identification/Selection Febreze or Dryel, P&G faces greater risks and higher expenses, and the complete new products process in all its detail will probably be followed. Thus, it is helpful to think of the process in Figure 1.5 as a guideline or framework, but to recognize that the new products process is really quite flexible. In fact, these characteristics (overlapping phases, fuzzy gates, and flexibility) are features of what is called the third-generation new products process, which is the way most firms interpret the process depicted in Figure 1.5.30 There is something else significant in Figure 1.5. The phases do not refer to functions or departments. Technical people may lead the technical portion of the development, but others participate, some very actively, including market research, sales, design, and others. Launch sounds like a marketing activity, but much of the marketing is done back during earlier phases. We discuss what we call the “marketing ramp-up” in detail in Chapter 15. Also, during launch, the manufacturing people are busy setting up production capability. Legal people are clearing brand names, and lab people are running tests on early product output. It is clear that the new products process is a job for a well-organized, efficient cross-functional team. Additionally, different firms group the new product activities differently. There is certainly no agreement on the exact number of steps. That is not a cause for concern. Rather than thinking of the process as some number of discrete phases, look for the bigger picture of a large, evolving, general-purpose process, which we break up into five phases partly for our benefit in presenting the story about new product activities. Different firms simply break up the same underlying process differently. We will go much deeper into the new products process in Chapter 2. The Other Strategic Elements The process depicted in Figure 1.5 is part of a firm’s new product strategy, but it leaves some questions unanswered. First, what is the firm’s underlying strategy for new products? What market and/or technology opportunities is it seeking to exploit? What is the strategic arena within which the firm will compete? How innovative does management want to be? Lacking a new product strategy, the firm will approach new product development in an unfocused manner. Without a clear boundary defining what new market or technology opportunities to pursue, any idea would seem to be all right, which leads to too many underfunded products. We call this new product strategy a product innovation charter, or PIC. The PIC is developed by senior management and provides guidance to all functional areas involved in innovation. It defines a scope of activity for new product development, helping the product team identify what opportunities lie within the boundaries and where they should focus their efforts. That way, perhaps fewer projects may be pursued, but they will generally be of higher value 30 See Robert G. Cooper, “Perspective: Third-Generation New Product Processes,” Journal of Product Innovation Management, 11(1), 1994, pp. 3–14; also Cooper (2008), op. cit.; also Robert G. Cooper, “Effective Gating,” Marketing Management, 18(2), 2009, pp. 12–17. Chapter One The Strategic Elements of Product Development 23 to the firm. And the advantages of establishing a PIC are clear-cut: In Robert Cooper’s research, firms with a strong product definition had about an 85 percent chance of success and averaged a 37 percent market share, while those with a weak product definition showed a 26 percent chance of success and a market share of about 23 percent.31 Additionally, many new product concepts may seem to be technically feasible and marketable. Before committing scarce financial and human resources, top management must also consider whether the new product, if developed, would fit the firm’s overall business strategy: whether it adds strategically to the products already being offered, or whether it throws the firm’s product line off balance. This is an issue of product portfolio management. While almost every firm will consider financial criteria such as expected sales revenues or profits when approving a new product development project, the best performing firms balance financial criteria with strategic considerations, such that the firm’s longterm objectives will be met and there will be a dependable flow of new products into the future.32 The product innovation charter, product portfolio management, and related issues are covered more deeply in Chapter 3. Product Development in Action To see the ongoing efforts of the best product developers in the business, check the Web site for the Product Development & Management Association (www.pdma .org). Among other things, the PDMA sponsors an Outstanding Corporate Innovator award. This award is not for a single great new product, but rather for a sustained program of new product success over at least five years. And award winners must tell attendees at the association’s annual conference how they did it. As we noted before, innovation can be taught—and managers from the best innovating firms serve as the teachers in these conference sessions. In most of these cases, one could take their systems right from this book. Winners have included Corning, Royal DSM, Merck, Hewlett-Packard, Dow Chemical, Maytag, Bausch & Lomb, Harley-Davidson, and many others (the full list is on the PDMA Web site). The PDMA Web site also provides links to their academic journal, the Journal of Product Innovation Management, and their practitioner-oriented newsletter, Visions, as well as to the glossary mentioned earlier. As you take this course, you may want to check these publications for the most recent and timely articles on many aspects of new product development and innovation, and for the current hot topics among new product development professionals. 31 Robert Cooper, Winning at New Products: Accelerating the Process from Idea to Launch, 2nd ed. (Reading, MA: Addison-Wesley, 1993). 32 Gary E. Blau, Joseph F. Pekny, Vishal A. Varma, and Paul R. Bunch, “Managing a Portfolio of Interdependent New Product Candidates in the Pharmaceutical Industry,” Journal of Product Innovation Management, 21(4), July 2004, pp. 227–245. 24 Part One Overview and Opportunity Identification/Selection Summary Applications This chapter has introduced you to the general field of new products management. You read how the activity is (or should be) found in all organizations, not just business. You read how this course of study relates to others, what a new product actually is, and that services and business products are covered, not just cake mixes, cell phones, and cars. You learned about where the field stands today, the hallmarks of our activity, our problems with vocabulary, and possible careers. Chapter 2 will take us directly into the new product process. At the end of each chapter are a few questions that arose (or could have) one time or another in a job interview. The candidate was a student who took a course in new products management, and the interviewer was a high-ranking person in the firm (here portrayed as the president). The questions came up naturally during discussion, and they are tough. Often, the executive didn’t intend them to be answered so much as talked about. Occasionally, the executive just made a comment and then paused for the applicant’s reaction. Each question or comment relates to something in the chapter. Imagine you are the person being interviewed. You do not have the option of ducking the question or saying “I really don’t know.” If, in fact, you really don’t know, then glance back over the reading to see what you missed. It’s also a good idea to exchange answers with another student taking the course, given that most of the applications involve opinions or interpretations, not recitation of facts. 1. “When you were talking a while ago about taking risks, I wondered just whose money you were talking about. A fellow I know out in California insists that all new product team members invest their own money (with his) in their projects. Fifty thousand dollars is not unusual. In that system I’ll bet you would be seeking to avoid risks, not trying to find them.” 2. “Funny thing, though, it sure does frustrate me when I hear a division general manager’s strategy is to imitate other firms. Now, I know some firms might reasonably use imitation, but none of my divisions should. Should they?” 3. “I would like to be sure as many of our people as possible support innovation, but I know some people in the firm just can’t react positively to proposed innovation, no matter how much we need it. Tell me, how do you think I should go about spotting the worst offenders, and what should I do with them when I find out who they are?” C H A P T E R T W O The New Products Process Setting Chapter 1 provided a view of the overall new products process—the phases and evaluative tasks that, if performed well, will churn out the new products the organization needs. This process appeared in Figure 1.5, which serves as a framework for the rest of this book. As noted in the introduction to Part I, the five figures that introduce each part of this book (Figures I.1, II.1, and so on) are indeed the five boxes of Figure 1.5, but expanded to show more detail on what happens at each phase in the process. In this chapter, we go more deeply into the phases of the new products process model of Figure 1.5, illustrating what tasks are required at each phase and who is responsible for what. We then explore several issues important to product managers: how the new products process can be sped up (without sacrificing product quality or running up the budget), how the process would have to be adapted for the development of new services, how to develop breakthrough innovations, and how the skills and resources of external partners can be leveraged to improve the process. We begin by relating a short new product story to illustrate some of the key activities in the new products process in action. This will lead into a deeper discussion of the new products process and its managerial aspects. In particular, the story clearly shows how the new products process is interwoven with the other strategic elements introduced in Chapter 1, that is, the product innovation charter and the new product portfolio. It also introduces the idea of the cross-functional team and the importance of effective team management in implementing the new products process. The Procter & Gamble Cosmetics Saga1 In 1989, leading U.S. consumer goods manufacturer Procter & Gamble established itself in the cosmetics business by acquiring two leading cosmetics brands, Cover Girl and Clarion. By 1991, they also owned Max Factor. P&G established a new Cosmetics strategic business unit (SBU) and tried developing and launching 1 The P&G saga is drawn from Robert G. Cooper and Michael S. Mills, “Succeeding at New Product Development the P&G Way: A Key Element Is Using the ‘Innovation Diamond,’” Visions, 29(4), October 2005, pp. 9–13. 26 Part One Overview and Opportunity Identification/Selection new products using the procedures it had mastered in the detergent, food, and other packaged-goods lines. By the mid-1990s, however, P&G senior management was considering its options for its Cosmetics SBU. Some new products had done poorly, Clarion was sold, and some managers were even wondering whether P&G should leave the cosmetics business for good. P&G, one of the leading product developers worldwide, certainly knew the long-term importance of new products to the firm’s bottom line. P&G’s CEO, A. G. Lafley, said that “innovation is a prerequisite for sustained growth. No other path to profitable growth can be sustained over time. Without continual innovation, markets stagnate, products become commodities, and margins shrink.” Senior management felt that this was not the time to get out of cosmetics; rather, the challenge was to turn new product development around within the ailing SBU. With this strong commitment to new products at the highest levels within the firm, the Cosmetics SBU’s challenge was clear: fix the new products process, so that new products can become a sustaining, vital part of their business into the long term. Cosmetics SBU management understood its new product weaknesses. At the time, there was little evidence of a clear product development strategy in the Cosmetics SBU. Product initiatives were all over the place, too many different product categories were being pursued, and too many customer segments were being targeted. In short, the SBU lacked a product development focus. Within a few years, P&G’s Cosmetics SBU had engineered a complete turnaround by following and implementing the three strategic elements we introduced in Chapter 1. It’s not so simple, of course; senior management had to recognize new products as the lifeblood of their SBU and a key component of their success. They had to adequately fund new products and assign the right people to the tasks. However, the cosmetics SBU turnaround is almost a textbook case of the value of the strategic elements. It also clearly shows that one or two of the strategic elements alone won’t be enough. All three were put into action, and each complemented and supported the others. Let’s examine each in turn. The Product Innovation Charter (PIC) The starting point for the turnaround was a clear product innovation charter (PIC), which starts with an honest situation assessment and opportunity identification. At the time, the Cosmetics SBU was trying to develop products for the entire body and having difficulty carving out a competitive position. The situation assessment showed that there was an underserved consumer market, who wanted quality products for facial use only (facial cleanser, eye or lip products, and so on). Also, a problem was identified in the supply chain, which had become uncoordinated. Production and shipments were not tied to fluctuations in market demand, with the result that products spent too long in the supply network. Some new products were almost obsolete (not really competitive any more) by the time they were launched! Management realized that the supply chain needed to be under better control, such that market demand forecasts drove production and shipping schedules. If they succeeded in improving the supply chain, fewer launches would be delayed, and new products would be more competitive at the time of launch. Chapter Two The New Products Process 27 While you will learn more about the PIC in Chapter 3, the important thing to know is that it is a systematic way for managers to develop a new product strategy that considers the goals for their product innovation efforts and how these efforts fit overall business strategy. It involves identifying a strategic focus (which markets and which technologies will be targeted). In this case the market was defined in terms of products for the face. Essentially, this became a statement of the strategic arena or battlefield the Cosmetics SBU was going to compete in. Any new product opportunities that did not clearly help Cosmetics achieve its objectives in this arena would no longer be pursued. The New Products Process A second strategic element is the new products process, which is the path the new product takes from idea to the time of launch and beyond. P&G’s Cosmetics SBU did not have an effective new products process in place, with the result that product development often proceeded without clear inputs from customers early in the process. Cosmetics implemented a new products process quite similar to the five-phase process introduced in Chapter 1, which ensured that project teams were established early in the process, that consumer research was done early, and that consumer insights were actually used in the development of new product concepts. (A term you will see in upcoming chapters is voice of the customer or VOC. Think back to this example and consider how the VOC was used to drive product development in the Cosmetics SBU.) Note here that having a new products process, and actually implementing it correctly, are two different things. Cosmetics made its new products process work by having tough evaluation steps between the phases. At each phase, the project team had a set of current best practices against which to evaluate the product, as well as clear end-points or expectations. In addition, the evaluations were two-step, consisting of both a team recommendation and a decision by top management. In the past, the evaluations were not so carefully done, with the result that too many poor ideas would be allowed to pass. The New Product Portfolio In addition to a well-functioning new products process, there needs also to be an assurance that the firm is developing the right products with respect to its product portfolio. P&G’s Cosmetics SBU established a plan for managing its product portfolio and systematically adding new products to the portfolio. Given the nature of the products they were making, it was important for new launches to create buzz and excitement in the marketplace—for that reason, excellent launch timing was important. Cosmetics added new products to the existing portfolio strategically, in terms of both product selection and launch timing. For example, a new eye makeup would not be launched onto the market if there were already too many similar products in the portfolio, or if a similar product had just been launched. SBU management spoke of establishing an “initiative rhythm” for its products: New products would come through the development pipeline such that they would be ready for market launch at the best moment. We will see much more about portfolio management in Chapter 3. 28 Part One Overview and Opportunity Identification/Selection Supporting the Strategic Elements: Effective Team Management Finally, new product team management within the Cosmetics SBU was excellent. First and foremost, senior SBU management was committed to turning new product development around and to properly implementing the new products process. Certainly, senior P&G management such as CEO A. G. Lafley was behind them as well, judging from his quote above. Cosmetics management ensured that a positive innovative culture was in place within their SBU, one in which employees worked effectively in empowered, cross-functional teams. One important step, established in Cosmetics as well as in other SBUs, was to create initiative success managers (ISMs), who reported to senior SBU management. These ISMs led strategy development sessions, managed new product evaluation meetings, took part in resource planning, trained employees, and (importantly) shared what they learned with ISMs working in other business units, so that their expertise could spread quickly throughout the firm. A solid set of metrics was also established, so that the performance of each development team could be honestly assessed every six or 12 months on key indicators established by SBU management. As with all other SBUs within P&G, great emphasis was placed on identifying the best team leaders, who could come from any of the functional areas (marketing, engineering, R&D, or elsewhere) and were rewarded based on how they performed relative to the established indicators. We will learn more about cross-functional teams and other organizational issues in Chapter 14. What Happened in That Saga? We just read several years’ worth of product development activity in a few minutes. The story began with an ongoing operation that was facing a difficult situation. The saga illustrates how the managers involved applied the strategic elements effectively. We also saw how important it was to get support for the process from top management—in this case, from the CEO himself. This situation is typical in that the new products process does not usually begin with a new product idea. It is folklore that someone, somewhere, wakes up in the middle of the night with a great insight. It can happen, but successful new product programs are not built on such slender hopes. As the saga shows, the process usually begins with what amounts to strategy. With top management’s support and good execution of all the strategic elements, P&G was able to turn the weak Cosmetics SBU around. Note too that development does not take place behind the closed doors of a research lab. The cross-functional team includes personnel from many departments, not just the product engineers or R&D people. Also, marketing doesn’t start when the product is finished. It becomes involved very early in the process—in this saga, marketing provided key information for the development of the PIC. Last, the process is not over when the new product is launched. It ends when the new product is successful, usually after some in-flight corrections. P&G monitors the sales, profits, and market shares of its new products and takes corrective actions if interim goals are not reached. Chapter Two The New Products Process 29 The next section looks more deeply at the phases of the new products process, first introduced in Chapter 1. The Phases in the New Products Process Figure 2.1 shows a more detailed version of the new products process. Let’s examine each of the phases individually to understand the basics. FIGURE 2.1 The Phases of the New Products Process Phase 1: Opportunity Identification and Selection (Figure I.1) Generate new products opportunities as spinouts of the ongoing business operation, new products suggestions, changes in marketing plan, resource changes, and new needs/wants in the marketplace. Research, evaluate, validate, and rank them (as opportunities, not specific product concepts). Give major ones a preliminary strategic statement to guide further work on them. Phase 2: Concept Generation (Figure II.1) Select a high potential/urgency opportunity, and begin customer involvement. Collect available new product concepts that fit the opportunity and generate new ones as well. Phase 3: Concept/Project Evaluation (Figure III.1) Evaluate new products concepts (as they begin to come in) on technical, marketing, and financial criteria. Rank them and select the best two or three. Request project proposal authorization when in possession of product definition, team, budget, skeleton of development plan, and final PIC. Phase 4: Development (Figure IV.1) A. Technical tasks B. Marketing tasks Specify the full development process Prepare strategy, tactics, and launch and its deliverables. Undertake to details for marketing plan, prepare design prototypes; test and validate proposed business plan and get prototypes against protocol; design approval for it, stipulate product and validate production process for augmentation (service, packaging, the best prototype; slowly scale up branding, etc.) and prepare for it. production as necessary for product and market testing. Phase 5: Launch (Figure V.1) Commercialize the plans and prototypes from development phase; begin distribution and sale of the new product (maybe on a limited basis); and manage the launch program to achieve the goals and objectives set in the PIC (as modified in the final business plan). 30 Part One Overview and Opportunity Identification/Selection Phase 1: Opportunity Identification and Selection The first phase is strategic in nature; successful completion of this phase yields strategic guidance to the new products team, which guides idea generation and all remaining phases in the new products process. At least three main streams of activity feed strategic planning for new products. They are as follows. • Ongoing marketing planning. Example: The annual marketing plan for a CD-ROM line calls for a line extension to meet encroachment of a new competitor selling primarily on price. • Ongoing corporate planning. Example: Top management adopts a strategy that says either own a market (meaning get either a first- or second-place share) or get out of it. This will require new product activity in all desirable markets where the firm holds a minor position. • Special opportunity analysis. One or more persons (in the firm or a consulting firm) are assigned to take an inventory of the firm’s resources (people, facilities, reputations, whatever). Example: A firm in the auto parts business called for an audit of its manufacturing operation. It turned out that manufacturing process engineering had been overlooked or just not appreciated; that skill could serve as the base for a new products program. From these activities, the opportunities identified can be sorted into four categories. Here are illustrations: • An underutilized resource: A bottling operation, a strong franchise with dealers, or that manufacturing process engineering department. • A new resource: DuPont’s discovery of Surlyn, a material with hundreds of potential uses. • An external mandate: The market may be stagnant, the competition may be threatening, or customer needs may be evolving. Challenges like this will cause the firm to search for new opportunities, as did the Tasty Baking Company in the case at the end of this chapter. • An internal mandate: Long-range planning often establishes a five-year-out dollar sales target, and new products people often must fill part of the gap between current sales and that target. That assignment is called the product innovation (and/or acquisition) gap. Other common internal mandates are simply upper management desires, such as Steve Jobs’s stated goal to “reinvent the phone” with the iPhone project.2 The process of creatively recognizing such opportunities is called opportunity identification. The opportunities are carefully and thoroughly described, then analyzed to confirm that a sales potential does, indeed, exist. Recall that one of the first things P&G Cosmetics did was recognize that the new products process for 2 Henry Robben, “Opportunity Identification,” in Jagdish N. Sheth and Naresh K. Malhotra, Wiley International Encyclopedia of Marketing, Volume 5, Product Innovation and Management (West Sussex, UK: John Wiley, 2011), p. 153. Chapter Two The New Products Process 31 cosmetics could be fixed, and that new products in this category could be a viable business direction. Opportunities can be anywhere, and firms like Corning have “opportunity scouts” and “tech scouts” who work with networks of external technologists and business leaders to find promising opportunities.3 Of course, no firm wants to exploit all opportunities; some are better than others. Some may not fit with company skills, some are too risky, some require more money than the firm has. So, most firms have ongoing strategies covering product innovation. For example, Waterford had a strategy that no new product would jeopardize the firm’s great image. Gillette and Sony usually choose leadingedge innovation strategies. Once an opportunity is approved, managers turn to various techniques to guide new product people in exploiting it. This we will call the product innovation charter (PIC), and it will be explained in Chapter 3. Phase 2: Concept Generation In some cases, merely identifying an opportunity spells out what is wanted (for example, an opportunity to add a small size of deodorant for travelers). Most times, however, it is not so clear, so an immense set of ideation tools has evolved. Creating new product ideas, usually called product concepts by new products people, sounds like fun, but is hard and sometimes frustrating work. The most fruitful ideation involves identifying problems people or businesses have and suggesting solutions to them. For example, if the opportunity focused on “people moving their families over long distances,” the first ideation step is to study those people and find what problems they have. This is the problem find-solve approach. While this problem-based ideation is going on, unsolicited ideas are pouring in via phone, mail, and e-mail from customers, potential or former customers, employees (especially sales, technical, and operations), and every other source imaginable. These ideas are reviewed briefly by whomever receives them to see if they are even relevant to the firm and its strategies. They are then put into the pool with the ideas that came from problem-solving activities. Concept generation is covered in Part II, Chapters 4 through 7. Phase 3: Concept/Project Evaluation Before development work can begin on new ideas, they need to be evaluated, screened, sorted out. This activity, sometimes called screening or pretechnical evaluation, varies tremendously. But most firms generally follow a sequence from quick looks to complete discounted cash flows and a net present value. The quick look is necessary because the flow of new product concepts is huge and can easily reach the thousands in many firms. But what happens next is the first formal type of evaluation. Depending on the idea, this may be end-user screening or technical screening, or both. The work may be extensive and difficult, or it may take no more than a few phone calls or 3 Jacquelin Cooper, “How Industry Leaders Find, Evaluate and Choose the Most Promising Open Innovation Opportunities,” Visions, 36(1), 2012, pp. 20–23. 32 Part One Overview and Opportunity Identification/Selection e-mails. In the P&G Cosmetics example, some of the proposed new products may have originated among the technical people; this would have to be followed by a concept test to see what potential consumers thought about it. Ultimately, these views all come together in what is often called the full screen. It uses a scoring model of some type and results in a decision to either undertake development or quit. If the decision is to go ahead, the evaluation turns into project evaluation, where we no longer evaluate the idea but rather the plan we propose for capitalizing on that idea. This involves preparing a statement of what is wanted from the new product. Firms using Quality Function Deployment (a method of project management and control, which we explore in Chapter 12) see this as the first list of customer needs. A more common generic term is product description or product definition. In this book it will be called product protocol. Protocol here means a kind of agreement, and it is important that there be agreement between the various groups before extensive technical work gets under way. The protocol should, to the extent possible, be benefits the new item is to yield, not the features the new item is to have. The lack of good hard information complicates all pretechnical evaluation. In fact, the first three phases (strategic planning, concept generation, and, especially, concept/project evaluation) comprise what is popularly called the fuzzy front end (of the new product process). By the end of the project, most fuzz will have been removed, but for now, we move with more daring than the data allow.4 The various pretechnical evaluation actions are covered in Part III, Chapters 8 through 12. Phase 4: Development This is the phase during which the item acquires finite form—a tangible good or a specific sequence of resources and activities that will perform an intangible service. It is also the phase during which the marketing plan is sketched and gradually fleshed out. Business practice varies immensely, but we often find the following components. Resource Preparation Often overlooked by new products managers is a step called resource preparation. For product improvements and some line extensions, this is fine, because a firm is already up and running in a mode that fits products that are close to home. The culture is right, market data are more reliable, and ongoing managers are ready to do the work. But a particular innovation charter may leave familiar territory, forcing problems of fit. If a firm wants new-to-the-world products (more about them later in this chapter), then the team will need to be adequately prepared: it may need special training, new reward systems, revisions in the firm’s usual project review system, and special permissions. 4 The fuzzy front end has been the subject of much research the past few years. A good resource is Peter A. Koen, Greg A. Ajamian, Scott Boyce, Allen Clamen, Eden Fisher, Stavros Fountoulakis, Albert Johnson, Pushpinder Puri, and Rebecca Seibert, “Fuzzy Front End: Effective Methods, Tools, and Techniques,” in P. Belliveau, A. Griffin, and S. M. Somermeyer, The PDMA Toolbook for New Product Development (New York: John Wiley, 2002), Ch. 1. Chapter Two The New Products Process 33 The Major Body of Effort Next comes what all of the previous steps have been leading up to—the actual development of not one thing, but three—the item or service itself, the marketing plan for it, and a business (or financial) plan that final approval will require. The product (or concept) stream involves industrial design and bench work (goods) or systems design (services), prototypes, product specifications, and so on. It culminates in a product that the developers hope is finished: produced, tested, and costed out. While the technical developers are at work, marketing planners are busy making periodic market scans (to keep up with changes out there) and making marketing decisions as early as they can be made—first strategic and then tactical. Marketing decisions are completely interlaced with technical ones and involve package design, brand name selection, and tentative marketing budgets. A technical disappointment down the line may junk the early package design, name, or whatever. But we have to pay that price; we can’t wait for each step to be conclusive before going to the next one. Along the way, concept evaluation continues; we evaluated the concept well enough to permit development work (discussed earlier), but we have to keep evaluating technical and marketing planning results. We evaluate prototypes primarily, checking to be sure that the technology being developed meets the needs and desires of the customers in a way that creates value for them, while at the same time being profitable commercially.5 By the time this phase winds down, we want to be assured that the new product actually does solve those problems we began with. Comprehensive Business Analysis If the product is real and customers like it, some firms make a comprehensive business analysis before moving into launch. The financial analysis is still not firm, but it is good enough to assure management that this project will be worthwhile. The financials will gradually be tightened during the launch phase, and where the actual Go/No Go point is reached varies with the nature of the industry. Approval for a new food product can be held until just before signing TV advertising contracts, but a new chemical that requires a new manufacturing facility has to Go much earlier, and the pharmaceutical industry really makes the Go decision when it undertakes the 10-year, $50 million R&D research effort. The development phase is covered in Part IV, Chapters 13 through 15. Phase 5: Launch Traditionally, the term launch, or commercialization, has described that time or that decision when the firm decides to market a product (the Go in Go/No Go). We associate this decision with building factories or authorizing agencies to proceed with multimillion-dollar advertising campaigns. 5 Edward U. Bond, III and Mark B. Houston, “Barriers to Matching New Technologies and Market Opportunities in Established Firms,” Journal of Product Innovation Management, 20(2), March 2003, pp. 120–135. 34 Part One Overview and Opportunity Identification/Selection But launch is more complex than that. The launch is not a single point in time, the “opening night,” so to speak. Rather, product teams think of launch as a phase, including the last few weeks or months before and after the product is launched. During the launch phase, the product team is living life in the fast lane (or in the pressure cooker). Manufacturing is doing a gradual scale-up of output. The marketing planners, who got a good look at their ultimate target market as early as the opportunity, are now deep into the hundreds of tactical details required for launch. The critical step (if a company takes it) is the market test, a dress rehearsal for the launch, and managers hope any problems discovered are fixable between dress rehearsal and opening night. If not, the opening has to be delayed. Given the time pressures involved, managers have come up with many new ways to do reliable market tests quickly, to complement the familiar test market, which can be inordinately time-consuming and costly. We will review many market test techniques in Chapter 18. Sooner or later, the preparation activities lead to a public announcement of the new product through advertising, sales calls, and other promotional tactics. The announcement is often called launch. Most firms today execute the launch gradually, over a period of at least several weeks, since there are suppliers to bring on line, sales forces to be trained, distributors to be stocked and trained, and a large set of market support people to be educated (columnists, scientists, government people, and others). One thing that is often overlooked at this point is the activity of planning for launch management. When spacecraft are launched, a plan of tracking has been carefully prepared. The space control center implements the tracking plan, seeking to spot every glitch that comes up during launch and hoping it was anticipated so that a solution is on board, ready to use. New products managers often do the same thing, sometimes formally but often very informally. The launch phase is covered in Part V, Chapters 16 through 20. Evaluation Tasks Throughout the New Products Process Figure 2.2 illustrates the evaluation tasks encountered in the new products process. As shown, different kinds of questions need to be asked after different phases. For example, once concepts are generated, each is subject to an initial review: Is it any good, and is it worth refining? At the concept evaluation phase, careful screening is required, as concepts that pass this phase move on to development and begin incurring significant costs. In development, relevant questions are “Are we done yet?” and “If not, should we continue to try?” These questions are best answered through progress reports. Finally, at launch, the main questions concern whether the product should be launched, and later, how well it has done relative to expectation. We pick up discussion of Figure 2.2 later, in Chapter 8, when we go much more in depth into which evaluation techniques are the most useful at each point in the new products process. You may have noticed by now that the new products process essentially turns an opportunity (the real start) into a profit flow (the real finish). It begins with Chapter Two The New Products Process 35 FIGURE 2.2 The Evaluation Tasks in the New Products Process New Product Process Phase Evaluation Task Opportunity identification and selection Direction Where should we look? Concept generation Initial review Concept/project evaluation Is the idea worth screening? Full screen Should we try to develop it? Development Progress reports Have we developed it? If not, should we continue to try? Launch Market testing Should we market it? If so, how? In retrospect something that is not a product (the opportunity) and ends up with another thing that is not a product (the profit). The product comes from a situation and turns into an end. What we have, then, is an evolving product, or better, an evolving concept that, at the end, if it is successful, becomes a new product. Even a new product announcement just tells the world about a concept, hopefully a winner, but actually just in temporary form. Forces are standing by to see what revisions need to be made, even now, if it is off track. This evolution is linked to the phases of the new products process (see Figure 2.3). Here are the phases in that process, using a new skim milk product as an example: 36 Part One Overview and Opportunity Identification/Selection Phase 5: Launch Phase 4: Development Phase 2: Concept Generation Phase 1: Opportunity Identification The Evolution from Concept to New Product Phase 3: Concept/Project Evaluation FIGURE 2.3 100% Successful concept Marketed concept Clarity Pilot concept Protocol concept Prototype concept Full screened concept Tested concept Process concept Batch concept Stated concept Idea concept Opportunity concept 0 Low High Market value Phase 1: Opportunity Identification • Opportunity concept—a company skill or resource, or a customer problem. (Assume that skim milk drinkers tell us they don’t like the watered look of their favorite beverage.) Phase 2: Concept Generation • Idea concept—the first appearance of an idea. (“Maybe we could change the color . . .”) • Stated concept—a form or a technology, plus a clear statement of benefit. (See Chapter 4.) (Our firm’s patented method of breaking down protein globules might make the liquid more cloudy; emphasis on the might, at this time.) Phase 3: Concept/Project Evaluation • Tested concept—it has passed an end-user concept test; need is confirmed. (Consumers say they would very much like to have such a milk product, and the method of getting it sounds fine.) • Fully screened concept—it passes the test of fit with the company’s situation. • Protocol concept—a product definition that includes the intended market user, the problem perceived, the benefits that a less watery skim milk would have Chapter Two The New Products Process 37 to have, plus any mandatory features. (Our new product must taste as good or better than current skim milk, and it must yield exactly the same nutritional values.) Phase 4: Development • Prototype concept—a tentative physical product or system procedure, including features and benefits. (A small supply of a full-bodied skim milk, ready to consume, though not yet produced in quantity.) • Batch concept—first full test-of-fit with manufacturing; it can be made. Specifications are written stating exactly what the product is to be, including features, characteristics, and standards. (Skim milk ingredients: Vitamin A source, fat, fiber, and so on.) • Process concept—the full manufacturing process is complete. • Pilot concept—a supply of the new product, produced in quantity from a pilot production line, enough for field testing with end users. Phase 5: Launch • Marketed concept—output of the scale-up process from the pilot—a milk product that is actually marketed, either for a market test or for full-scale launch. • Successful concept (i.e., new product)—it meets the goals set for it at the start of the project. (New, Full Body Skim has achieved 24 percent of the market, is very profitable, and already competitors are negotiating licenses on our technology.) The idea that a new product suddenly emerges from R&D—like a chicken from an egg—is simply incorrect. In fact, throughout this book we will be examining how analytical techniques are applied throughout the new product process, from early idea generation and concept evaluation, through screening, and on to positioning, market testing, and launch management. Speeding the Product to Market One of today’s most discussed management goals in product development is accelerated product development (APD), or speeding the product to market. Accelerating time to market offers many benefits to the firm. The product will be on the market for a longer period of time before becoming obsolete, it can attract customers early and possibly block competitors with similar products that hit the market at a later time, or it can help to build or support a firm’s reputation. A firm that implements the strategic elements outlined in Chapter 1—the product innovation charter, the new products process, and portfolio management—has advantages in reducing cycle time. New product consultant Robert Cooper identifies five sure methods to accelerate time to market, some of which have been mentioned previously: • A clear product innovation charter—doing the opportunity identification homework and having a clean product definition—leads to better product design specifications and less time lost due to “recycling” (returning to earlier phases in the process to fix errors). 38 Part One Overview and Opportunity Identification/Selection • A third-generation new products process that permits overlapping phases or parallel processing results in more getting accomplished in a shorter span of time; streamlined evaluation tasks means that less time is wasted in evaluation. • A portfolio management approach minimizes the chance that the firm’s human and financial resources are spread too thinly over too many projects; better project selection focuses the firm’s scarce resources and uses them more efficiently. • A focus on quality at every phase complements the PIC; by following the adage “do it right the first time,” the firm will avoid unnecessary recycling. • An empowered cross-functional team, including individuals from marketing, R&D, manufacturing, and other functional areas, that works on the project from the earliest phases, supports parallel processing and eliminates “overthe-wall” product development (for example, marketing or production do not even begin their participation until the product is out of technical product development).6 Notice that the first three methods are the three strategic elements, while the last two (focus on quality and multifunctional product teams) are methods that help the firm to implement the strategic elements. There is plenty of evidence that these techniques contribute greatly to increasing speed to market. Software development is often marked by intensive “crunch time” periods due to approaching deadlines, and many firms in this industry rely on small, cohesive cross-functional teams to meet time goals while at the same time not sacrificing quality.7 Parallel processing is typical in the car industry: A car’s drive train may be 70 or 80 percent designed (but not 100 percent) before body design work is initiated. Then, an early prototype (but not the final car) may be built and ready for controlled test-driving. The use of parallel processing by Japanese automakers was a big factor in their emergence on the world market.8 Figure 2.4 shows many techniques that have been advocated for shortening cycle times. Note that the cycle time metric, that is, the way management measures speed to market (or, frequently, time to market), is often “getting the idea to the shipping dock faster.” This assumes that there has already been technical accomplishment—the 6 Robert G. Cooper, Winning at New Products: Accelerating the Process from Idea to Launch, 2nd ed. (Reading, MA: Addison-Wesley, 1993), p. 210. 7 B. J. Zirger and Janet L. Hartley, “The Effect of Acceleration Techniques on Product Development Time,” IEEE Transactions on Engineering Management, May 1996, pp. 143–152, look at cross-functional teams in electronics firms; and Alfredo M. Choperena, “Fast Cycle Time: Driver of Innovation and Quality,” Research-Technology Management, May–June 1996, pp. 36–40, examines the development of an immunoassay diagnostic system. Both found evidence that teams drive speed to market without sacrificing quality. 8 K. B. Clark and T. Fujimoto, Product Development Performance: Strategy, Organization, and Management in the World Auto Industry (Boston, MA: Harvard Business School Press, 1991). Chapter Two The New Products Process 39 FIGURE 2.4 Techniques for Attaining Speed in a New Product Project Organization Phase 1. 2. 3. 4. 5. 6. 7. Use dedicated cross-functional teams. Use small groups and other techniques to minimize bureaucracy. Empower a team, motivate it through incentives and rewards, and protect it. Destroy turf and territory. Make sure the supporting departments are ready when called on. Develop effective team leadership. Encourage organizational learning; transfer knowledge from one project to the next. Intensify Resource Commitments 1. 2. 3. 4. 5. 6. Integrate vendors; reduce numbers as necessary. Integrate other technology resources. Integrate resellers; reduce numbers as necessary. Get users involved early; capture the voice of the customer. Use simultaneous or concurrent engineering. Get suppliers involved through alliances, ventures, etc; develop long-term relations with them. Design for Speed 1. 2. 3. 4. 5. 6. Computer-aided design and other forms of rapid prototyping. Design-aided manufacturing, reduce number of parts, consider the manufacturing process. Use common components across families. Make the product easy to test. Design in the qualities that lead to fast trial, including relative advantage. Use effective design practices; minimize costly design changes late in the new products process. Prepare for Rapid Manufacturing 1. 2. 3. 4. 5. Simplify documentation. Use standardized process plans. Use computer-aided manufacturing. Go to just-in-time delivery of materials and components (flexible manufacturing). Integrate product use testing, and start it early. Prepare for Rapid Marketing 1. 2. 3. 4. 5. Use rollouts in place of test markets. Seed the firm’s reputation ahead of marketing. Spend what it takes to get immediate market awareness. Make trial purchasing as easy as possible. Get customer service capability in place ahead of need, and test it. Sources: Compiled from many sources, but a good overview of the topic is found in Pinar Cankurtaran, Fred Langerak, and Abbie Griffin, “Consequences of New Product Development Speed: A Meta-Analysis,” Journal of Product Innovation Management, 30(3), 2013, pp. 465–486. R of R&D has been concluded successfully. But from the point of view of technical development, speed to market success means not just time to the shipping dock, but also postshipping technical speed: For example, are corporate services (such as legal and environmental) in place? Also, if one uses the metric of “time 40 Part One Overview and Opportunity Identification/Selection to success” rather than “time to the shipping dock,” marketing has a much bigger role to play in accelerating cycle time. Marketing can strive to accelerate premarket speed (i.e., pretesting the marketing plan more quickly, or getting up to speed on field coverage through alliance formation), and also postannouncement speed (i.e., speeding up coupon redemption, or getting the sales reps into the field more quickly). We are now also hearing about the value of being first to mindshare rather than being first to market. The firm with mindshare in a given product category is the one that the target market associates with the product category and that is seen as the standard for competitors to match (such as Intel microprocessors or HewlettPackard laser printers). Firms that strive for mindshare think not about the speed of an individual product’s development and launch, but rather about creating a dominant position in the mind of the customer.9 Finally, the role of top management in speeding products to market cannot be ignored. It is not enough for top management simply to say “Cycle times are to be cut by 50 percent, effective now!” Employees will no doubt interpret such blanket statements as a thinly disguised command to work twice as hard. Real resources need to be committed to a cycle time reduction program. An expert in cycle time reduction, Preston Smith, reports that many firms expect the process to be quick and easy. Executives sometimes ask for a one- or two-day training program in cycle-time reduction, believing that to be adequate training. The idea is not to skip critical steps in the new products process, but to get through the process faster without sacrificing quality. Senior management will also know the value of strategic alliances to obtain technical and marketing resources and assistance. Alliances can be upstream to vendors, downstream to resellers and customers, and even sideways to competitors. Apple, for example, turned to Sony for assistance in speeding up the development of the PowerBook notebook.10 Risks and Guidelines in Speeding to Market There are plenty of advantages to speeding to market, not the least of which is that the product that is launched early is on the market for a longer period of time before becoming obsolete. A launch delay of, say, six months means six months less to earn profits and may give a competitor a chance to be first to market and establish a positive reputation. Nevertheless, there are lots of costs involved in speed, costs that are not evident and which can sometimes be disastrous. A firm facing increased competitive intensity, rapid technological change, and fast-changing market demographics 9 Denis Lambert and Stanley F. Slater, “First, Fast, and On-Time: The Path to Success. Or Is It?” Journal of Product Innovation Management, 16(5), September 1999, pp. 427–438. 10 For the Apple example, see Douglas W. LaBahn, Abdul Ali, and Robert Krapfel, “New Product Development Cycle Time: The Influence of Project and Process Functions in Small Manufacturing Companies,” Journal of Business Research, June 1996, pp. 179–188. Chapter Two The New Products Process 41 may be tempted to concentrate on only easy, incremental product projects, or to cut critical steps in the new products process in order to get cycle time down. Cutting corners in technical product development may result in quality sacrifices, resulting in annoyed customers and distributors. By rushing the early steps, the firm may decide late in the process that the product quality is inadequate, which delays the launch, further infuriates dealers, and encourages customers to drift to the competition. Alternatively, rushing through the marketing ramp-up may result in inadequate attention to key marketing tasks in readying the product for launch. In these cases, the firm wins the speed-to-market battle but may lose the war. The temptation to go too fast must be resisted, so that the firm does not mishandle a new-to-the-world opportunity, miss out on key customer information, or develop a technologically inferior product.11 A better way to cope when facing a high-turbulence environment is to keep product development as flexible as possible: Do not freeze the product concept until the last possible moment, but allow later phases in the new products process to run concurrently with concept development.12 This is the principle of postponement, which we revisit in our discussion of the launch phase in Chapter 17. Another related concern is that accelerating time to market might result in bringing the product out too soon, while it still has bugs. In some situations, where there are high opportunity costs and relatively low development risks (such as with a new personal computer), it would be better to speed up cycle time. When Boeing develops and launches a new aircraft, however, there are relatively low opportunity costs (less direct competitors) but much higher development risks. In this case, getting the product “100 percent right” is the more appropriate goal.13 Another risk of focusing exclusively on speed to market is that management might be tempted to concentrate on quick, close-to-home innovations at the expense of really new products, thus putting new product development efforts out of strategic balance.14 11 Christer Karlsson and Pär Åhlström, “Technological Level and Product Development Cycle Time,” Journal of Product Innovation Management, 16(4), July 1999, pp. 352–362; R. G. Cooper and S. J. Edgett, “The Dark Side of Time and Time Metrics in Product Innovation,” Visions, April–May 2002, pp. 14–16; see also C. Merle Crawford, “The Hidden Cost of Accelerated Product Development,” Journal of Product Innovation Management, 9(3), September 1992, pp. 188–199; and Abdul Ali, Robert Krapfel, Jr., and Douglas LaBahn, “Product Innovativeness and Entry Strategy: Impact on Cycle Time and Break-Even Time,” Journal of Product Innovation Management, 12(1), January 1995, pp. 54–69. 12 Marco Iansiti, “Shooting the Rapids: Managing Product Development in Turbulent Environments,” California Management Review, Fall 1995, pp. 37–58. Roger J. Calantone, Jeffrey B. Schmidt, and C. Anthony Di Benedetto, “New Product Activities and Performance: The Moderating Role of Environmental Hostility,” Journal of Product Innovation Management, 14(3), May 1997, pp. 179–189, looked specifically at high-hostility environments. 13 E. G. Krubasik, “Customize Your Product Development,” Harvard Business Review, 66, November–December 1988, pp. 46–52. 14 C. Merle Crawford, op. cit. 42 Part One Overview and Opportunity Identification/Selection FIGURE 2.5 Other Considerations in Cycle Time Acceleration Do the job right the first time. A small amount of time in the early phases can save many times that later, in rework alone. Seek lots of platinum BBs rather than one silver bullet. This means look at every step, every action, every meeting; small savings add up. Train everyone involved. People who don’t know their jobs, who are assigned work without proper skill-building, won’t know how to speed things up. Communicate. Huge amounts of delay can be traced to someone, somewhere, waiting for a piece of information. E-mail and the Internet have made collaboration much easier and quicker, speeding up communication. Be flexible. Look for machines that can do many jobs, people who can switch from one job to another, stand-by vendors, and more. Attitudes too: a new product may require finding a more open-minded designer. Make fast decisions. Managers know that people sometimes get blamed more for things they do than for things they don’t do. Retraining them to make decisions as soon as they reasonably can, and managing them in such a way that we don’t destroy that willingness, is a key step to a fast program. Cut things wisely. There is a common bureaucratic practice of meeting a budget cut of 10% by cutting all of its components 10%. A better method is to take perhaps a 50% cut in noncritical steps, and 0% in the key ones. It’s all risky, but why not take the risk on things that are more forgiving? Some other considerations in cycle time acceleration may be summarized as in Figure 2.5. P&G successfully cut development time on a pharmaceutical product by over 80 percent, while improving quality by over 60 percent, using many of the techniques described above. New product personnel carefully documented all work activities involved in product development and set aggressive goals for time reduction—“This activity should take 50 percent of the time it currently takes” (more modest goals could be easily achieved through only minor improvements). Stretch goals were set: To achieve a 75 percent reduction, one could set a goal of 50 percent reduction in the first year and a further 50 percent reduction in the second year. They also practiced several of the techniques described in this chapter and the previous one: team motivation through clear goal setting, empowerment, and reward mechanisms; and senior management commitment.15 In Chapter 16 we will see other metrics that can be used to complement speedto-market, such as the cash-to-cash metric. Using this tool, the firm will measure not just how quickly the product is launched, but also how long it takes to break even. Using metrics such as this helps the firm manage the whole launch phase, not just the moment of launch. 15 R. W. Boggs, Linda M. Bayuk, and David A. McCamey, “Speeding Development Cycles,” Research-Technology Management, September–October 1999, pp. 33–38. Chapter Two The New Products Process 43 What about New Services? Before we leave the new products process, let’s consider products that seem not to have a technical component to their development—services. Services and goods are often arrayed on a scale of (1) pure service, (2) primarily service and partly a good, (3) primarily a good and partly service, and (4) pure good. Examples, in order, are counseling, insurance policy, automobile, and candy bar. Only in the first category does the product provider have nothing tangible on which to do R&D, and there are very few of them. Even on pure or primarily service products, there are tangible support items such as ads, warranties, policies, and instructions, which need design and production. For an example of a good/service blend, consider a smartphone. The phone itself is tangible, yet it provides services: communication, as well as entertainment (music, games), information (maps), and other applications. The creation of service products tends to mirror the systems used for goods. The strategic elements all fit (the product innovation charter, the new products process, and the balanced portfolio). Perhaps the concepts must be applied creatively, but still the parallels are there. Indeed, a study of successful new services found that these tended to come from companies that used a systematic, comprehensive new service development process with clearly defined phases and regular evaluations and reviews. In fact, according to the most recent CPAS study, the new services process is very similar to the new products process we have presented in this chapter, with only a few key differences. While the basic phases of Figure 2.1 apply, on average, new services take substantially less time in development, and this is true for radical as well as incremental service innovations. For example, a radical new tangible product may require on average over 122 weeks in development, while a radical new service could be completed in about 55 weeks. On other measures, though, services show little difference from tangible products.16 The new products process needs a little refinement to be most useful in service development, mostly due to fundamental differences between services and manufactured goods.17 Services are individualized to the individual customer. Whereas goods are mass-produced, services are provided through interaction between service provider and customer, and the most successful service providers are those that can deliver a “customized” experience to each customer. Services, unlike goods, are also intangible, which means that a key component of the service is indeed the experience of receiving the service. For this reason, the human interaction between service provider and customer is of utmost importance; service 16 Stephen K. Markham and Thomas Hollmann, “The Difference Between Goods and Services Development: A PDMA CPAS Research Study,” in K. B. Kahn, S. E. Kay, R. J. Slotegraaf, and S. Uban (Eds.), The PDMA Handbook of New Product Development (Hoboken, NJ: John Wiley, 2013), Ch. 25, p. 408. 17 The next paragraphs, and the JetBlue example, are drawn from Thomas D. Kuczmarski and Zachary T. Johnston, “New Service Development,” in Kenneth B. Kahn, George Castellion, and Abbie Griffin (eds.), The PDMA Handbook of New Product Development (New York: John Wiley, 2005), pp. 92–107. 44 Part One Overview and Opportunity Identification/Selection providers must strive to meet customer expectation and leave a positive impression. Services are also instantly and continuously being evaluated by customers at every interaction with the service provider. The service provider therefore needs to obtain feedback from the customer and act on it quickly so as to continuously improve performance. Finally, services are often evaluated by customers as the sum of their parts. A family looking back on a trip to a theme park considers the ease of transfer from the airport, the convenience of parking, the number and entertainment value of rides and other activities, the cleanliness of the park, and the friendliness of park personnel when forming an overall opinion of the experience. Poor performance on any of these leads to a lower evaluation of the whole trip. These fundamental differences between services and products pose challenges to the service provider, but the same basic new products process can still be used. Consider how JetBlue has been successful in an extremely competitive air travel market through excellent service development. Rather than focusing on cost-cutting, as many of its competitors have done, JetBlue strove to provide a customized experience, offering travelers free television, friendlier flight crews, comfortable seats, and a simple but useful Web site. JetBlue also is a leader in adding safety measures, such as “paperless cockpit” flight technology and security cameras in the passenger cabin. These safety measures, important to today’s air travelers, help to differentiate JetBlue from competitors. Also, JetBlue obtains opinions (good and bad) from customers throughout the service experience: when the customer is on the company Web page, at the ticket counter, or on the actual flight, JetBlue strives to obtain customer satisfaction information so they can target any areas for improvement and increase the customer’s overall level of satisfaction with the firm. The guiding objective at JetBlue, according to its founding CEO, David Neeleman, was to bring “humanity back to air travel.” This is clearly a service provider that understands the importance of customer interaction! Given the importance of customer interaction in service success, it is no surprise that getting customer participation early is critical to successful new service development. Service delivery personnel—the staff that actually deals with customers, obtains their feedback, and handles their complaints—are in the best position to identify unmet customer needs and are therefore critical players at the concept generation phase. Involving them this early in the new products process increases their motivation and excitement about the new service, which results in more enthusiastic service delivery and more satisfied customers. As the service progresses through development, the best prototype concepts can be taken to customers and tested in product use tests much like those outlined above. Unfortunately, prototype testing is not always well done by service providers. Since services are by definition unpatentable and often easy for competitors to replicate, it is important to ensure that the service has been “tweaked” as much as possible before launch to make sure customers are very satisfied with the offering. A prototype test would be an ideal opportunity to do this kind of tweaking. Finally, the launch phase for services can be particularly challenging. For one thing, services need constant monitoring to ensure they are efficiently meeting customer needs and expectations; this is why the best service providers (think restaurants, hotels, and hospitals) are constantly getting customer feedback. Also, Chapter Two The New Products Process 45 the successful launch of a new service depends greatly on the service delivery personnel training. Coca-Cola employees, for example, rarely interact with the final consumer; by contrast, services of all types are delivered by company personnel (the bank teller, the hotel clerk, the hairdresser, the financial advisor). Excellent training of the service delivery personnel is a key component of any service firm’s customer retention program. A training program will include instruction on the strategic importance to the firm of excellent service delivery as well as lessons on crisis management and troubleshooting, in addition to basic service delivery training. FedEx is a prime example of a service provider that excels in new service development.18 FedEx places the customer experience at the center of its new products process. Customers are involved early in the new products process as co-innovators; this helps FedEx identify their needs early. FedEx has set up market councils comprising executives, sales and marketing personnel, and even corporate lawyers to call on key customers and learn from them. They complement these activities with ethnographic studies (such as the observational techniques to be discussed in Chapter 5) to get at the heart of emerging customer needs. A classic example of this was FedEx’s realization that the customer’s experience would be enhanced if they provided greater access and more digital services. The solution was the 2004 acquisition of Kinko’s (now FedEx Office), which immediately increased the number of shipping points, and at the same time widened service offerings to include photocopying, faxing, binding, and printing. As customer needs evolved (in particular those of small business customers), FedEx was able to grow with them and keep pace. A key factor in FedEx’s new products success was the establishment of the Portfolio Management Team (PMT), a group of senior executives that lead business units and functional areas. The PMT is charged with developing strategic direction, conducting the evaluation tasks in the new products process, and maintaining a balanced portfolio of projects. FedEx has found that the principles stated in Chapter 1 are effective: Following a phased new products process, risk is managed down through time, such that it is effectively reduced by the time costly development and launch phases are reached, and there is a high degree of confidence that a newly launched service will be successful and provide expected returns on investment. For its excellent service development program, FedEx won the 2007 PDMA Outstanding Corporate Innovator award. Service providers can be very successful with creative new service offerings that offer competitive advantages. Noticing the wide variety of languages spoken by its customers, Walgreen’s instituted a Dial-A-Pharmacist service that permitted non-English speakers access to pharmacists that spoke their language using instore phones. The service, which gets about 1,000 calls per month, reduced patient error and boosted customer satisfaction. Since the core benefit of a service is nontangible, the goal often is to improve the overall customer experience, which Netflix did when it initiated its DVD service. Customers appreciated the easy-to-use 18 The FedEx example is drawn from Donald Comer, “How FedEx Uses Insight and Invention to Innovate,” Visions, 31(4), December 2007, pp. 12–14. 46 Part One Overview and Opportunity Identification/Selection Web site and convenient mailing. Finally, Amazon’s Kindle has succeeded where other reader devices have not, because users don’t buy just the reader—they buy the whole service, and this is where Kindle excels. It offers the largest selection of books, an easy-to-use Web site, and wireless syncing to the user’s computer or other devices.19 New-to-the-World Products As seen in Chapter 1, the term new products can refer to new-to-the world products, close-to-home extensions of existing products, or just about anything in between. But the phased process seen in Figure 2.1 may not work as well with new tothe-world products. Research confirms that firms that launch new-to-the-world products incur a significantly lower long-term survival rate than those that enter the market later. But the lower survival rate for a new-to-the-world product is offset by higher profits, since the market for such a product is often larger and can offer bigger profit margins.20 Managers are enticed to take educated risks on newto-the-world products, buoyed by the phenomenal success of products such as Corning’s optical fiber, General Electric’s CT scanner, Apple’s iPhone, and many others.21 Part of the reason for the higher failure rate for new-to-the-world products is that they are difficult to manage. Almost by definition, new-to-the-world products, like the first cell phones or the first personal computers, require discontinuities (sometimes several of them) in order to succeed. Consider the introduction of the personal computer. Contributing to its rapid adoption were discontinuities in technology (computer companies, including some new startups, had to design essentially a totally new computer), in the market (individual homeowners and small businesses now were buying computers, and not just big firms), organizational (personal computers were sold in electronics shops and department stores, not through a professional sales force), and social (millions of people realized how much they needed a computer).22 Much new research is aimed at understanding the management processes that are most appropriate for high-uncertainty, highambiguity environments.23 19 The characteristics and examples are taken from Thomas D. Kuczmarski and Rishu Mandolia, “Service Development,” in K. B. Kahn, S. E. Kay, R. J. Slotegraaf, and S. Uban (Eds.), The PDMA Handbook of New Product Development (Hoboken, NJ: John Wiley, 2013), Ch. 3, pp. 52–54. 20 Sungwook Min, Manohar U. Kalwani, and William T. Robinson, “Market Pioneer and Early Follower Survival Risks: A Contingency Analysis of Really New Versus Incrementally New Product-Markets,” Journal of Marketing, 70(1), 2006, pp. 15–33. 21 Much of this section derives from Gina C. O’Connor, Richard Liefer, Albert S. Paulson, and Lois S. Peters, Grabbing Lightning: Building a Capability for Breakthrough Innovation (San Francisco, CA: Jossey-Bass, 2008). 22 Rosanna Garcia, “Types of Innovation,” in V. K. Narayanan and Gina C. O’Connor (eds.), Encyclopedia of Technology & Innovation Management (Chichester, UK: John Wiley, 2010), Chapter 13. 23 A good reference is O’Connor et al., op. cit. Chapter Two The New Products Process 47 While we still have a long way to go, a good starting point is the recognition that new-to-the-world products must be nurtured. Many business opportunities may present themselves, and some may look promising, but all have uncertain outcomes. Experimentation in the marketplace can assess the opportunities and may even identify new ones. The goal is to consider the innovation, together with the identified business opportunities, and to develop from this a new business model that provides high customer value and, ultimately, is also profitable for the firm. This can be an expensive proposition, and certainly management must not plan on instant decision making, but the due diligence is required due to the uncertainties involved. This process sounds like a business laboratory, and it is sometimes called the incubation period.24 As leading researcher Gina O’Connor says, “Companies do not realize that breakthrough technologies do not yield breakthrough businesses without enormous investment far beyond the technology itself, requiring lots of experimentation on many fronts.” To do incubation correctly, failure must be tolerated, but at the same time learning from the failure so that the firm continues to move toward a successful launch. According to Marissa Meyer, formerly with Google and now president and CEO of Yahoo!, Google’s failure rate of innovative products is over 60 percent, but this is the cost of doing cutting-edge technology.25 It should be noted that incubation is not the same as business development (finding new customers or managing acquisitions). Business development is often done over a one- to two-year time horizon and may be completely done by marketing or management personnel. Due to its focus on business model development for a radical innovation (in an uncertain environment), the time horizon for incubation can be three to five years, and typically technical development, as well as customer and market interaction, is involved. Radical innovation requires a planning approach that acknowledges the unknowns and uncertainties involved. This approach, called discovery-driven planning,26 requires that managers make assumptions about the future in order to build their forecasts and targets, recognizing that these assumptions may be quite wrong. As more information becomes available, the targets are rethought, the forecasts adjusted, and the plan evolves. This is different from the approach more typically seen in less uncertain markets, where past results can be used to build predictable forecasts of the future. A guiding principle in discovery-driven planning is the reverse income statement, which starts from the bottom (required profits) and works backward to required levels of revenues and costs. The firm should also be pursuing a real-options orientation to investment.27 The firm can make low-cost test investments to gather information on the technology and its 24 O’Connor et al., Chapter 4; the quote is from p. 82. P. Sellers, “The Net’s Next Phase,” Fortune, November 13, 2006, pp. 71–72. 26 Rita Gunter McGrath, and Ian C. MacMillan, “Discovery-Driven Planning,” Harvard Business Review, July–August 1995, pp. 44–54. 27 Ian C. MacMillan, Alexander B. van Putten, Rita Gunther McGrath, and James D. Thompson, “Using Real Options Discipline for Highly Uncertain Technology Investments,” ResearchTechnology Management, January–February 2006, pp. 29–37. 25 48 Part One Overview and Opportunity Identification/Selection marketplace potential. The test investment can be thought of as buying an option to continue the development of the breakthrough innovation. If the small investment suggests there is great upside potential, the project is continued to the next phase; otherwise it is terminated. There should be a clear connection between the radical innovation and the firm’s strategic vision as articulated by senior management.28 Without top management encouragement, business units involved in product development will often focus on improving operational efficiency and will therefore be reluctant to accept radically new product projects because of the new product personnel involvement in corporate strategic planning when the environment is very changeable and turbulent.29 In order to move promising radical innovation projects forward, senior management at some firms establishes a transition management team charged with moving an R&D innovation project to business operating status. The transition team receives appropriate funding as well as support and oversight from senior management.30 Obviously, with a new-to-the-world product, it is especially important that the voice of the customer (VOC) be brought in as early as possible, preferably at the beginning of the process. A critical issue here is identifying the right customers to bring in: For example, a medical equipment manufacturer developing a next-generation diagnostic machine might want to partner with the leading research hospitals to determine what performance features need to be built in. Researchers at these hospitals, having found available products to be unsatisfactory, may already have internally developed simple working prototypes of their own. What better guidance could the equipment manufacturer obtain? Identifying and working with these customers is central to lead user analysis, which we will explore more fully in Chapter 5; we will more formally discuss the voice of the customer in Chapter 12. The Role of the Serial Innovator The most recent research on the new products process suggests that some rethinking of the traditional process is required for a firm to be consistently successful at commercializing new-to-the-world products. While many firms are capable of launching radical innovations successfully, few seem to be able to do it repeatedly over a long period of time. Those that can (Apple, Procter & Gamble, Caterpillar, 28 A good general reference on radical new products is Gary S. Lynn and Richard R. Reilly, Blockbusters: The Five Keys to Developing Great New Products (New York: HarperCollins, 2002). An influential article on this topic is Erwin Danneels, “Disruptive Technology Reconsidered: A Critique and Research Agenda,” Journal of Product Innovation Management, 21(4), July 2004, pp. 246–258. 29 Roger Calantone, Rosanna Garcia, and Cornelia Dröge, “The Effects of Environmental Turbulence on New Product Development Strategy Planning,” Journal of Product Innovation Management, 20(2), March 2003, pp. 90–103. 30 Gina O’Connor, Joanne Hyland, and Mark P. Rice, “Bringing Radical and Other Major Innovations Successfully to Market: Bridging the Transition from R&D to Operations,” in P. Belliveau, A. Griffin, and S. M. Somermeyer (eds.), The PDMA Toolbook 2 for New Product Development (New York: John Wiley, 2004). Chapter Two The New Products Process 49 and Intel come to mind) have something in common: serial innovators.31 These are usually mid-level, technical employees who think and work differently and follow their own new products process. In fact, a challenge for senior management is to be able to identify serial innovators (there aren’t many; they are estimated at one per 100 technical employees, or even one per 500), and once identified, to manage them and reward them properly. The problem many firms have with radical innovation is that technology-driven innovation may be very exciting from a technical viewpoint, but doesn’t really solve a customer problem, therefore there is no application that can be brought to market. (In Chapter 3 we will see how the Product Innovation Charter is designed to avert this problem by ensuring there is a market dimension that matches with the technology.) In other cases, a good technology may lack an internal product champion that ensures that the technology goes into the development process (see discussion of product champions in Chapter 14). The reason serial innovators are so good at breakthrough innovation is that they know how to bridge the gap between technology and market. They do this in iterative fashion. Generally, they begin by identifying and fully understanding a customer problem, and then discover possible technical solutions to those problems. They oscillate between customer need and technology solution. As time goes on and the serial innovators learn more about potential customer solutions, they will also bring in market information as well: is there a large enough market demand to justify bringing this technology to market? The “process” followed by serial innovators is arguably not a process at all, since that implies a series of steps and a fixed order. There are several activities that need to be done, but there is no particular order and a lot of recycling and rethinking is necessary and expected. These steps include: • Finding a problem that is important to customers, checking potential market size and revenue stream. • Understanding the problem, including technology, currently available solutions, competition, and customer requirements. • Determining if the problem is interesting to enough customers willing to pay for it, and also interesting to the firm in terms of fitting with product strategy. • Inventing a solution to the problem and checking for customer acceptance with a prototype. • Ensuring that the product goes into development, then gaining market acceptance for the product.32 There is much room here for circling back: if customers don’t accept the prototype, this may require further work to better understand the problem, or even rethinking if the right problem is being solved. It is clear how serial innovators differ from other technology employees. They have a deeper understanding of customers, the firm’s product strategy, and 31 Abbie Griffin, Raymond L. Price and Bruce A. Vojak, Serial Innovators: How Individuals Create and Deliver Breakthrough Innovations in Mature Firms (Stanford: Stanford Unversity Press, 2012). 32 Griffin, Price, and Vojak, op. cit. 50 Part One Overview and Opportunity Identification/Selection political processes, and can act themselves as the product champion. They focus not just on solving customer problems, but on understanding the situation so well, from so many different perspectives, that they find the optimal solution to the customer problem. They can handle discovery, invention, and launch themselves, and therefore are highly valuable to the firm, and allow the firm to be consistently successful with radical new products. So how can a firm recruit serial innovators, or recognize serial innovators among its employees? New products researcher Abbie Griffin suggests that serial innovators have five innate characteristics one should look for: • • • • • Systems thinking (can see ways to connect disjointed information). High creativity (though, interestingly, not exceedingly high!). Curiosity in several areas of interest. A knack for intuition based on expertise. A sincere desire to solve customer problems.33 Spiral Development and the Role of Prototypes In the case of radical product innovation, a fluid, agile new products process might lead to more innovative results. If the final form of the product is truly unknown, it may make sense for the firm to try several prototypes in rapid succession, showing them to customers, getting feedback, trying another prototype, then continuing in this manner until an acceptable form is identified. The term spiral development is sometimes used to describe this process; the name refers to the many iterations between firm and customer. Spiral development can be described as a “build-testfeedback-revise” process: • An early, nonworking version of the product, called a focused prototype, is built (this might be a new cell phone made of wood or foam, or perhaps it is a plastic nonfunctioning prototype that looks real but lacks wires). • The prototype is tested with customers, who express likes, dislikes, purchase intentions, and so on. • Customer feedback is obtained on what needs to be changed. • Based on the feedback, the next prototype is prepared and the cycle continues.34 Note that, essentially, the spiral development process allows phases in the new products process to be done out of order. Early prototypes are built even before customer specifications are determined! Early in 1994, Iomega’s first internally developed prototypes of the Zip Drive (a plug-in laptop drive that allowed the user to access large-storage floppy disks called zip-disks) were plain-looking gray boxes with a flip-up lid (much like many 33 Griffin, Price, and Vojak, op. cit. Robert G. Cooper, “New Products: What Separates the Winners from the Losers and What Drives Success,” in K. B. Kahn, S. E. Kay, R. J. Slotegraaf, and S. Uban (Eds.), The PDMA Handbook of New Product Development (Hoboken, NJ: Wiley, 2013), Ch. 1, p. 14. 34 Chapter Two The New Products Process 51 CD players); early focus groups disliked the top-loading feature and wanted something “personal, portable, and powerful.” Several days later, R&D hit upon the front-loading design (much like a floppy-disk drive); ultimately, over 50 prototypes were built to test many different ideas before one was selected. The Zip Drive ultimately marketed by Iomega featured the now-familiar front-loading mechanism and was indeed personal (attractive blue plastic cover), portable (small enough to fit in a briefcase), and powerful (capable of storing 100 megabytes of data).35 The spiral development process employed by Iomega is sometimes called probe-and-learn: Through interaction with customers, designers are inspired to probe, experiment, and improvise, and as a result, may come up with a successful new-to-the-world product. Another term sometimes used to describe this iterative process is lickety-stick: The developing team develops prototypes from dozens of different new product ideas (“lickety”), eventually settling on a prototype that customers like (“stick”).36 As Mike Santori of National Instruments said, the goal at this early stage is not to determine how to cut costs, but to see what functionality customers are looking for. Rolling out several prototypes quickly and efficiently “gives you flexibility to try out different ideas and audiences.”37 The story of General Electric’s computed tomographic (CT) scanner illustrates the development of a new-to-the-world product with a large assist from the voice of the customer. The original instrument was developed as a head scanner; later versions included a breast scanner and a full-body scanner. In each case, the physicians said the product did not work appropriately. On the fourth try, GE developed the 8800 full-body scanner, which was a huge success when launched, eventually gaining a 68 percent market share. GE did not simply create a solution looking for a problem—that implies having no strategy. GE did have a strategy: Develop a breakthrough scanner technology for medical diagnostic applications, and learn from early trials specifically what applications would be the most valuable to their physician customers.38 Closing Thoughts about the New Products Process As we discussed in Chapter 1, many firms use a new products process much like the one shown in this chapter (though of course the details will vary), and the CPAS studies have consistently shown that the majority of the Best firms implement new products processes and enjoy more success with new products as a result.39 35 For the Iomega story, see Gary S. Lynn and Richard R. Reilly, Blockbusters, op. cit. “Lickety-stick” was coined by Gary S. Lynn and Richard R. Reilly in Blockbusters, op. cit. 37 Quoted in Heidi Bertels, “The 7th Annual Front End of Innovation Conference Adopts a New Format and Content,” Visions, 33(3), October 2009, pp. 34–37. 38 Gary S. Lynn, Mario Mazzuca, Joseph G. Morone, and Albert S. Paulson, “Learning Is the Critical Success Factor in Developing Truly New Products,” Research-Technology Management, May–June 1998, pp. 45–51. 39 Stephen K. Markham and Hyunjung Lee, “Product Development and Management Association’s 2012 Comparative Performance Assessment Study,” Journal of Product Innovation Management, 30(3), 2013, pp. 408–429. 36 52 Part One Overview and Opportunity Identification/Selection Some exciting developments are being noted in the auto industry. Almost half of the automotive engineers in a recent survey said their companies used a traditional new products process, while about a third used a modified process, which allows them to improve efficiency without sacrificing product novelty.40 Using a modified process has allowed as much as a 50 percent reduction in time to market while maintaining new product quality and novelty. The role of senior management cannot be overlooked, especially in the case of radical new products. Speaking at a PDMA meeting, Al Lopez, former vice president of R&D at ExxonMobil, mentioned his company’s many radical new products, including high-strength steel for pipelines, low-sulfur fuel processes, improved catalysts, and so on. Senior management supports productive R&D in several key ways: by recognizing and building the firm’s core technical competencies and capabilities, by encouraging knowledge flow (both internal and external) throughout the firm, by developing effective, streamlined work processes, by clearly linking basic and applied research, and by assuring an exciting work environment in which learning and achievement are rewarded.41 Finally one can ask whether firms can be ambidextrous—that is, be excellent in both new-to-the-world and incremental innovations. Barriers to ambidexterity no doubt exist: Fears of brand dilution, channel conflict, or even a “we’ve always done it this way” culture. There is also the issue of resource allocation: Investing in developing competencies that lead to radical new products may mean divesting those resources away from bettering one’s existing competencies. These are serious concerns of managers at highly innovative firms; to avoid problems of this sort, often a highly projectized venture is spun out as a separate organizational unit to pursue the radical innovation;42 more on this topic in Chapter 14. Summary In this chapter, we studied the system of phases and activities used in the process of developing and marketing new products. We looked at a simplistic version of this process as used by P&G’s cosmetics division and showed how it interacted with P&G’s product portfolio and PIC. We then went through the basic process phase by phase. Be careful: don’t think that this, or any, new product process is etched in stone. It is a guide and an integrator, not a straitjacket. We now turn to Chapter 3, in which we will discuss two more strategic elements: the product innovation charter and managing the portfolio of new products. Chapter 3 introduces the first of the five major phases in the process—Opportunity Identification and Selection. This will include the various forms of strategy to guide the evaluation of available opportunities. That will prepare us to begin the study of concept generation. 40 John E. Ettlie and Jorg M. Eisenbach, “Modified Stage-Gate® Regimes in New Product Development,” Journal of Product Innovation Management, 24(1), January 2007, pp. 20–33. 41 For more information, see Peter Koen, “Tools and Techniques for Managing the Front End of Innovation: Highlights from the May 2003 Cambridge Conference,” Visions, October 2003. 42 Erwin Danneels, “From the Guest Editor: Dialogue on the Effects of Disruptive Technology on Firms and Industries,” Journal of Product Innovation Management, 23(1), January 2006, pp. 2–4. Chapter Two The New Products Process 53 Applications 1. “I’ve got to make a speech down in Dallas next month. It’s part of a conference SMU is having on the general topic of opportunity identification (OI). They want me to explain why OI is sometimes more important than brainstorming and other techniques of concept generation. Seems to me it isn’t. What do you think?” 2. “I work for a financial services firm. We do new product development all the time, and a lot of it is of the incremental variety. You know, bundle credit card access to a savings account, bundle the savings account to a money market account, add an IRA investment option, things of that sort. Explain how the new products process is relevant in my industry, and to my company. Seems like it’s more tailored to physical goods. Isn’t it a little misleading?” 3. “We are increasingly committed to really new products—we see them as the future of our company. Can you explain to me again what the new products process looks like for them? I’m not really convinced that the process you outlined is applicable to them. Seems like it will generate more incrementally new products rather than bold new ideas.” Case: LEGO43 The LEGO group, manufacturer of those plastic bricks so popular with children worldwide, was founded in 1932 in Denmark by Ole Kirk Christiansen. The name LEGO was picked as a short form of “leg goet,” or “play well” in Danish, but quite by coincidence also means “I connect” in Latin, and one cannot imagine a more appropriate name for the little connecting bricks. LEGO manufactured wooden toys for the first few years of its existence, but in 1947 Christiansen bought a plastic injection molding machine, and by 1958, the company was manufacturing the familiar plastic bricks. From 1958 to 1978 sales grew at a steady pace to about $180 million worldwide. By 1978, LEGO had introduced the Castle, Space, and Fabuland play themes and the Technic building system. Buoyed by these and other product innovations, sales really took off, doubling once every five years up to about 1993. Faced with slowing sales in the early 1990s, LEGO began product line extensions, tripling the number of stock-keeping units (SKUs), but these only cannibalized existing products. Overall company sales were unaffected, and the product development efforts ate into LEGO profits. In 1998, LEGO lost money (a company first) and laid off 1,000 workers. Customer and industry research suggested several reasons for the declines in sales and profits. A generation earlier, kids would have played with LEGO until they reached about 10 or 11 years old; by the late 1990s they were losing interest 43 This case is adapted from David Robertson, “Innovation at Lego,” Visions, 35(3), 2011, pp. 10–17; Anonymous, “So What Did Lego Do Anyway,” Visions, 36(1), 2012, pp. 24–25, and Innovation at the LEGO Group, International Institute for Management Development, Case IMD-380, 2008. 54 Part One Overview and Opportunity Identification/Selection earlier. In addition, many children, including the very young, were getting into video and electronic games, and preferred these to building blocks. These consumer trends were unlikely to change soon. Industry changes were also taking place. Most competitors were manufacturing in China, while LEGO still produced bricks at more expensive European locations. The channel power in the toy and game industry had shifted away from manufacturers, and now belonged to huge retailers like Toys ‘R’ Us. LEGO’s patent on the plastic brick was also about to expire, truly leaving LEGO in a “change or die” situation. In 2000, LEGO developed a new mission statement: “To become the world’s strongest brand among families with children by 2005.” This mission statement was designed to boost innovation throughout the company. To pursue the mission and spur the innovative process, LEGO followed the wisdom of the best business consultants and academics. Some of the most notable activities were as follows: • LEGO hired creative people internationally, from Italy to Japan to the United States, for their “Concept Lab” new products center to increase diversity and stimulate creativity. • LEGO created a wide spectrum of innovations including LEGOLAND amusement parks and education centers, and also opened LEGO retail stores. • LEGO pursued open innovation for the first time, partnering with movie producers to develop Star Wars and Harry Potter LEGO sets in addition to Steven Spielberg Movie-Maker toys. • New electronic toys such as Galidor, Bionicle, and Mindstorms were added. Galidor interacted with a TV program of the same name, and Bionicle was paired with a movie. A line of Explore electronic toys was also designed for very young children. • LEGO launched the Digital Designer, with which children could use virtual LEGO bricks to design creations on the computer. • Innovation was valued and rewarded, and the ideas kept flowing. Unfortunately, nothing worked. By 2003, LEGO had lost about $300 million and was almost bankrupt. It was very likely that the company would be sold within a year. The big turnaround to save the company began in earnest in 2003. LEGO sold most of the LEGOLAND parks and its headquarters building. Brick production was outsourced to cheaper locations (Mexico and Czech Republic). The number of SKUs was cut in half, as many of these were redundant (for example, at one time, there were seven different lines of LEGO people, all with slightly different faces). Within a year, LEGO had raised enough cash to stave off bankruptcy, but now a long-term plan was required. The plan centered around innovation, or rather, how LEGO defined and managed innovation. LEGO seemed to have implemented all the right strategies: boosting creativity, opening innovation, looking for disruptive innovation opportunities, and building an innovation culture. All of these activities were investments in the innovation “engine.” With the power of this engine, LEGO was capable of rapid new product development. This did not guarantee successful new products, however. Chapter Two The New Products Process 55 The Concept Lab was charged with developing new products, and not held responsible if its products were not very innovative or very good (hence, adding a seventh line of LEGO people to an already wide-enough product line). According to Professor David Robertson of the Wharton School, what was missing was a guidance system: “Like a rocket car without a steering wheel, LEGO’s innovation engine had launched them down a path at high speed, without the ability to navigate the curves in the road ahead. And like such a car, the end of such a ride is destined to be disastrous.” To be sure, there were successes, such as the very popular Mindstorms, which LEGO could build upon. But to improve the success rate substantially, LEGO would have to institute an effective guidance system, which, according to Robertson, would have to provide answers to three questions: Where are you now, where do you want to go, and how will you get there? What would you recommend to LEGO in order for the company to institute this guidance system? In your opinion, what are the strengths that LEGO could build upon, and where is there room for improvement? How could LEGO use some of the concepts in this chapter to get back on track and institute a long-term plan for success, centered around innovation? Case: Tastykake Sensables44 For generations, Tastykakes have been one of the most popular snack foods in and around the Philadelphia area. The local Tasty Baking Co., founded about 90 years ago, turns out about five million snack cakes, pies, cookies, and doughnuts each day. Sales, however, had been stagnant in recent years. The top sales year was 2001, when Tasty Baking hit $166 million in sales, netting $6 million profit. About this time, new CEO Charles Pizzi announced to shareholders that in 2004, an innovative line extension would be launched. While this would not be the only company action designed to boost performance, the new line would certainly be an important step. In the snack business, the previous decade had seen a major trend toward healthier products. Nabisco SnackWell’s lowfat cakes and cookies were a prominent example from the early 1990s. Even Tastykake had some low-fat products out at this time. By the late 1990s, a newer diet-conscious trend, low-carb products, was emerging, due to the popularity of the low-carbohydrate Atkins Diet. By 2003, dozens of food companies had launched about 600 low-carb products onto the store shelves, and the healthy, low-carb product trend showed no signs of abating. It seemed logical to all concerned that Tasty Baking’s new line would be a low-carb version of Tastykakes. As noted by chief marketing officer Vince Melchiorre, “It was a wave, and we wanted to ride it.” If successful, the low-carb Tastykake could be the first of several new lines targeted at a variety of health concerns. 44 This case was based on Marian Uhlman, “A Trimmer Tastykake,” The Philadelphia Inquirer, May 16, 2004, pp. E1, E8; and Marian Uhlman, “Low-Cal Strategy to Fatten Profit,” The Philadelphia Inquirer, August 10, 2004, pp. E1, E12. 56 Part One Overview and Opportunity Identification/Selection Karen Schutz had about 20 years of marketing experience at Campbell Soup before becoming a marketing manager at Tastykake. In January 2004, she was given the task of making the low-carb Tastykake a reality. The deadline was short: The product was to be out by fall. From her Campbell days, Schutz was aware of the challenge. A new consumer packaged good of this type might require a year to 18 months for product formulation, assessing shelf life, market testing, and advertising planning. As an added constraint, the new line would have to be produced using existing equipment and personnel. By mid-January, John Sawicki, Tasty Baking’s manager of research, obtained the first trial batches of low-carb cookies and doughnuts from an ingredient supplier and arranged for a private tasting by Tasty Baking managers, including Schutz, at company headquarters. Schutz and her colleagues liked the taste (she feared her doughnut would taste like “hamster food”) and agreed that this supplier’s mixes were a good starting point. At this time a code name was selected for the stillsecret low-carb project: Greta (for “Greta Carbo”). Sawicki and his team began development of a low-carb chocolate cookie bar on January 27, 2004. A sugar alcohol called maltitol would be the sugar substitute, and some flour in the mix would be replaced by modified cornstarch. Meanwhile, Melchiorre asked Schutz if it would be possible to make Greta sugar-free. He explained, “I needed to address the issue of people who had grown up on Tastykakes who can’t eat it any more. It was good for business, and good for them.” Schutz knew this would be difficult, as sugar is in milk, berries, and other ingredients. She e-mailed Sawicki to see if this were feasible. Sawicki’s response: “Possibly. It probably depends on the product. Should we be targeting this?” Things were soon going to get exciting for Schutz and Sawicki. In February, senior management rescheduled Greta’s launch to late June—three months earlier than expected. But early research results were promising from a low-carb viewpoint. Early batches of low-carb chocolate cookie bars contained only seven net carbs (the statistic used by low-carb dieters), comparable to Atkins low-carb cookies. Sawicki brought these, as well as chocolate cookies and blueberry muffins, to a taste test attended by Schutz and her co-workers. They liked the taste, but other details like product shape and toppings still had to be decided on. Schutz reminded the team that “people eat with their eyes,” emphasizing that the products had to look good. She also noted that the low-carb cookies, muffins, and doughnuts were planned for preview at the upcoming March 10 board meeting. Considering that the doughnut mix wasn’t ready, the blueberries sank in the muffins, and the cookie bars needed icing, this would be difficult. Somehow, Sawicki pulled it all together for the meeting, even arranging the snacks on serving trays, and the board thanked Schutz and the Greta team for having come so far so fast. Later that day, Schutz was speaking to a supplier, who happened to mention that he could not eat products with maltitol, because it gave him side effects in the lower intestine. As it turned out, some people are more sensitive to maltitol. She spoke to Melchiorre the next morning. He needed no convincing that maltitol levels had to be decreased: He had the same discomfort. In addition, two days after the board meeting, the FDA announced they were going to monitor the usage of Chapter Two The New Products Process 57 terms like “carb free” or “reduced carb” on product labeling; violating companies would face sanctions. Schutz and Sawicki thought fast to solve these problems. A new cookie with polydextrose and glycerin (a sugar alcohol with fewer side effects than maltitol) was in preparation. Portion sizes were also reduced. To avoid the FDA low-carb regulatory web, Schutz decided to position Greta as a sugar-free product, with low-carb being a secondary attribute. The product was also about to get a name: “Sensables” evoked diet moderation and could potentially be reused on other snacks with different health benefits. By May 12, Sensables were introduced to Tasty Baking’s district sales managers. The team was unable to work out problems with the blueberry muffins (too much sugar in the blueberries) and replaced them with orange and chocolate-chip finger cakes. The rest of the lineup was plain and chocolate doughnuts, and the chocolate and chocolate-chip cookie bars. Schutz and her team had tried out the reformulated line, with no intestinal side effects. In her presentation, Schutz stressed the Sensables message: no sugar, low-carb, and portion control, and announced that the product would hit the shelves on July 15. At the end of her presentation, something happened that is rare for a district sales manager meeting: She received a standing ovation. Jim Roche, a Pennsylvania district sales manager, said, “This is a winner.”45 After Sensables were introduced to the sales force, consumer taste testing was undertaken. According to Schutz, a few of the products were modestly “tweaked”: the chocolate chip finger cakes and cookie bars received more chips, and more orange flavor was added to orange finger cakes. The chocolate doughnut was dropped, as customers didn’t like its taste or its appearance. The consumer testing delayed the launch by a few weeks. Once the final adjustments were made to the product line, Sensables were launched in and around Philadelphia on August 10, 2004. The launch received coverage in Philadelphia-area newspapers and on news radio. How does the Sensables process compare with the new products process in this chapter? Would you question anything Tastykake did? Do you think the Sensables line will succeed? Why or why not? Case: The Levacor Heart Pump46 Since 1982, when the first artificial heart (the Jarvik-7) was implanted in the chest of Barney Clark, a major quest of the medical device companies has been to improve the well-being of heart failure patients. The goal for the patient is independent existence. Dr. Clark needed to be attached to a large external machine that powered his mechanical heart, which managed to prolong his life for 112 days. 45 Not everyone was convinced. A local food critic tried the snacks and wasn’t impressed, offering the opinion that devotees of real Tastykakes would not find the Sensables line to be an acceptable substitute. 46 This case was based on information in Reena Jana, “A Smaller, Sleeker Heart Pump,” businessweek.com, January 16, 2007. 58 Part One Overview and Opportunity Identification/Selection Today, of course, the goal is to make the devices as thin as possible, such that they can be implanted in the body and allow the patient essentially to go about a normal existence. In view of the ultra-thin smartphones and other similar products already available to consumers, medical device engineers are eager to use similar technologies to develop slim devices to support the functioning of the heart. According to the American Heart Association, about 80 million adults in the United States have cardiovascular disease of one type or another, and about 5 million suffer from heart failure. When looking at the demand for medical devices to aid the heart, the aging baby-boomer market cannot be ignored. This active age group “wants to live, and demands, a full, rich life. . . . now we have medical consumers, a market that didn’t exist 20 years ago,” notes designer Allan Cameron. This target audience would certainly be receptive to a device that would allow long-term freedom and independence, even if major heart disease strikes. In fact, the heart pump industry is profitable and growing. In 2005, the leading heartpump maker, Thoratec, had annual sales of $201 million on its HeartMate XVE. Analysts see the market as going nowhere but up, especially after Medicare announced that it will be expanding the number of hospitals permitted to do heartpump implants. Recent efforts have been to develop implantable heart pumps that assist the patient’s own heart, rather than mechanical devices that actually replace the heart. One of the most promising of these is the Levacor, which by late 2006 was in development at WorldHeart, based in Oakland, California. By this time, the Levacor had been in feasibility trials in Europe only for a few months; clinical trials in the United States (and ultimate FDA approval) were still far into the future. The most distinctive feature of the Levacor is that it uses magnetically levitated rotary technology to power the pump. The Levacor story begins in the early 1990s at a company called Medquest (since acquired by WorldHeart). Pratap Khanwilkar and his team at Medquest were studying the heart pumps of the time and identified several problems associated with their use. Their size limited their usefulness: A pump that fit into the body of a large man might not be supported by a small adult, teenager, or child. There was also the problem of longevity. The heart pumps needed to be replaced every so often, exposing the patient to the risks and stresses of repeat surgery; this would be a concern especially in the case of a very young patient who might be relying on the pump for decades. The term used in the medical community for an implant that will never need to be replaced is “destination therapy.” Another concern is the actual functioning of the pump: It must be gentle enough not to rupture blood cells, cause as little vibration as possible, and not require much power to operate. The Medquest team settled on magnetic levitation technology as a possible solution. Magnetic levitation involves suspending a rotor using a balance of magnetic fields so that it moves without touching other parts: It literally levitates. Since nothing contacts the rotor, there is no friction or heat buildup, and also no erosion due to wear and tear, leading to longer life. The technology had been used for some time in large-scale projects such as power turbines, but had never been tried on such a small commercial application and certainly never in a heart device. Chapter Two The New Products Process 59 Together with an engineering firm, LaunchPoint Technologies, Medquest developed a small, proprietary magnetic levitation system that could serve to pump blood from the heart throughout the rest of the body. The “suspension in air” of the rotor had a distinct advantage in a heart pump application: since there was less to obstruct blood flow, life-threatening clots would be unlikely to form. The development team designed a three-dimensional version of the pump using computer-aided design (CAD) software, which also was used to make a real-size, clear plastic prototype using rapid-prototyping technology. Using a blood substitute, the team was able to watch liquid flow through the prototype. By early 2006, a working prototype made of a titanium alloy was available, about the size of a hockey puck and one-fourth the size of WorldHeart’s previous model (which did not use the magnetic levitation technology). The device provides full mobility: The pump itself is implanted in the patient’s abdomen, and the external device is a small battery pack and controller that the patient straps on. The first patient, a 67-year-old Greek man, was well enough 50 days after the implant to climb stairs on his own and was released from hospital to live a normal life at home not long thereafter. By this time, WorldHeart and LaunchPoint were also working on an even smaller device designed for babies. As of early 2007, it was still unclear whether the magnetic levitation heart pumps would be the long-term industry standard; however, Mr. Khanwilkar (by now serving as WorldHeart’s vice president for rotary systems and business development) was optimistic. Based on the description in this case, discuss the new products process apparently underway at WorldHeart, in comparison to that outlined in this chapter. How is it similar or different? The launch phase is, of course, still well into the future at the time the case occurs. What are the problem areas the company might face at the time of launch? At the time of the case, what are the uncertainties that still exist? What could the company do now to manage these uncertainties? C H A P T E R T H R E E Opportunity Identification and Selection: Strategic Planning for New Products Setting Chapter 1 introduced us to the strategic elements of new product development, and to the first of these elements, the new products process. Chapter 2 expanded on this process, showing us the phases beginning with opportunity generation and ending with the launch of a new product. Chapter 3 details the first phase of the process, opportunity analysis and strategic planning. It is in this context that we present the two remaining strategic elements: the product innovation charter (PIC) and product portfolio management, since these two strategic elements are such essential parts of this first phase. We will explore in detail the process shown in Figure I.1 in the Introduction to Part I. The first part of this chapter discusses the importance of product strategic planning, focusing on the role of product platforms and also on the process of opportunity identification. This leads up to the second part of the chapter, the development of the PIC. This is essentially the product team’s new product strategy, and it can be thought of as a foundation for new products management that serves as a loose harness for the integration of all people and resources used in generating new products. We look at what a team needs in its strategy statement and then at where its inputs originate—that is, in corporate strategy, in platform strategy, and in influences from many other sources. We explore the components of the PIC—its drivers, its goals and objectives, and its rules of the road. The final part of the chapter presents new product portfolio strategy: the strategic importance of having a portfolio strategy, what the components of a good portfolio are, and how some of the top firms develop their portfolios. Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 61 A Product Strategy for a “Company within a Company” The group of people who lead the development of a new product act as a company within a company. They may be loosely tied together in a committee, or they may be fully dedicated (full-time) managers sent off somewhere in a skunkworks to address a difficult assignment. Regardless of the precise form, the group represents all of the necessary functions. They are led by a group leader, a team manager, or a project manager. They, as a group, essentially do everything the company as a whole would do: develop and allocate a budget, do financial analysis and projections, assign and implement tasks and responsibilities, and so on. For these people, a new products strategy does several things. It charts the group’s direction—where it must go, and where it must not go. It also tells the group its goals and objectives and provides some rules of the road. As new product researcher Peter Koen said, managers of the best firms at product innovation ask, “What sandbox should I be playing in?” before thinking of specific products— much as successful venture capitalists ask first what market areas they should be looking for new businesses in.1 We first explore the inputs to this new products strategy—which we will later define as the product innovation charter (PIC)— then we detail the PIC’s components and explore ways in which it can be built. New Product Strategy Inputs and Identifying Opportunities Corporate leaders make many strategy statements. Figure 3.1 shows a list of such statements, and you can see how important they would be to a new products team. Top-level statements like these guide a whole firm and are parts of what are sometimes called mission statements. Explicit consideration of the role of new products in the organization is strongly related to success. In Robert Cooper’s research, 59 percent of managers from top-performing firms report their new products are a key part of their stated business goals, while only 3 percent of the lowest-performing firms do so.2 Product Platform Planning A product platform is defined as a set of systems and interfaces that form a common structure. It is from this common structure that a family, or stream, of products can be developed efficiently. In simple terms, a product platform can be thought of as a basis for all individual product projects within a family of products.3 1 Peter Koen, “Tools and Techniques for Managing the Front End of Innovation: Highlights from the May 2003 Cambridge Conference,” Visions, October 2003. 2 Robert G. Cooper, Product Leadership: Pathways to Profitable Innovation, 2nd ed. (New York: Basic Books, 2005). 3 The formal definition is from Moreno Muffatto and Marco Roveda, “Developing Product Platforms: Analysis of the Development Process,” Technovation, 20, 2000, pp. 617–630; for more information see also Johannes Halman, Adrian Hofer, and Wim van Vuuren, “Platform-Driven Development of Product Families: Linking Theory with Practice,” Journal of Product Innovation Management, 20, 2003, pp. 149–162. For a good discussion of the top-down platform procedure, see Timothy Simpson, “Product Plaform Design and Customization: Status and Promise,” AI EDAM, 18(1), pp. 3–20. 62 Part One Overview and Opportunity Identification/Selection FIGURE 3.1 Corporate Strengths These are examples of actual corporate strengths that managements have asked be used to differentiate the firm’s new products. Many others are discussed in this chapter. These terms can be used to complete the following sentence: New products in this firm will: Technologies Guidelines Herman Miller: Utilize our fine furniture designers. Braun: Utilize innovative design in every product. Otis Elevator: Build in new levels of service as a key benefit. Coca-Cola: Gain value by being bottled in our bottling system. White Consolidated: Be made on our assembly lines. Lexus: Offer genuine value. Polaroid: Be almost impossible to create. Cooper: Never be first to market. Bausch & Lomb: Use only internal R&D. Sealed Air: Offer more protection with less material. Argo: Copy Deere, at a lower price. Markets Gerber: Be for babies and only babies. Nike: Be for all sports and not just shoes. IBM: Be for all people in computers, not just techie types. Budd: Be specially created to meet the needs of Ford engineers. In many industries, corporate strategy affects product platforms as well as individual product projects. Product platform strategy will affect all projects related to the common platform. So a new products team has some strategies from corporate, some from platforms and other parts of the firm, and some developed by the team itself. A PIC (defined, in these cases, at the product project level) will contain corporate mandates as well as product platform-level mandates. The term platform originally gained usage in the car industry, but now a platform can be a technology, a design, a subsystem—anything that can be shared by one or more product families. Due to rapid market turbulence and the number of product varieties demanded by customers, many firms find that it is not efficient to develop a single product. It can cost as much as $3 billion to develop a new car platform, for example, so carmakers look for ways by which they can spread these costs over several models. It makes sense for carmakers, and other manufacturers, to think in terms of product families that share similarities in design, development, or production processes. Among many others, consider the following examples: • Honda used the same engine and transaxle combination in several vehicles, including the Civic, Accord, and CR-V sport utility vehicles, despite the differences in shapes and sizes of these cars. • From Toyota’s U.S. Camry platform, at least five different made-in-America cars were developed. • Boeing makes passenger, cargo, short-haul, and long-haul aircraft from the same design, with many shared components. Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 63 • Intel develops a microprocessor generation (platform), then concentrates on developing derivative products such as MMX or speed doubler technology.4 These companies are making use of modularization, that is, decomposing complex systems into subsystems or modules. A car, for example, can be decomposed into its engine-transaxle combination, interior, body, dashboard, and so on.5 In these examples, Honda and Toyota combine these subsystems into modular products built on the same platform. The procedure by which these subsystems are combined into modular products is tied to the product’s architecture, which we shall revisit in Chapter 13. Platforms can also be used to gain competitive advantage globally. Firms that take a multinational approach to new products, on average, do better than those that develop products just for their home market.6 There are several strategies for selling one’s products internationally. If customer needs are not that different, one way is to sell a global product (one product is sold worldwide, such as Gillette blades or Canon cameras). If customers differ in their preferences, firms will often resort to a platform strategy. This provides the benefit of standardization and scale economies, while still allowing for adaptation to particular market needs. A firm may offer a “glocal” product (one platform but several product variants adapted to local needs; a single Ford platform may be used to make cars with stick shifts in Europe, but with automatic transmission and standard air conditioning in North America), or multiple glocal products (Volkswagen sells some kinds of cars in Europe and North America, and different lines in Asian markets).7 As examples, consider: • From a common set of ingredients, P&G developed Liquid Ariel, Liquid Tide, and Liquid Cheer for the European, U.S., and Japanese markets, respectively. • Honda’s World Car platform is used to make Accords for the North American, European, and Japanese markets, each slightly different in size according to market preferences. Honda also makes minivans, SUVs, and Acura luxury cars from the same platform. 4 Definition is from William L. Moore, Jordan J. Louviere, and Rohit Verma, “Using Conjoint Analysis to Help Design Product Platforms,” Journal of Product Innovation Management, 16(1), January 1999, pp. 27–39. Examples are from Tucker J. Marion, “Product Modularity,” in Jagdish N. Sheth and Naresh K. Malhotra, Wiley International Encyclopedia of Marketing, Volume 5, Product Innovation and Management (West Sussex, UK: John Wiley, 2011), p. 194; Tamar Krichevsky, “Leveraging and Managing Platforms,” Visions, 24(1), January 2000, pp. 12–13; and Marc H. Meyer and Arthur DeTore, “Perspective: Creating a Platform-Based Approach for Developing New Services,” Journal of Product Innovation Management, 18(3), May 2001, pp. 188–204. 5 Tucker J. Marion, op. cit., p. 194. 6 U. De Brentani, E. Kleinschmidt, and S. Salomo, “Success in Global New Product Development: Impact of Strategy and the Behavioral Environment of the Firm,” Journal of Product Innovation Management, 27(2), 2010, pp. 143–160; see Robert G. Cooper, op. cit., p. 15. 7 Robert G. Cooper, op. cit., p. 15. 64 Part One Overview and Opportunity Identification/Selection The right choice depends on the situation, of course, but it should be guided by concept testing and product testing in each international market; it’s not a safe bet to test the product only in the home market and then hope for the best internationally. It also makes sense to have global product teams with representatives of multiple countries.8 Many firms report promising results. Black & Decker redesigned its power tool groups into product families, allowing for more sharing of components. For example, where once 120 different motors were used in consumer power tools, a single universal motor is now used. The emphasis on platforms reduced product costs by 50 percent and helped Black & Decker achieve the highest market share in the category. IBM used a standard set of subcomponents for all ThinkPad products, cutting both number of parts needed and base manufacturing costs in half.9 The examples illustrate a couple of different ways platforms evolve. In the Black & Decker example, the procedure was bottom-up: The firm found a way to consolidate components within an existing family of products to gain scale economies. But the Sony and Honda examples show a top-down platform procedure: The platform was designed at the outset to become the basis for a family of products, possibly for years into the future. Managers may need to be convinced to commit to top-down platform development rather than development of a single product, as it will certainly be more expensive and time-consuming. But the benefits are the cost and time efficiencies that will be obtained with future products built from the same platform, and ultimately, greater future competitive advantage. (Actually, once Black & Decker successfully converted to platform-based design, it then was able to generate even more products by “reusing” its platforms: It transformed its product design into a kind of top-down procedure.10) Platforms are a possibility in service industries as well as for manufactured goods. From a single platform for managed health care services, one provider offered several derivative insurance products: self-insurance, group insurance, and extra coverage insurance.11 There is a trade-off involved here: Customers (or segments) want distinct products, while common products produce the greatest cost efficiencies.12 To find the best balance, the manufacturer needs to decide on the level of commonality to be attained (that is, which designs or processes to standardize, and which to adapt). Suppose a product team is designing the dashboard for a new line of cars. The desired attributes in a dashboard no doubt depend on type of car being designed. 8 U. De Brentani, E. Kleinschmidt, and S. Salomo, op. cit. Identified industries and examples are from Niklas Sundgren, “Introducing Interface Management in New Product Family Development,” Journal of Product Innovation Management, 16(1), January 1999, pp. 40–51; Krichevsky, op. cit.; and Meyer and DeTore, op. cit. 10 Marc Meyer and Alvin Lehnerd, The Power of Product Platforms (New York: Free Press, 1997). 11 James Walter, “Managing Services Platforms: The Managed Comp Experience,” presentation at the 1999 Product Development & Management Association International Conference, Marco Island, FL. 12 As the U.S. carmakers found out in the 1980s, it may have been more cost efficient for compacts and luxury sedans to share parts and design features, but customers complained that the cars looked too much alike. 9 Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 65 A sports coupe buyer would probably like a roadsterlike dashboard, while a family sedan buyer would prefer a more functional look. Once the key attributes have been identified, the team considers the components of the dashboard: HVAC, electrical, steering system, radio, insulation, and so on, and decides where commonalities could be found. Possibly the electrical and radio designs could be shared, as could some HVAC parts (only the ends of the air conditioning ducts might need to be adapted). To achieve the desired differences between the two types of cars, the steering systems may need to be completely different, as would the insulation system: One might want to design the insulation for the sporty car such that it lets in more road noise!13 Planning for product platforms clearly is difficult work. Excellent crossfunctional communication and serious top management involvement and support are needed to ensure everyone agrees on the platform’s architecture and how it is to be adapted to market segment needs.14 It is also clear that firms can have very different philosophies on platform planning. Within the car industry, for example, Volkswagen might have a single platform underlying its lowest-priced Skoda Oktavia model and its Audi TT sports car. BMW continues to develop each model individually, believing that sharing a common platform would hurt its cars’ appeal.15 Ford established a common platform from which small Fords, Mazdas, and Volvos are produced. Through its strategic alliance with Suzuki, GM gains access to small-car platforms as well as low-cost manufacturing skill; Suzuki obtains access to GM’s low-cost global sourcing of car parts and its skill at alternative energy systems.16 Brand platforms can also be strategically important and are widely used. Brands may be billion-dollar assets, so many brand platforms are personally driven by CEOs. Brands can serve as the launching pad for scores of products, all having in common the brand and any strategies applying to that brand. Think back to how Kellogg’s used the company name as a brand platform and extended the Special K brand to a whole range of diet-related products. Note, however, that any team using a platform brand must conform to the strategy of that brand; in the case of Waterford glass, all products had to have top quality, no exceptions.17 13 David Robertson and Karl Ulrich, “Planning for Product Platforms,” Sloan Management Review, 39(4), Summer 1998, pp. 19–32. 14 Mohan V. Tatikonda, “An Empirical Study of Platform and Derivative Product Development Projects,” Journal of Product Innovation Management, 16(1), January 1999, pp. 3–26. 15 Thomas Osegowitsch, “The Art and Science of Synergy: The Case of the Auto Industry,” Business Horizons, March–April 2001, pp. 17–23. 16 Larry J. Howell and Jamie C. Hsu, “Globalization within the Auto Industry,” ResearchTechnology Management, 45(4), July–August 2002, pp. 43–49. 17 Guidance on using brand platforms for product innovation can be found in Dennis A. Pitts and Lea Prevel Katsanis, “Understanding Brand Equity for Successful Brand Extension,” Journal of Consumer Marketing, 12(4), 1995, pp. 51–64. For a discussion of benefits and issues in platform planning, see David Robertson and Karl Ulrich, “Planning for Product Platforms,” Sloan Management Review, Summer 1998, pp. 19–32. 66 Part One Overview and Opportunity Identification/Selection The value of an established brand is called its brand equity. Market research can measure the value of any brand for any particular market (for example, the Duracell brand, if put onto a line of heavyweight industrial batteries). The measurements actually tell the amount of free promotion and integrity the brand equity brings to a new item that uses it. Estimating this value accurately can be challenging; furthermore, a poor product concept won’t succeed just because of a good brand name, and may indeed damage that brand’s equity. We will return to the issue of managing brand equity in Chapter 16. Another common platform is the category platform, either product type or customer. Most marketing effort today is conducted at category group levels— one overall plan for cake mixes, for do-it-yourself tools, or for finance courses in a college. For example, DuPont has special finishes platforms for doing business with the automobile industry, the marine industry, and the furniture industry, among others. Any strategic change in one of those areas influences all new products developed under that umbrella. Oddly, though Intel has its microprocessors as corporate strategic platforms (remember “Intel inside”?); they are not new product platforms because each is a product, not a group of products. Customers such as Gateway will have the latest Intel chip as a platform for a line of products using it. In sum, any new products team that sets out to develop its own product innovation charter for management’s approval ought to look around for all the baggage that comes from corporate and platform strategies. Most of them would hope they could be so lucky as the teams at Calvin Klein Cosmetics, where the CEO has a rule that there will be “no rules”—he thinks it permitted the first (and very successful) unisex scents. Opportunity Identification Many firms have persons working full time looking for new opportunities. They essentially audit the firm and any environment relevant to it. Throughout the firm people in the course of doing their jobs discover new opportunities—a salesperson learns that a customer is moving into a new market, a scientist finds unexpected activity in a compound, a finance VP notes a fall in the prime rate, a director urges that we look more carefully at what the Environmental Protection Agency is doing. A new regulation, for example, may restrict the use of petroleum-based synthetics, and a CEO may want all divisions to seek new products that actively capitalize on the regulatory change. As shown in Figure I.1, new opportunities can come from many different sources: underutilized or new resources, mandates originating external to the firm (e.g., new regulatory restrictions), or internal mandates (e.g., from new corporate leadership). Figure 3.2 suggests several ways in which firms can identify opportunities for growth in new markets as their existing ones become less desirable. Many futurists advocate studying the emerging trends in society and deriving product opportunities from them. A team of experts from the consulting firm Social Technologies identified six important and provocative modern trends including: Just-in-time life: People like making spur-of-the-moment decisions based on real-time information. Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 67 FIGURE 3.2 Identifying the “Greenfield Markets” 1. Find another location or venue. Once McDonald’s had taken up the best locations for traditional fast-food restaurants, it continued its U.S. expansion by placing stores inside Wal-Marts, in sports arenas, and elsewhere. Starbucks coffee complemented coffee-shop sales by selling its coffee beans and ice creams in supermarkets. 2. Leverage your firm’s strengths in a new activity center. Nike has recently moved into golf and hockey, and Honeywell is looking into casino opportunities. 3. Identify a fast-growing need, and adapt your products to that need. Hewlett-Packard followed the need for “total information solutions” that led it to develop computing and communications products for the World Cup and other sporting events. 4. Find a “new to you” industry: P&G in pharmaceuticals, GE in broadcasting (NBC), Disney in cruises, Rubbermaid in gardening products—either through alliance, acquisition, or internal development. Recommendations for scouting for such opportunities: 1. Look for emerging trends: increased globalization of freight flow meant more global opportunities for FedEx. 2. Find fringe markets that are becoming mainstream: gourmet coffee, extreme sports, home carbon monoxide testing are recent examples that spelled opportunities for many firms. 3. Find bottlenecks in the flow of trade, and seek to eliminate them. Need for better hospital patient record retrieval led 3M to develop its Health Information Systems business. 4. Look for “ripple effects” on business opportunities. The trend toward “immediacy” has led to products such as electronic banking and 24-hour food stores. Health concerns have opened up opportunities in fitness products, vitamins, seminars, etc. Source: Allen J. Magrath, “Envisioning Greenfield Markets.” Across the Board, May 1998, pp. 26–30. Reprinted with permission. Sensing consumers: People can sense their environment better now than ever before; what might be “too much information” for some might be essential information for others. The transparent self: There is more information about consumers available to product managers now than ever before. In search of “enoughness”: Consumers are increasingly adopting simpler lifestyles marked by fewer material possessions and an increasing concern about quality of life. Virtual made real: As more people become accustomed to virtual spaces, the boundary between these and the real world will become increasingly blurred. Co-creation: Due to increases in e-commerce and online communities, it is easier for customers to communicate with each other, cooperate, and share information.18 18 Andy Hines, Josh Calder, and Don Abraham, “Six Catalysts Shaping the Future of Product Development,” Visions, 33(3), October 2009, pp. 20–23. 68 Part One Overview and Opportunity Identification/Selection FIGURE 3.3 Product Opportunities as Derived from Six Societal Trends Trend Related Product Opportunities Trend 1: Just-in-Time Life PhillyCarShare or Zipcar: carsharing systems with hourly rentals. Twitter or related services that allow instantaneous updates about friends. Real-time people tracking services such as Loopt. Trend 2: Sensing Consumers Home-testing kits for cholesterol, allergens, and so on. Technology that allows parents to track their children around the clock. Consumers taking part in environmental sensing networks. Trend 3: The Transparent Self GyPSii displays friends’ whereabouts. Services that generate personal data such as bank accounts. 23andme, a home DNA test (Time Invention of the Year in 2009). Trend 4: In Search of “Enoughness” Products servicing environmental concerns. “Slow food” and “slow life” related products. Products supporting leisure time activities. Trend 5: Virtual Made Real Products and services related to virtual economies. Websites offering avatars for socialization and play in virtual cities. Virtual nightclubs and similar activities. Trend 6: Co-Creation iPhone apps number in the tens of thousands and are still growing. Lego has an online factory for visitors to make their own Lego toys. NikeID for custom shoes, and other similar product “configurators.” Source: Andy Hines, Josh Calder, and Don Abraham, “Six Catalysts Shaping the Future of Product Development,” Visions, 33(3), October 2009, pp. 20–23. Each one of these trends suggests possible opportunities for new product development, as shown in Figure 3.3. As an example, Tremont Electric developed the nPower PEG, a “Personal Energy Generator” that allows the user to charge a phone or other electronic device just by plugging it in and putting it in his or her pocket—the device charges from the kinetic energy generated from walking or running. Product developers at Tremont may have been thinking about the “just-in-time life” trend. If one is traveling or camping, or just forgot to plug in the phone the night before, this product can help the person use the phone again quickly and without waiting to find a plug. It also adds a cost-saving benefit as well as an eco-friendly charging option.19 Truly, there is no end to these opportunities, every one of which may reveal additional opportunities for new products. Unfortunately, each opportunity takes time and money to investigate, so we don’t exploit nearly as many as we would like. Noncorporate Strategic Planning Although the major thrust of strategic planning comes from the top down (i.e., corporate and platform strategy development), much of it also comes from the heads of the functions (silos or chimneys) in the firm—marketing, technical, 19 Tremont Electric’s Web site for the charger is www.nPowerPEG.com. Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 69 FIGURE 3.4 Degree of Innovativeness as a Matter of Strategic Risk Risk Change in use/ user mode Change in operations or marketing mode None Some Great None Low Low Medium Some Low Medium High Great Medium High Dangerous Application: This matrix has gone by several names: Product/Market, Technology/Application, and Market-Newness/Firm-Newness. In all cases, the issue is the risk of innovativeness. Risk on the user side is just as much a concern to us as risk within the firm. Every new product can be positioned on this chart somewhere, and that position is important if it is accepted as a project. Selecting one section to be preferred over the others is a matter of strategy. manufacturing, and finance, and from the planning of suppliers, customers, and others. Such groups frequently have the power to affect new product work. For example, paper manufacturing is done on huge, expensive machines; such firms often have strategies with a statement: “All new items, if paper-related, must be manufacturable on our current lines.” Financial conditions may warrant restrictions such as “no new products that require more than $3,000,000 capital investments.” Suppliers of materials (e.g., chemicals or metals) often require (usually smaller) firms to buy and use what they make. But the greatest functional inputs may come from technical, especially in technology- or supply-push conditions, or from marketing, where ongoing planning uses a range of techniques designed to give sharper market focus and new positionings. For example, look at Figure 3.4. It shows a variation on the traditional product-market matrix. The cells show variations in innovativeness risk as a firm brings in new product types or technologies (operating mode change, across the top) or markets products that require changes in how people buy or use them (use mode down the left side). A simple flavor change (product improvement) would probably involve little or no risk, but substituting a computer line for face-to-face dealings in the field of medicine (diversification for a computer services firm) would involve dangerous risk to the producer of the service (great change in both technology and use mode). 70 Part One Overview and Opportunity Identification/Selection Miscellaneous Sources In contrast to the corporate-platform downward pressure approach and the horizontal functional pressure approach, some inputs can start at the lower level of activity and influence upward, as when a new product is so successful it drives corporate strategy to change. For example, an ethical pharmaceutical firm once unintentionally marketed a very successful new proprietary food product, with the result that a new division was created (to isolate the consumer advertising activity from the rest of the firm) and new strategies created to optimize its opportunity. Sometimes, a slow and gradual restructuring of business practice can influence new product strategies almost without anyone realizing it. For example, managers of service products frequently add a tangible component (FedEx and UPS offer branded packaging materials and even require that their drivers look neat), while managers of tangible products may add or emphasize a service component (such as a car-warranty program). The Product Innovation Charter All of the above inputs (corporate mission, platform planning, strategic fit, and so on) are potentially used in the development of a company’s new product strategy. Because of the importance of this step in driving all that comes later in product development, we advocate a special name for this strategy: the product innovation charter (PIC). Typically, the PIC is a document prepared by senior management designed to provide guidance to the business units on the role of innovation.20 The term PIC reminds us that the strategy is for products, not processes and other activities, it is for innovation, and it is indeed a charter (a document that gives the conditions under which an organization will operate). The PIC can be thought of as a kind of mission statement, but applied at a more micro level within the firm and adapted to new product activities.21 It allows delegation, permits financing, and calls for personnel assignments, all within an agreed-upon scope of activity. For new product teams plowing off into unknown waters, such a charter is invaluable.22 Most firms have a PIC, though it may not go by that name. In fact, some firms claim they have no strategy and then go on to describe methods of project management that are clearly strategic!23 In an empirical study of PDMA member 20 Erika B. Seamon, “Achieving Growth Through an Innovative Culture,” in P. Belliveau, A. Griffin, and S. M. Somermeyer, PDMA Toolbook 2 for New Product Development (New York: John Wiley, 2004), Ch. 1. 21 Christopher K. Bart, “Product Innovation Charters: Mission Statements for New Products,” R&D Management, 32(1), 2002, pp. 23–34. 22 See Robert G. Cooper and Elko J. Kleinschmidt, “Winning Businesses in Product Development: The Critical Success Factors,” Research-Technology Management, July–August 1996, pp. 18–29. An example at Kodak is given in Diana Laitner, “Deep Needs and the Fuzzy Front End,” Visions, July 1997, pp. 6–9. 23 Albert L. Page, “Product Strategy for Product Development,” Visions, July 1997, pp. 15–16. Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 71 FIGURE 3.5 The Product Innovation Charter Background Key ideas from the situation analysis; special forces such as managerial dicta; reasons for preparing a new PIC at this time. Focus At least one clear technology dimension and one clear market dimension. They match and have good potential. Goals-Objectives What the project will accomplish, either short-term as objectives or longer-term as goals. Evaluation measurements. Guidelines Any “rules of the road,” requirements imposed by the situation or by upper management. Innovativeness, order of market entry, time/quality/cost, miscellaneous. managers, about three-fourths of the firms investigated had a formal new product policy of some type (that is, at least a partial PIC), while 29 percent reported having a formal, written PIC.24 A more recent study of senior executives found that innovation rates are substantially higher in cases where the PIC has detailed and specific content, and where there is general satisfaction with the new products process within the firm. The more specific the corporate mission is presented in the PIC, and the more clearly senior management’s strategic directions are spelled out, the better the performance of new products developed by the firm.25 The value of PICs is clearly shown in the CPAS study introduced in Chapter 1. In that study, 86 percent of the “Best” firms had a PIC, as opposed to only 69 percent for the “Rest.”26 The components of a PIC are provided in Figure 3.5. In the PDMA study, well over 80 percent of the firms had formalized at least some of these components. To ensure that the PIC is effective, it should be put in place early by senior management, and the latter should stay involved and not delegate its implementation.27 24 Bart, op. cit. Chris Bart and Ashish Pujari, “The Performance Impact of Content and Process in Product Innovation Charters,” Journal of Product Innovation Management, 24(1), January 2007, pp. 3–19. 26 Gloria Barczak, Abbie Griffin, and Kenneth B. Kahn, “Perspective: Trends and Drivers of Success in NPD Practices: Results of the 2003 PDMA Best Practices Study,” Journal of Product Innovation Management, 26(1), January 2009, pp. 3–23. 27 Seamon, op. cit. 25 72 Part One Overview and Opportunity Identification/Selection FIGURE 3.6 An Illustrative PIC for the Apple iPadi Focus: Apple’s technology strengths include Apple’s operating system, hardware, applications, and services. It also has product design and development strengths by which it can provide products that are intuitive, simple, and fun to use. On the marketing side, the customer wants the newest products on the “cutting edge” that provide seamless integration, high performance, and ease of use. Goals: Revolutionary new products should also be platforms for a line of products into the future. This is a necessity due to the high costs of product development for Apple’s “really new” products. New products should also become the “standard bearer” establishing a leadership position in its market. Special Guidelines: Apple aims to be the best, not necessarily the first, in new product categories. One can see how this PIC would lead to the development of the iPad. Much of the required technology had been previously developed for the Mac, iPod, or iPhone. The product itself would be Apple’s first tablet computer, which was a revolutionary new product that was seen by many to be the “next big thing” in computers. Some skeptics were less impressed, noting that it lacked a “killer app,” but the iPad is designed to be the first of a line of tablet computers with increasing numbers of capabilities and applications. At the time of launch, no one tablet computer had really established a dominant position yet in that industry. Apple’s goal appears to be to have the iPad become the standard bearer of tablet computers much as iPhones have become almost synonymous with music players. i This PIC is speculative only and meant to be an illustration, yet it is realistic and derived from published news reports about the time of the iPad launch. These include: Apple’s 2009 annual report; Scott Steinberg, “Apple iPad Impressions: The Skeptic’s Take,” digitaltrends.com, January 27, 2010; Reena Jana, “Apple iPad Product Development Approach,” blog appearing on the Harvard Business Review site, hbr.org, January 27, 2010. An illustrative example of what the PIC might have looked like for the Apple iPad is shown in Figure 3.6. We will expand the discussion of the PIC soon, but let’s first examine the various inputs that help managers make strategic decisions. Strategy statements take almost as many forms as there are firms preparing them, but they tend to build around the structure given in Figure 3.5. They can be for an entire firm (if very small or very narrowly conceived), or for a standing platform of activity within a larger firm (for example, Black & Decker brand of tools), or for a specific project (for example, Hewlett-Packard’s newest laser printer). A PIC generally speaks to an opportunity (the focus), not to the specific product or products the group is yet to create. Oscar Mayer embarked on the development of the “big wiener” only to find later they needed Big & Juicy in six different flavors for U.S. regions. Of course, when products are very complex (a new car platform, an air express service for the Asian market, or a nation’s new health plan), one product is all the team can handle. The PIC should be in writing, but for various reasons it often is not, and it should be given to all participants, but again it often is not. This is unfortunate, because a secret, mind-only strategy will not do much for a team of 30 people. Why Have a PIC? Think back to Peter Koen’s “sandbox” comment mentioned earlier. We have already seen how many different places we can find opportunities for new product development. Without a strategy, it is easy to lose focus and to spend time and resources chasing the wrong opportunities. The PIC provides the direction, Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 73 or directions, the team should concentrate on in new product development—in other words, it defines what sandbox the team is in, or wants to be in, and also where it does not want to be. Without putting down the boards that define the size and shape of the sandbox (metaphorically speaking), then any opportunity would seem to be a good one! Consider a team developing a small, portable computer printer. One member is thinking of using a new battery-based technology, while another team member is concentrating on potential customers who happen to work in environments where wall plugs are available. Marketing research people plan to pretest the product extensively, while manufacturing engineers assume time is critical and are designing finished production capability from the beginning. A vendor picked to supply the tractor mechanism has to check with the team leader almost every day because the team has not decided exactly what functions the printer will serve or the target user. And the team is being guided by requests from the sales department, which is currently calling on smaller firms although, in fact, the biggest potential is thought to be in large firms and governments. This team has not developed strategy. Team guidance, just like corporate or strategic business units (SBUs) guidance, comes partly in the form of strategy. Its purpose is to focus and integrate team effort and to permit delegation. Bausch & Lomb almost lost its market position when its managers concentrated for too long on improving old products and thus almost missed new products like extended-wear contact lenses. Being forced to review their strategy, they found many more opportunities and went on to capitalize on them (e.g., disposable contact lenses). Lacking a focused and integrated effort, new products teams are likely to face the related problems of scope creep and unstable product specifications.28 Both of these problems occur if the “sandbox” is not defined, or only poorly or vaguely defined. Scope creep refers to the constant changing of a project’s definition: Is the project meant to be a product designed for a specific customer, a large number of users, or a platform for a whole new line of products? Unstable product specifications refer to the product requirements or desired performance level changing as the product goes through the development phase. In either case, the product team is chasing an elusive target (Robert Cooper would call this the “moving goalpost”), with inevitable wastes of both time and resources. A clear-cut PIC, designed to over-arch the entire new products process, helps to minimize these costly and time-consuming problems. There is even value to be gained in the very process of working together and formulating a PIC. A good process fosters high levels of commitment from the participants, consensus on goals and objectives, and agreement on the ways the goals will be achieved. Indeed, in a recent empirical study, the most innovative firms in the sample were those that had a clear PIC and also had a satisfactory process for PIC development.29 This finding complements an earlier study, in 28 For more on scope creep and unstable specifications, see Robert G. Cooper, “What Separates the Winners from the Losers and What Drives Success,” in Kenneth B. Kahn, George Castellion, and Abbie Griffin, The PDMA Handbook of New Product Development, 2nd ed. (Hoboken, NJ: John Wiley & Sons, 2005). 29 Chris Bart and Ashish Pujari, “The Performance Impact of Content and Process in Product Innovation Charters,” Journal of Product Innovation Management, 24(1), January 2007, pp. 3–19. 74 Part One Overview and Opportunity Identification/Selection which 70 percent of the highest-performing firms had a specific PIC, while only 51 percent of low-performing firms did.30 The Sections of the PIC The PIC is a plan for the firm’s initiatives in new product development. It provides clarification of goals and a common language for all personnel involved in new products. The roles of all participants throughout the new products process are clearly specified. An effective PIC communicates to the product team members exactly how their efforts fit into the big corporate picture.31 Background Section of the PIC This section of the PIC answers the question: “Why did we develop this strategy, anyway?” To the extent necessary, it recaps the analysis behind it. The Arena (Area of Focus) Section of the PIC In today’s competitive marketplaces, it takes focus to unlock the necessary power of innovation. Just as a laser can take a harmless light and convert it into a deadly ray, so can a commitment to, say, the delivered-pizza business or to the Web site construction process convert limited resources into a strong competitive thrust. As one developer said, “We like to play on fields that tilt in our direction.” In recent years we have heard a great deal about core competencies. They are an excellent place to start the search for charter arena definitions. Marketers narrow their focus by targeting and segmentation. Technical people, all too often fenced in by time, limited facilities, and money, don’t relish yet another focus mechanism. But the idea of a new products arena, or area of focus, is growing. Focus is generally achieved by the use of four types of strengths or leverage capabilities: technology (such as Kimberly-Clark’s paper processing technology), product experience (Stroh’s chose to focus on the beer business), customer franchise (Stanley Tools’ hold on the woodworker), and end-use experience (Chase Manhattan’s international division). Licensing or acquisition to acquire technologies or market strengths are also fair game for inclusion in strategies. The original Star Wars creator, George Lucas, opened bids from toy makers for licenses when planning Episode 1. Some of them approached $1 billion.32 Relying solely on a technology is risky, as no one knows whether the technologybased product is something customers want (Polaroid’s Polavision instant movies, for example, or the Iridium satellite-based telephone, are examples of new 30 Abbie Griffin, “PDMA Research on New Product Development Practices: Updating Trends and Benchmarking Best Practices,” Journal of Product Innovation Management, 14(6), November 1997, pp. 429–458. 31 Roger J. Calantone, S. K. Vickery, and Cornelia Dröge, “A Business Performance and Strategic New Product Development Activities: An Empirical Investigation,” Journal of Product Innovation Management, 12(3), May 1995, pp. 214–223. 32 Lisa Bannon and Joseph Pereira, “Toy Makers Offer the Moon for New ‘Star Wars’ Licenses,” The Wall Street Journal, August 19, 1997, p. B1. Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 75 technology-based products that did not catch on). Likewise, letting customers’ stated wants drive product innovation is unlikely to work well, except in a market with enormous unmet needs and very slow-reacting competitors. One recent report told how two PC makers differed on drivers. One, Fujitsu, bet on technology and lost, while NEC bet on customer needs and won.33 Gambles like this are too expensive today. So consumer giants Frito-Lay and P&G have major laboratory research facilities, and technology-driven Hewlett-Packard has announced that it wants a strong market commitment behind every new product program. These firms have realized that the best option is a balanced, or dual-drive, strategy. Let’s explore technology and market drivers separately, then see their value when taken in combination. Technology Drivers The most common technological strengths are in the laboratories. Corning used to say it would develop those products—and only those products—that exploited the firm’s fabulous glass technology. Today’s global competition makes it tougher for Corning (and others) to hold a superior position in a technology defined so broadly. Many times, a firm finds it has a valuable non-laboratory technology. Avon has an efficient small-order-handling technology. Other operations technologies include soft drink distributed bottling systems and White Consolidated’s efficient appliance production lines. Big business consulting firms have built new services around capabilities of analysis and interpretation of financial information. For a firm with superior technical skills, applying the dual-drive idea means turning technical specifications into product features that satisfy market needs. Consider a firm that manufactures semiconductors and has developed the capability to produce smaller-size, highly efficient, higher-resistance semiconductors (the technical specifications). These specifications by themselves may mean little to customers or end users, but they do provide capabilities and features, such as longer battery life, lower temperature operation, or lower manufacturing or maintenance costs, that might provide useful benefits to customers. The firm will need to think first of what products might be developed from the basic technology, such as a chip for use in smartphones, laptops, or electric motors. Then, what particular market segments would be interested in such products? Here, the firm will need to match these products, offering these features and benefits, to unmet market needs. Smartphone or laptop users who require longer and more dependable battery life, possibly, or electric motor users who need lighter-weight, lower-cost units that run at lower temperatures. This procedure of converting technical specifications to product features and benefits, to market needs, has sometimes been called the T-P-M linkage.34 33 David T. Methé, Ryoko Toyama, and Junichiro Miyabe, “Product Development Strategy and Organizational Learning,” Journal of Product Innovation Management, 14(5), September 1997, pp. 323–336. 34 Stephen K. Markham and Angus I. Kingon, “Turning Technical Advantage into Product Advantage,” in P. Belliveau, A. Griffin, and S. M. Somermeyer, The PDMA Toolbook 2 for New Product Development (New York: John Wiley, 2004). 76 Part One Overview and Opportunity Identification/Selection Even harder to see are the technologies in marketing. For example, some packaged goods firms view their product management departments as technologies. Other examples include physical distribution systems, customer technical service, or creative advertising departments. Market Drivers The other half of the dual-drive strategy also comes from two market sources: customer group and end-use. The best new product ideas are based on customer problems, and these problems serve as the heart of the concept generation process described in Chapter 4. The Hoover Company once had a strategy of developing new vacuums for people who already had one—the two-vacuum-home concept. (Today they may be working on the five-vacuum home!) Other firms have relied on demographic dimensions for focus, for example, Toro’s young couples and Olivetti’s banks and law offices. As examples of more abstract dimensions, Hallmark famously concentrates on “people who care enough to send the very best.” Welch Allyn, maker of high-tech medical devices that are used in doctors’ offices and hospitals, once said, not jokingly, if you have a cavity we want to see it, and if you don’t have a cavity but need one we will make it. The latter part of that strategic focus brought about their device for doing noninvasive gall bladder removal. Firms producing services find customer-focus comfortable, since many of their operations involve the customer as an actual coproducer of the service. The logic of this arrangement has led many service firms, in all industries, to involve the customer as an integrated partner in the new product development process. Occasionally, a firm can concentrate on one single customer; for example, an auto-parts firm may build new items for Ford or for General Motors. A variation on the single-customer focus is mass customization—where we offer all customers a product of their individual choice. Marriott’s Courtyard, for example, has made this successful in the motel business. We will see more of mass customization in upcoming chapters. The second way of focusing on the market side is on a particular end-use, such as sports or skiing. For example, focusing on skiers or skiing would both provide new equipment, but skiing would also lead to new lodges, new slopes, new travel packages, and services for lodge owners (who may not even be skiers). Industrial firms make great use of end-use. You may say, but how do we know when to focus on the customer, and when on the end-use? The answer lies in the opportunity analysis that took place earlier—you studied markets, people in them, and activities they engage in. You selected a given opportunity because you thought its needs fit the firm’s capabilities. A variation on market drivers is the distributor—when a producer develops new products to meet the needs of, or capitalize on the franchise of, resellers. Hallmark’s line of small gift items, for example, was originally developed to help their card shop franchisees make more money. Combinations: Dual-Drive Now, putting one technical driver together with a market driver yields a clear and precise arena focus. University Microfilms International (UMI) has been using Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 77 the technology of microfilming and the market activity of education as their original mainstay, but later added photocopiers for schools and microfilm readers for law offices. Penn Racquet Sports switched species of markets, putting their tennis ball technology to work making a line of ball toys for dogs.35 Toro had years of success with a series of dual-drives, one of which is global-satellite technology and golf course superintendents.36 The Signode Corporation set up a series of seven new product venture operations and asked each group to select one company technology and one market opportunity that matched that company strength. The first team chose plastics extrusion (from Signode’s primary business of strapping materials) and food manufacturing. This team’s first new products were plastic trays for packaged foods headed into microwave ovens. Goals and Objectives Section of the PIC Anyone working on product innovation ought to know the purpose, because work can change in so many ways if the purpose changes. The PIC uses the standard definition that goals are longer-range, general directions of movement, whereas objectives are short-term, specific measures of accomplishment. Thus, a PIC may aim for market dominance (as a goal) and 25 percent market share the first year (as an objective). Both goals and objectives are of three types: (1) profit, stated in one or more of the many ways profit can be stated; (2) growth, usually controlled, though occasionally a charter is used defensively to help the firm hold or retard a declining trend; and (3) market status, usually increased market share. Many top managements insist that new product teams entering new markets plan to dominate them. But the American Regitel Corporation, marketers of point-of-sale machines, aimed to be number three in its markets, even though the parent firm wanted to be number one as a general policy. There has been lots of criticism of market share as a new product goal, but it is still a popular objective. Wendy’s, Burger King, Dunkin’ Donuts, and Starbucks all rolled out breakfast menus in recent years in order to capture a larger share of the enormous breakfast market that has been dominated by McDonald’s.37 Special Guidelines Section of the PIC Up to this point, we have filled out three sections of the PIC form. We know the team’s arena or focus and we know what they are supposed to accomplish there. But research shows that almost every new product strategy has a fourth 35 Dennis Berman, “Now, Tennis Balls Are Chasing the Dogs,” BusinessWeek, July 13, 1998, p. 138. 36 Richard Gibson, “Toro Charges into Greener Fields with New Products,” The Wall Street Journal, June 22, 1997, p. B4. This article gives lots of details on a very sophisticated use of the dual-drive system of defining an arena. 37 For more on goals and objectives used by business, see Abbie Griffin and Albert L. Page, “PDMA Success Measurement Project: Recommended Measures for Product Development Success and Failure,” Journal of Product Innovation Management, 13(6), November 1996, pp. 169–195. For developments in the fast-food industry, see Bruce Horovitz, “Fast-Food Rivals Suit Up for Breakfast War,” USA Today, February 20, 2007. 78 Part One Overview and Opportunity Identification/Selection section—some guidelines or rules of the road. They may be managerially imposed, or the consensus thinking of team members. They are certainly strategic. We have no research study that shows what such guidelines should be, but we do have lots of research showing what firms put into this section, right or wrong. Degree of Innovativeness How innovative does a management want a particular group to be? The options range from first-to-market (whether a synthetic fiber or a Frisbee) to strict imitation. First-to-market is a risky strategy. It goes by several other names, including pioneering. There are three ways to get it, the first of which is by state-of-the-art breakthrough. Pharmaceutical firms use that route most of the time. Other products that came from such programs include bubble memory, the pacemaker, compact discs, and television. But most first-to-market products do not extend the state of the art; instead they tweak technology in a new way. This second way, sometimes called leveraged creativity, constitutes the most common first-to-market category. For example, DuPont researchers find the special properties (such as durability or oil and grease resistance) of synthetic materials such as Surlyn and Kevlar, then think up creative applications to arrive at new products. Surlyn’s grease resistance led to its application in the meat packing industry. The third way to be first is applications engineering, where the technology may not be changed at all, but the use is totally new. Loctite has done this dozens of times, for example, by using glue to replace metal fasteners in electronics and automotive products. Far more common than pioneering is the strategy of developing an adaptive product. Being adaptive means taking one’s own or a competitive product and improving it in some way. The improvement may be technical (a CD drive for the PC) or nontechnical (the 17-inch PC screen). It may be useful, or trivial. Adaptation is especially popular where the firm needs cash, fast. Some adapters seek almost any change that can be used in advertising. Others follow what is called second but best; the improvement is to be major, and the follower intends to take over the market, if possible. Maytag followed this strategy for many years. Harris Corporation, on the other hand, entered markets where others had pioneered and used its great technical know-how to create a niche with a slightly improved product. The firm’s chairman said Harris tried to be strong in technology and to enter a product in a timely manner. Adaptation alone is risky. The pioneer often obtains a permanent advantage; if other things are equal, the first product in a new market gains an average market share of around 30 percent. But the second firm can take over the market and win the category if its adaptation is clearly superior. Often, the firm that enters first-tomarket will follow the successful entry with less innovative adaptive extensions or even straight imitations, opening up an opportunity window for competitors. The third level of innovativeness is imitation, or emulation. Firms such as Cooper Tire & Rubber, Matsushita, and White Consolidated (appliances) deliberately wait to see winners emerge from among the pioneers and early adopters. Imitation has its risks, too: A firm cannot wait too long to enter the market, by which time the Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 79 earliest firms to enter have well-established, loyal customer bases and ties to the supply networks and distribution channels. Furthermore, an incumbent firm may take an innovator to court over alleged patent, trademark, or copyright infringement; we return to trademark protection in Chapter 16.38 Timing This category of guidelines variation has four options: first, quick second, slower, and late. The decision to be first is pioneering, just discussed. A quick second tries to capture a good second-share position, perhaps making no significant improvement, or just enough to promote. The strategy is very demanding, because such a firm has to make the decision to enter the market before the innovator is successful or has even come to market. This turns the quick second into a forecaster—how successful will the innovator be? Waiting risks letting the second spot go to aggressive competitors. Striving for a slower entry is safer in the sense that a firm knows the outcome of the pioneer’s efforts and has time to make a more meaningful adaptation. But the good market opportunities may be taken by quick seconds. The last timing alternative, late entry, is usually a price entry keyed to manufacturing skills. Miscellaneous Guidelines Innumerable specialized guidelines can be found in product innovation charters. Some are surprising. Hewlett-Packard was trying to decide what to do with its new digital photography for use in printers and scanners. But it took hard selling for the technical people to persuade HP’s printer division it should try to compete with Kodak. An unwritten guideline had banned it for many years.39 Some firms recognize weaknesses. For example, a large mining machinery firm told its product innovators to come up with products that did not require strong marketing; the firm didn’t have it and didn’t want to invest in getting it. A pharmaceutical firm said, “It must be patentable.” A small computer firm said all new products must be parts of systems, while an even smaller computer firm said, “Nothing that must be part of a system!” A food firm said, “Don’t put anything in a can that Frito-Lay can put in a bag.” Another miscellaneous guideline is product integrity, meaning that all aspects of the product are internally consistent. An example: Honda was very successful using the new four-wheel steering system because it put the innovation into a two-door coupe with a sporty image, whereas Mazda failed when putting it on a five-door hatchback that was positioned for safety and durability. 38 For a discussion of pioneering benefits and risks, see M. B. Lieberman and D. B. Montgomery, “First-Mover (Dis)advantages: Retrospective and Link with the Resource-Based View,” Strategic Management Journal, 19(12), 1998, pp. 1111–1125. Information on several later entrants who overtook pioneers can be found in Steven P. Schnaars, Managing Imitation Strategies (New York: Free Press, 1994). 39 Eric Nee, “What Have You Invented for Me Lately?” Forbes, July 28, 1997, pp. 76–82. 80 Part One Overview and Opportunity Identification/Selection How to Prepare a Product Innovation Charter The process for developing a PIC lies in its contents. First, we are always looking for opportunities, inside the firm or outside it. Each strategy can be traced to a strength of the company involved. No one firm can be strong in everything. Second, we have to evaluate, rate, and rank them. Third, we simply begin filling out the PIC form—focus, goals, and guidelines. Usually there is no shortage of suggestions for all the sections—not unlike any marketing situation analysis. Consider first the opportunity identification step. Potentially fruitful options in technologies or marketplaces may seem hard to find, but we are surrounded by them. Figure 3.7 shows a partial list. Every one has been the basis for a team’s new product assignment, at least once. The second step, evaluating and ranking the opportunities, is extremely difficult. In fact, one of the most valuable creative skills in product innovation is the ability to look at a building, an operation, a person, or a department and visualize how it could be used in a new way. This skill can be developed and should be practiced. Not only is there no ready quantitative tool for measuring, say, the strength of the pharmaceutical chemistry department of a small drug manufacturer, there is also politics, because people are involved. And, unfortunately, it is a lot easier to see the potential in some technology or market after the fact. Take Amazon.com, for example. Thousands of people have said how obvious was their idea of selling books on the Internet, but where were they when Amazon.com stock was selling for $10 a share? FIGURE 3.7 Market and Technology Opportunities Market Opportunities Technology Opportunities User (category) User (for our product) Customer (buyer) Influencer Potential user Nonuser Demographic set Psychographic set Geographic set Retailer Wholesaler Agent Use Application Activity Franchise Location Competitor Regulator Product type Specific product Primary packaging Secondary packaging Design process Production process Distribution process Packaging process Patent Science Material Individual Management system Information system Analytical skill Expert system Project control Quality attainment Project design Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 81 To get through these disputes and arrive at a mutually agreeable PIC, a firm needs to do an honest assessment of itself, its goals, its strengths, its customers, and so on. Figure 3.8 illustrates the procedure followed by a real firm, Creative Problem Solving Group—Buffalo, in developing its PIC. The figure shows a set FIGURE 3.8 PIC Development for Creative Problem Solving Group—Buffalo* Who are we, and what do we do? • We provide professional services and resources that help people • Ignite creative potential • Lead creative change • Achieve creative results What technology/core competency do we leverage? • Understanding of creative problem solving, leadership, etc. • Understanding of curriculum design • Understanding of assessment techniques • Human interface and high-touch Who are our customers? • Human resource professionals • Business unit and line leaders • Senior managers • Researchers (potential) • Professors (potential) • New product professionals What are our growth expectations? • 40 percent per year What are their demographics? • Western Europe and North America • Aiming for global industrialized nations • Professionals in organizations (profit, not for profit, government) What are their behavioral traits? • Work with their minds • Interested in research • Personally committed • Bright • Possess integrity and honesty • Willing to collaborate and partner What is our marketing plan? • Target a new level of distribution What do we want to accomplish? • Fill gaps in existing line • Maintain or improve marketplace image • Take existing products to new markets • Enhance brand name How much risk are we willing to take? • Strive for integration and synthesis with current operations How should we time our entry to the market? • ASAP Any other factors important to consider? • Pay attention to patents Based on the above question and answer session, the following PIC was developed: Creative Problem Solving Group—Buffalo provides professional services and resources that help people ignite creative potential, lead creative change, and achieve creative results. These services and resources are used in training courses, learning programs, and creativity and change-related consultancy. Our customers are adult professionals in organizations with interest in affecting change and enhancing creativity. These customers are primarily located in Western Europe, North America, and increasingly in industrialized nations around the globe. Existing and potential customers and associates will be the primary distribution channels. We aspire to grow at least 40% each year by expanding our own efforts and increasing our offering to associates. New products will fill gaps in our existing line as well as improving our image and brand identity. Leveraging our brand and reputation, we will take existing and new products to new markets. Integration and synthesis is our key approach and we will release products to new markets as they are validated. New products will be protected as much as possible by copyright, patent, trade secrets, and trademarks. *The authors wish to thank Len Kistner for providing this detailed example of PIC development. 82 Part One Overview and Opportunity Identification/Selection of assessment questions, together with some of the answers provided by firm’s managers in a PIC workshop. The PIC that was ultimately developed is shown at the end of the figure. At the end of this chapter are two cases involving Kellogg and Honda. In each case, the assignment is to write out what the PIC might have been. Doing that will demonstrate some of the difficulties we have been talking about, because you will have to replicate their knowledge and process. Product Portfolio Analysis: The New Product’s Strategic Fit Now that a new products manager has written a PIC, is that it? Not at all. Upper management must approve it. Importantly, the newly chartered product must fit as part of the firm’s overall business strategy. It should provide an appropriate balance to other products already being offered before any scarce financial resources are allocated to it. Many firms use a product-portfolio approach in which management allocates R&D and other scarce resources across several categories defined by strategic as well as financial dimensions.40 As noted in Chapter 1, virtually all firms consider financial criteria when selecting which products to add to their portfolio. But the best-performing firms also include strategic criteria in their evaluations. This can be tricky: the product team can use standard accounting criteria, such as return on investment, payback period, or net present value, for financial evaluation. But there is no one correct way to do strategic evaluation, which will depend on what is prioritized in the firm’s PIC. While no means an exhaustive list, some common strategic criteria might include • Strategic goals (defending current base of products versus extending the base). • Project types (balancing fundamental research, process improvements, and maintenance projects). • Short-term versus long-term projects. • High-risk versus low-risk projects. • Market familiarity (existing markets, extensions of current ones, or totally new ones). • Technology familiarity (existing platforms, extensions of current ones, or totally new ones). • Geographical markets (balancing sales or profits in North America, Europe, and Asia). 40 For discussions of the portfolio approach, see Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, Portfolio Management for New Products (Hamilton, Ontario: McMaster University, 1997), pp. 59–69; and Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, “Portfolio Management: Fundamental to New Product Success,” in P. Belliveau, A. Griffin, and S. Somermeyer (eds.), The PDMA Toolbook for New Product Development (New York: John Wiley, 2002), pp. 331–364. Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 83 Forthcoming examples show the wide range of possibilities for strategic criteria. Current spending within each of the categories is assessed and compared to desired spending (which may be expressed as a dollar amount or as a percent), and adjustments are made. Thus, a firm would not allocate funds to yet another close-to-home, low-value-product project when it would be strategically more advisable to take on a riskier, higher-potential-return project. Whatever criteria are used, the objectives of developing the product portfolio remain the same:41 • Strategic alignment: most importantly, the portfolio ensures that the mix of products reflects the PIC. Any new projects should be “on strategy” (they support the firm’s innovation strategy and/or are critical to the strategy). • Assessing portfolio value: projects should be selected so that the commercial value of products in the pipeline is maximized. The familiar metrics such as net present value or return on investment can be used. • Project balance: the portfolio should make it easy to select projects that complement the existing product line; for example, too many high-risk projects can be balanced by selecting a couple of lower-risk ones. There should be a nice mix of new-to-the-world products, improvements and revisions, costreducing innovations, and so forth. • Number of projects: one must also consider the number of products in the pipeline, as resource commitments to too many projects inevitably leads to underfunding and gridlock. Resources required by the portfolio should be in balance with the amount of resources available. Managing a strategic product portfolio in order to maintain a dependable, continuous flow of products is a reality in most industries. Consider pharmaceuticals, agrochemicals, or other highly regulated industries. Product managers face incredibly difficult challenges: low likelihoods of success, high development and regulatory costs, limited financial and human resources, even the sheer difficulty of coming up with a good new product idea! Add to this the need to time product launches with marketplace demand, and it’s easy to see why managers in such industries resort to complex decision models to help them manage their product portfolios.42 One SBU within Exxon Chemical uses a strategic portfolio approach for funding allocation, using two dimensions: product newness and market newness (see Figure 3.9). In this figure, the percentages represent resource allocations. If current allocations to, say, improvements to existing products total much more than the desired 35 percent, another product project of this type would be less likely to be funded. The SBU would rather invest in a project of higher product and/or market 41 Scott Edgett, “Portfolio Management for Product Innovation,” in K. B. Kahn, S. E. Kay, R. J. Slotegraaf, and S. Uban (Eds.), The PDMA Handbook of New Product Development (Hoboken, NJ: Wiley, 2013), Ch. 9, p. 156. 42 Gary E. Blau, Joseph F. Pekny, Vishal A. Varma, and Paul R. Bunch, “Managing a Portfolio of Interdependent New Product Candidates in the Pharmaceutical Industry,” Journal of Product Innovation Management, 21(4), July 2004, pp. 227–245. 84 Part One Overview and Opportunity Identification/Selection FIGURE 3.9 Strategic Portfolio Model for One SBU in Exxon Chemical Low Market Newness High Market Newness Low Product Newness Improvements to Existing Products (35%) Additions to Existing Product Lines (20%) Medium Product Newness Cost Reductions (20%) New Product Lines (15%) Repositioning (6%) New-to-the-World Products (4%) High Product Newness Source: Adapted from Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, Portfolio Management for New Products, McMaster University, Hamilton, Ontario, Canada, 1997, p. 63. Reprinted with permission. newness. Both Eastman Chemical and Dow Corning are among the firms that use similar dimensions of technology and market newness to define their strategic categories.43 As another example, Allied Signal has three strategic categories: platform projects, new products, and minor projects, and maintains a portfolio within each category.44 Procter & Gamble uses a phased new products process much like that seen in Chapter 2 in conjunction with a portfolio that includes all its new product initiatives to ensure the right balance and mix of products. This portfolio method permitted P&G to build a pipeline of new products for each of its product lines (face, lips, eyes, etc.) to be released or upgraded at the most appropriate times. Among their recent winning products are Prilosec OTC and Crest White Strips Premium Tooth Whiteners.45 Senior management can also check strategic balance using a portfolio diagram. While a wide variety of dimensions can be used to construct the diagram, the example in Figure 3.10 (similar to one used by a division of Hewlett-Packard) uses extent of product change and extent of process change. Incremental change on both dimensions leads to enhancement products; major change on the product dimension leads to breakthrough (or really new) products. Next-generation products and new product platforms are also represented in the diagram. Too many products in any one region of the diagram represent an imbalance that would have to be rectified.46 43 The examples are from the Cooper et al. book, pp. 62–63. Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, “New Products, New Solutions: Making Portfolio Management More Effective,” Research-Technology Management, March–April 2000, pp. 18–33. 45 Robert G. Cooper and Michael S. Mills, “Succeeding at New Product Development the P&G Way: A Key Element Is Using the Innovation Diamond,” Visions, October 2005, pp. 9–13. 46 Randall L. Englund and Robert J. Graham, “From Experience: Linking Projects to Strategy,” Journal of Product Innovation Management, 16(1), January 1999, pp. 52–64. 44 Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 85 FIGURE 3.10 A Sample Portfolio Diagram Extent of Product Change Large Large Small Breakthroughs Extent of Process Change Next Generation/ Platform Enhancements Small Breakthroughs Source: Adapted from a real portfolio diagram used by a division of Hewlett-Packard, as reported in Randall L. Englund and Robert J. Graham, “From Experience: Linking Projects to Strategy,” Journal of Product Innovation Management 16, no. 1, January 1999, pp. 52–64. Another format is given in Figure 3.11, which is a portfolio evaluation model proposed by the Strategic Decision Group (SDG).47 This method uses expected commercial value (or ECV, measured as the net present value of the future earnings stream) and probability of technical success to build the grid that appears in Figure 3.11. As shown in the grid, four categories emerge. Oysters and Pearls are projects with high ECV. Pearls are projected to have high technical success as well and are therefore highly desirable. Oysters are currently assessed to have a lower likelihood of technical success but potentially are highly profitable; with additional investment, the firm can “cultivate” some of these into Pearls. On the other side of the grid are the projects with lower ECV. The Bread and Butter projects are low risk but low ECV, and typically include incrementally new projects such as extensions and product modifications. White Elephants have low ECV and low probability of success and should be avoided. This portfolio model stresses balance between the three desirable categories. As with other portfolio models, the Strategic Decision Group model alerts the firm if it is investing too heavily in incremental Bread and Butter projects, or if it has taken on too many risky Oyster projects. 47 See Robert G. Cooper, Winning at New Products: Accelerating the Process from Idea to Launch, 2nd ed. (Reading, MA: Addison-Wesley, 1993), pp. 184–185. 86 Part One Overview and Opportunity Identification/Selection FIGURE 3.11 Strategic Decision Group Portfolio Evaluation Model Probability of Technical Success High Pearls Bread and Butter High Low Expected Commercial Value White Elephants Oysters Low Finally, a multiple-objective strategic portfolio model, recommended by product portfolio expert Scott Edgett, is shown in Figure 3.12. This one is a kind of extension of the model of Figure 3.9 in that it includes additional objectives to be considered in project selection. In Figure 3.12, a hypothetical SBU will allocate about 18 percent of its resources to disruptive projects: these should number about 10 percent of all projects and will result in about 22 percent of incremental sales. That is, disruptive projects will be a little more expensive on average, but will also have higher-than-average payoffs in terms of sales. The percentages are interpreted the same way as in Figure 3.9: they are targets for the SBU in making FIGURE 3.12 Portfolio of Product Types DISRUPTIVE INNOVATION (New-to-world or new-to-firm) Number of Projects Resource Allocation Incremental Sales 10% 18% 22% TECHNOLOGY UPGRADE (Next-generation) 12% 22% 40% PRODUCT LINE EXTENSION (Additions to product line) INCREMENTAL INNOVATION (Improvements to existing products) 32% 25% 15% 46% 35% 23% Source: Adapted from Scott Edgett, “Portfolio Management for Product Innovation,” in K. B. Kahn, S. E. Kay, R. J. Slotegraaf, and S. Uban (Eds.), The PDMA Handbook of New Product Development (Hoboken, NJ: John Wiley), 2013, Ch. 9, p. 162. Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 87 project selection decisions, and if the number of disruptive projects in the pipeline falls below 10 percent, they should be prioritized. Persistence in applying portfolio techniques is rewarded: Many firms report that their portfolio efforts are weakly implemented, resulting in an unhealthy preference for incremental projects and inefficient resource allocation.48 Summary Chapter 3 has dealt with the most important and difficult step in the entire new products process: developing a sound strategy to guide the company within the company—the subset of people and resources charged with getting new products. Strategy turns such a group into a miniature firm, a microcosm of the whole. We looked at what such strategic guidance might be—a format here called a product innovation charter. We then studied the opportunities and mandates that yield the charters and how the charters can vary. The chapter ended by looking at some important issues that often arise when discussing new product strategy. We can now begin the study of concept generation—the subject of four chapters in Part II. Applications 1. “I’m afraid I don’t follow your reasoning very well when it comes to this matter of innovativeness—being a pioneer, an adapter, quick second, and so on. Seems you’ve always got to come up with something new, or it simply won’t sell. I believe we agreed on that earlier when we discussed the concept that winners market unique, superior products. Further, if you’ve got something new, why in the world would you ever want to be less than first to market with it? You’ll lose your uniqueness that way. Sounds like you’ve taken a simple practice and made it complex.” 2. “Somewhere along the line, R&D gets the short end of the stick. Now, I know about the arguments for strategy, but I really do feel that R&D deserves a better shake than to simply be told to do this or that. Some of our top people are in R&D—our electronics division has a couple of the world’s best fax technicians. If I were doing it, I think I would have R&D prepare the first draft of a PIC, at least their areas of a PIC, and then have other areas like manufacturing add to it. When all of the interior departments have their sights properly set, I would ask marketing to reconcile the PIC with the marketplace. Otherwise, we’d have the tail wagging the dog when it comes to the new products function.” 48 Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, Improving New Product Development Performance and Practices, Benchmarking Study (Houston, TX: American Productivity and Quality Center, 2002). 88 Part One Overview and Opportunity Identification/Selection 3. “I saw the other day where filmmakers (large ones as well as small ones) are finding profits in low-budget films. It seems they aim for narrow, but very reachable market segments (e.g., young kids), and they use standard filmmaking technologies but use only what they call emerging actors and directors (meaning cheap now). They try hard to capture the interests of their core target group, and they mean it when they say low-budget. I also read where several of them are trying to move out rapidly from the core when they have a winner: little kids, bigger kids, and so on. They think this approach yields the best return on investment even though it causes them to miss out on the occasional blockbuster winner. Some of these low-budget specials included American Pie, There’s Something About Mary, and The Wedding Singer. That last one focused on boys and men, but they added a love story line with Drew Barrymore that brought women in too. Now, can you fit all that into what might be the PIC of these films? What are the negatives of this approach?”49 Case: New Product Strategy at Kellogg50 The Kellogg Company of Battle Creek, Michigan, has been producing cereals since 1906. With reported annual sales in the $9 billion range, it is a leading manufacturer of cereals and other convenience foods (cookies, crackers, toaster pastries, frozen waffles, etc.). Aside from the familiar Kellogg’s cereals, company brands include Keebler (acquired by Kellogg in 2001), Kashi (acquired in 2000), Pop-Tarts, Eggo, Famous Amos, and Morningstar Farms. Kellogg advertising has long featured cartoon spokespersons such as Tony the Tiger, Toucan Sam, and Snap, Crackle, and Pop, the Rice Krispies elves. Kellogg manufactures in 19 countries and has a presence on store shelves in over 150 countries. Many Kellogg’s brands are particularly popular in Europe, where peanut- and chocolate-flavored corn flakes coexist on store shelves with the original recipe. Kellogg, long the number-one cereal maker, was bypassed by General Mills in 1999. Many factors seemed to contribute to Kellogg’s loss in cereal market share (from over 40 percent to about 31 percent): few successful new product introductions, high prices, and slashed advertising budgets. Meanwhile, its many competitors were thinking up new ways to compete: General Mills successfully launched line extensions such as Honey Nut Cheerios; Post focused on the adult market; and Quaker slashed prices by switching to bag packaging. 49 This application is taken from Bruce Orwall, “Hollywood’s Champs: Cheap Little Flicks,” The Wall Street Journal, November 11, 1998, p. B1. 50 This case is based on Keith Naughton, “Crunch Time at Kellogg,” Newsweek, February 14, 2000, pp. 52–53; Stephanie Thompson, “Kellogg Has Megabrand Ambitions for Special K,” Advertising Age, November 6, 2006; Anonymous, “Adwatch: Kellogg’s Special K—Drop a Jeans Size,” Marketing, February 1, 2006, p. 21; Anonymous, “Special K Seals David Lloyd Tie,” Marketing, July 5, 2006, p. 6; Lawrence C. Strauss, “Barron’s Insight: Kellogg Seems Underrated,” Wall Street Journal, January 8, 2006, p. 2; and on information available at www.kellogg. com. Many thanks to Geoff Lantos who provided additional material to update this case. Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 89 Kellogg CEO Carlos Gutierrez, who took over in 1999 (and who later became U.S. Secretary of Commerce), planned to respond to the competitive challenges by getting Kellogg to rethink its corporate strategy. Under his leadership, Kellogg has turned to a greater emphasis on snack products while not ignoring its core cereal products. It is building its traditional cereal business through heavy advertising and promotions (such as a Special K sweepstakes offering a chance to meet model Cindy Crawford). At the same time, however, it is leveraging its well-known cereal and snack products to increase its presence in the snack food business. New snack products are being spun off familiar cereal brand names (Snack ’Ums are largesize Froot Loops that come in small cans), and new flavors of familiar snacks are constantly being tried out (such as butterscotch Rice Krispies Treats and S’Mores Nutri-Grain Bars). The 2001 acquisition of Keebler also quickly increased Kellogg’s snack portfolio. Gutierrez predicted that cereal would soon make up less than half of Kellogg’s business, as the company increasingly pursues the convenience foods market with products such as Nutri-Grain Bars. As Gutierrez said, “People snack— that’s the way the world is moving.” By year-end 1999, Kellogg’s snack food lines were already showing substantial sales and profit increases. New product development efforts at Kellogg have focused on new snack items, many being line extensions of familiar Kellogg snacks. Among the items tested were kimchi- and seaweed-flavored Rice Krispies Treats (both aimed at overseas markets, neither ultimately launched). Another was Krave, a refueling snack bar intended as a midday snack. Krave was to be supported with a $2–$3 million advertising budget; on the packaging, the “K” in Krave was written as the familiar red-script Kellogg’s K. Snack ’Ums (see above) made the cut. At the same time, major cereal brands such as Special K are being supported with significant advertising and sales promotion budgets: At one time, Beanie Babies were stuffed inside Kellogg cereal boxes. By 2002, Kellogg had regained the leadership position, largely due to its successful new products and shrewd marketing investments. In 2003, more new cookies were launched: E. L. Fudge in Butterfingers and S’Mores varieties. In keeping with the line extension strategy, new cereals included Special K Red Berries and Special K Vanilla Almond, as well as Maple Brown Sugar Frosted Mini Wheats, Smart Start, and Tony’s Cinnamon Crunchers. In 2004, low-sugar Frosted Flakes and Froot Loops were launched, as well as Fruit Twistables (a snack). Kellogg has also got into licensing in a big way recently with SpongeBob SquarePants items such as Pop-Tarts, Eggo waffles, and Cheez-Its. By 2006, Kellogg was focusing on Special K as a “mega brand” under which it launched several new snack products (Snack Bites, K2O Protein Water, and Protein Bars) as well as fruit- and yogurtflavored Special K cereals. These launches have been tied to aggressive promotions aimed at the calorie-counting crowd. During 2006 it sponsored a “Drop a Jeans Size” ad campaign, ran a joint promotion with David Lloyd health clubs in Britain, supported a special “get-fit” Web site, and offered Special K Personal Trainer watches. Noting the obsession with healthy eating in North America and elsewhere, Kellogg also tied its promotion of All-Bran to better digestion. Given what you know about the cereal industry and the information provided above, choose one of the products or product lines marketed by Kellogg (it can 90 Part One Overview and Opportunity Identification/Selection be an older brand or a new release) and try to write out its PIC for 1999, around the time of Gutierrez’s assumption of the CEO position. Follow the format of Figure 3.5. Be sure to flesh out all components of the PIC, and include in the background section of your PIC what higher level strategic plans may have been in place. Comment also on how Kellogg’s PIC seems to have evolved from 1999 to 2006. As a further challenge, try to think of future products and/or product lines that would be consistent with your PIC and might be appealing directions for new product development at Kellogg. Case: The Honda Element51 Honda, like most automakers, is an expert in the use of product platforms. This case takes you through all new products process phases, highlighting how Honda applied its expertise in product platforms to develop a cost-efficient new light truck, the Element, that was highly appealing to the targeted market segment. The development of the Element began in 1998 with an idea for a new kind of light truck. At the time, Honda was already producing several lines of light trucks and SUVs, including the CR-V, Pilot SUV, and Odyssey minivan. At that time, a new cross-functional team was charged with developing a new light truck to add to this line, targeting a different customer segment and usage situation. In particular, the target was Generation Y males (aged 19–29) about to buy their first car. Gen Y was a potentially lucrative market: It was a sizeable segment, almost as large as the “baby boom” (individuals born between 1946 and 1964). Also, 52 percent of firsttime car buyers were in this demographic. In the original business model, Element sales were forecasted to reach about 50,000 units in the first year. This number was based on comparison against CR-V sales, which reached about 100,000 per year in North America. Senior salespeople at Honda recognized that several of their cars and light trucks were popular with young women or with families, but nothing appealed to young men. Honda also knew that several competitors had SUVs in the $20,000 price range that appealed to this segment. Getting loyalty at an early age has always been a strategy of automakers, as they expect that customers will trade up to more expensive or luxurious cars in the line as they become more affluent. For example, an Element buyer might trade up to an Accord, then an Odyssey, through time. Honda was clearly using demographics as a segmentation base and identifying a segment with very high growth potential. The original charge of the product team was to develop a compelling new design that target users would respond to, while keeping the retail price affordable. Therefore, the first task was to try to understand the core values and beliefs of this unfamiliar segment. Ethnographic “fly-on-the-wall” research was conducted at the X-Games, featuring competitions in extreme events such as hot-dog skiing, snowboarding, and dirt-course motorcycle racing. Researchers with camcorders watched X-Games participants and spectators before, during, and after 51 This case was derived from Marc H. Meyer, “Perspective: How Honda Innovates,” Journal of Product Innovation Management, 25(3), May 2008, pp. 261–271. Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 91 competitions. Later analysis of the videos provided a clear picture of the young males in the target market: They exhibit strong cohort identification, support social and environmental causes, are well educated, and tend to be less career driven than older segments. These observations provided clues to Honda designers on what features would need to be built in to appeal to this target. For example, typical users of this age group would need a vehicle that provided flexibility: it should be able to easily carry sporting equipment, dorm room furniture, or plenty of friends, and could even serve as sleeping quarters for weekend trips. Product planners recognized that the light trucks currently in the line each had a clear positioning statement. The CR-V was for single, active individuals or small families; the Pilot was for larger families; and the Odyssey appealed to more settled families. The Element could fill a gap in the positioning map: the light truck for the single individual with an unconventional lifestyle. Designers realized they would have to build flexibility into the Element’s design. It would need a unique appearance and would also have to provide a fun driving experience. In all, four design themes were identified for the Element: adaptability/modularity, authenticity, functionality, and attitude/expression. These were added to the three design themes that drive development of all Honda cars: performance, safety, and value, to get the seven design themes that guided designers and engineers working on the Element. Several different activities were then conducted simultaneously. Designers sketched several new versions of a bold new exterior appearance. Meanwhile, engineers worked on building in adaptability, focusing on fold-away seats that provided plenty of cargo or sleeping space when folded. Side doors were attached in such as way as to permit easier entry and exit, and the tailgate was also redesigned in a “clamshell” shape to improve access. A removable moon roof would allow the user to carry a tall piece of furniture vertically, with the top part sticking out. Armed with sketches of their progress so far, team members (both engineers and marketers) visited several universities and met with male students at frat houses. After obtaining feedback, they made adjustments and were able to achieve many “quick-turn” improvements. To get top management support for the Element, the product team invited Honda executives to San Onofre Surf Beach in California, together with several Gen Y university students, for a weekend camping trip. The group discussed Gen Y lifestyle as well as car issues. The team felt that top management would support the project if they “lived the life” of the target user. It worked. The top executives were convinced of the value of the Element to the Honda car line, and the project got approval. A launch date of late 2003 was chosen. Once the project was approved, stylists updated their sketches, quarter-size clay models were built, and eventually full-size prototypes were created and submitted to top executives for approval. At the same time, a user group of 30 men in the target age group, all living near Honda’s Design Center in Torrance, California, was selected. They also reviewed sketches and prototypes, and gradually a design that this group found really interesting was finalized. Here is where Honda’s platform experience was put to use. New car product development is usually broken down into subsystems. In the case of the Element, 92 Part One Overview and Opportunity Identification/Selection four subsystems were used: exterior, interior, suspension, and power train. For each, a design strategy was created, and work progressed with periodic review by top management. The exterior subsystem consists of frame, bumpers, windshield, sunroof, tailgate, and so forth. Many of these components were specifically designed for the Element target segment, such as the unique side doors and the clamshell tailgate. The exterior panels were also designed with extra durability. In short, the Element’s exterior was different enough from other Honda autos that it had to be designed uniquely, from the ground up. Similarly, the interior was a unique design. The driving principle behind the Element’s interior design was the flexibility in cargo storage. The seats could be easily reconfigured into many different positions, or removed entirely. It was also expected that sand or mud would likely find its way into the storage area, so easy cleaning would be required. The flooring was urethane-coated, and electronics were located above the floor or put into waterproof barriers. Even waterproof seat fabric was used. There was little need to develop a totally unique suspension for the Element, however. The ride needed to be maneuverable, sporty, and fun, and the current CR-V suspension would not have delivered the desired benefits. Honda engineers solved the problem by combining the basic CR-V chassis with the power steering gearbox used in the CR-V, MDX, and Pilot, making the Element wider and lower to the ground, and adding wider tires. Finally, for the power train, they used the existing 2.4 liter VTEC (variable valve timing and emissions control) engine, specifically adapted for the Element to deliver 160 horsepower at 5,500 RPM—plenty of power for the target customer. This engine also provided 26 miles per gallon (highway rating) and met all California emission standards. Since the power train accounts for about 20 to 30 percent of each car’s cost of goods, Honda has historically invested in excellent power trains; product teams such as the Element team are actually not authorized to design new power trains but indeed must work with Honda’s central Power Train Group. This same engine was used in the 2002 CR-V and Acura RSX, as well as the 2003 Accord. Together with the Element, four different products were supported by the same engine, and any advances made by the Power Train Group benefit all of these products. In summer 2003, initial manufacturing runs began and early versions of the Element were delivered to dealerships. Marketing worked on finalizing the brand name; “Element” was the favorite of the user panel and also in research studies with prospective buyers. Communications had to be carefully chosen, given Gen Y’s notorious aversion to traditional advertising. Honda selected a more grassroots approach, creating buzz in auto enthusiast groups, at auto shows, and at colleges. Honda sponsored surf events and tailgate parties at universities, highly unusual for an automaker. More traditional television advertising used a lifestyle theme, showing groups of young Gen Y friends going to the beach or to a party. The product team’s hard work paid off. The Element was named Automobile Magazine’s small SUV of the year for 2003, and sales have been good—2004 sales reached 75,000 cars, substantially above the forecast. The biggest surprise was that the Element proved popular across all age groups: 40 percent of Element buyers were in their mid- to late-30s, and baby boomers also bought the Element in large numbers. Still, the buyers were mostly (not totally) male, and lived more active Chapter Three Opportunity Identification and Selection: Strategic Planning for New Products 93 lifestyles than typical Civic buyers. Older buyers seemed to like the fact that it was clearly a young person’s car. Comment on the factors leading to the success of the Element. Include Honda’s platform strategy as well as any other aspects of the new products process that you feel are relevant. In your answer, try also to work out what the product innovation charter (PIC) might have been for the Element. What tangible benefits resulted from bringing in the voice of the customer? What could be learned from this case for firms in industries other than automobile manufacture? 94 Part One Overview and Opportunity Identification/Selection FIGURE II.1 Concept Generation From Figure I.1 (Opportunity Identification) Prepare for Ideation Establish a team or nucleus for the ideation and screening stages Train and otherwise prepare the ideation team Problem Identification In-depth analysis of markets, primarily through some form of problem analysis or scenario analysis Solve Problem Technology Seeks possible solutions May have found a new technology The team Analytical problem solving Surprise problem solving Collect concepts from others in the organization End-user Offers possible solutions May have a working prototype Collect concepts from outside the organization Pool of new product concepts, from vague ideas to working prototypes To Figure III.1 (Concept/Project Evaluation) P A R T T W O Concept Generation In Chapters 1 and 2, we saw the overall new products process and learned that the first phase is all about strategic planning—the rationale being that one should seek new products that are best for the particular firm. Ideation goes on constantly. Many employees of every organization come up with new product possibilities, and the act of creativity could never be constrained into a diagram. But there are common patterns, and we manage those. Look at Figure II.1. Starting at the top, we see Prepare for Ideation, the topic of Chapter 4. People inside and outside the firm don’t hold up ideating until we “prepare,” of course, but managed creativity is much more successful if we assign much of the task to people with strong creative capabilities. Then, early on, we want to focus on problems and needs. So, by one means or another, we try to identify and clarify one or more specific problems that creativity can be focused on. Identifying problems and finding creative ways to solve them are the subject of Chapter 5. Most of what follows in Part II does just that, but there still is a lot of freelance ideation going on. Activity takes place in five areas, shown in the figure. On the left, most firms have a technology operation (R&D or engineering) in which completely new technologies are being sought. Technical people are also on hand to help solve problems identified earlier. On the right side, end users (indeed, all stakeholders in the marketplace) also do freelance ideation, and some of them actually design their own products, produce prototypes, and put them to work. For example, a dentist or an X-ray technician may well conjure up some device this way. They also stand ready to help us solve problems we identify. In the meantime, in the middle of the diagram, the in-house team or group of people working on this project do their own problem solving. And they engage in other activities (Chapters 6 and 7) that produce “surprise” products. These, of course, are not problem-driven, so they must find out if someone has a problem that fits the solution. While all of this is going on, people everywhere are telling us about their ideas: employees throughout the organization, their families, even complete strangers. They come in on the left and right sides of the figure, lower down. The consequence is a pool of ideas, and filling this pool is the subject of 96 Part Two One Overview Concept Generation and Opportunity Identification/Selection Chapters 4–7. We will take up the issue of evaluating and refining these concepts in Part III of the book. Ideation is a huge topic, and there are hundreds of methods. The best are here, and a set of others often used is in Appendix B. What works on a pizza would not work on a fiber-optic sensor. And nothing in the world of creativity lends itself very well to research, so what most firms do is what satisfies them. C H A P T E R F O U R Creativity and the Product Concept Setting This chapter takes us through several topics. First, to managers comes the task of preparing the firm for ideation—the first step in Figure II.1.1 This means getting the right people, putting them in the correct environment, and generally getting them ready for the ideation process. Second, a creative person needs to know what is being searched for—that is, what is a concept and how is it typically found and identified? Third, you will explore a specific system of active (not reactive) concept generation, including approaches that seem to work. One part of that system— using employees and nonemployees in a search for ready-made ideas—will be discussed in this chapter, and the others will follow in Chapters 5, 6, and 7. Preparation Many people think of product innovation beginning with a new product idea. But Chapter 3 showed that it is far better to select a playing field and some rules (have a strategy) before starting the game. The Product Innovation Charter Think about these items from a hypothetical charter (Chapter 3) in a firm making bathtubs: • Our new product concepts should be useful to older people and others with physical handicaps. • New products coming from these concepts must make use of the firm’s strong design capabilities, as well as copper metal. Assuming the PIC work was well done, any person trying to come up with new bathtub ideas for this firm had better know the game plan, or many ideas created will simply be wrong. In a case like this, having a strategy helps. 1 Anonymous, “Inspiring Innovation,” Harvard Business Review, August 2002, pp. 39–49. 98 Part Two Concept Generation Finding the Right People Creativity has been described by Craig Wynett, a senior manager at P&G, as “the everyday task of making nonobvious connections.” Firms like P&G that are known for their innovative product programs are also known for being staffed with highly creative people—those that get original ideas with a high degree of usefulness. One such highly creative person was Harry Coover, the discoverer of superglue (cyanoacrylate adhesives). He was working on plastics from which to cast precision gunsights. He noticed that the plastic he was working with stuck to everything, and thereby ruined a refractometer he was using to study it with. He also was the first to get the idea that superglues could be used by doctors as an adhesive for human tissues.2 Harry Coover’s example demonstrates that originality and usefulness are both important characteristics of creative ideas. Most people think reproductively—solve problems in ways that have worked for us in the past. Creative geniuses think productively, rethinking how to visualize the problem. Nobel Prize–winning physicist Richard Feynman called it “inventing new ways to think.” For example, what is half of 13? Most of us would say 6½. But by redefining the problem we can identify other solutions. • One half of “thirteen” is “thir.” • One half of “1-3” is “1.” • Cutting XIII horizontally through the middle gives VIII. Can you think of others? The key here is to keep looking, even after you have found a solution!3 Several thinking strategies seem common to creative geniuses in all walks of life (see Figure 4.1). A common stereotype is that creative persons are eccentric. While this may not always be the case, creative individuals do announce themselves by leaving a lifetime trail of creative accomplishments. They are creative as children and never become uncreative. This is the bottom line for us, since people being considered for new product team assignments can be evaluated on their past. People without a lifetime trail usually blame unfamiliar environments, overpowering bosses, limited opportunities, and the like. Creativity can be measured using the standard MBTI® (Myers-Briggs Type Indicator) Creativity Index. This index is based on the MBTI personality measurement instrument, used to assess individuals on four personality scales (intuitivesensory, perceiving-judging, extraverted-introverted, and thinking-feeling). The MBTI Creativity Index uses an individual’s personality scores to assess his or her creativity: Creative types tend to be more intuitive, perceiving, extraverted, and thinking than other individuals.4 More recent studies of new product development personnel found that those with high MBTI Creativity Index scores did more new 2 Harry W. Coover, “Discovery of Superglue Shows Power of Pursuing the Unexplained,” Research-Technology Management, September–October 2000, pp. 36–39. 3 Michael Michalko, “Thinking Like a Genius,” The Futurist, May 1998, pp. 21–25. For dozens of problems of this type, try thinks.com/brainteasers. Good luck finding creative solutions! 4 Avril Thorne and Harrison Gough, Portraits of Type: An MBTI Research Compendium (Palo Alto, CA: Consulting Psychologists Press, 1991). Chapter Four Creativity and the Product Concept 99 FIGURE 4.1 Genius Thinking Strategies 1. Geniuses find many different ways to look at a problem. Einstein, for example, and da Vinci were well known for looking at their problems from many different perspectives. 2. Geniuses make their thoughts visible. Da Vinci’s famous sketches and Galileo’s diagrams of the planets allowed them to display information visibly rather than relying strictly on mathematical analysis. 3. Geniuses produce. Thomas Edison had a quota of one invention every 10 days. Mozart was among the most prolific composers during his short life. 4. Geniuses make novel combinations. Einstein found the relationship between energy, mass, and the speed of light (the equation E 5 mc2). 5. Geniuses force relationships. They can make connections where others cannot. The scientist August Kekulé dreamed of a snake biting its tail, immediately suggesting to him that the shape of the molecule he was studying (benzene) was circular. 6. Geniuses think in opposites. This will often suggest a new point of view. Physicist Neils Bohr conceived of light as being both a wave and a particle. 7. Geniuses think metaphorically. Bell thought of a membrane moving steel, and its similarity to the construction of the ear; this led to the development of the telephone earpiece. 8. Geniuses prepare themselves for chance. Fleming was not the first to see mold forming on a culture, but was the first to investigate the mold, which eventually led to the discovery of penicillin. Source: From Michael Michalko, “Thinking Like a Genius,” The Futurist, May 1998, pp. 21–25. Originally published in The Futurist. Used with permission from the World Future Society, 7910 Woodmont Avenue, Suite 450, Bethesda, Maryland 20814 USA. Telephone: 01-656-8274; www.wfs.org. product projects and identified new product opportunities that were much more profitable than those identified by other personnel.5 This suggests that choosing the right persons and getting them involved in the new products process in the earliest phases may be just as important as the process itself. Management’s Role in Creativity Certainly, management has a role in getting the best out of its “ideas people.” Some firms, like General Electric, seem to truly embrace new ideas, treating them like corporate initiatives, organizing learning sessions, and importantly, sticking with them—rather than moving on to the “next big thing.” This stress on business innovation allows GE to gain advantage over competitors who focus solely on financial results.6 Recent work on idea generation in large organizations suggests that top managers should keep control over innovative projects, while at the same time allowing the employees to do as much of the work as possible. In short, top 5 Greg Stevens, James Burley, and Richard Divine, “Creativity—Business Discipline—Higher Profits Faster from New Product Development,” Journal of Product Innovation Management, 16(5), September 1999, pp. 455–468; and Greg Stevens and James Burley, “Piloting the Rocket of Radical Innovation,” Research-Technology Management, 46(3), March–April 2003, pp. 16–25. 6 Thomas H. Davenport and Laurence Prusak with H. James Wilson, What’s the Big Idea? Creating and Capitalizing on the Best Management Thinking (Cambridge, MA: Harvard Business School Press, 2003). 100 Part Two Concept Generation FIGURE 4.2 Obstacles to Idea Generation Groupthink: We think we are being creative, when in reality we are only coming up with ideas that our group will find acceptable. Remember that we are not trying to find the “conventional wisdom,” but truly original ideas. Targeting error: We keep going back to the same simple demographic targets (for example, the under-35 or under-50 markets). Great new product opportunities may be missed as a result. Poor customer knowledge: Despite the money spent on market research by the top firms, the reality is that little is understood about prospective customers. Lavish research spending doesn’t guarantee that it was done well. Complexity: Creative types within organizations, as well as senior management, often think that the more complex the idea, the better it is (or the smarter and more promotable they seem). Complexity, however, is a major barrier to new product adoption (see discussion in Chapter 8). Lack of empathy: These same managers are also well-educated, high-income individuals accustomed to an upscale lifestyle. They may simply not understand the “typical” customer they are trying to sell to. Too many cooks: A small new product team works fine, but large companies especially are prone to internal competition for power and influence. This is not a healthy climate for a new product in the earliest phases of development. Source: Jerry W. Thomas, “In Tough Times, ‘Hyper-Creatives’ Provide an Advantage,” Visions, 33(3), October 2009, 24–26. management must stay involved; and participants who had a hand in the design of the innovation will be more likely to adopt it.7 Creative people can benefit from training. Training programs range from introductory classes in traditional brainstorming to elaborate sessions that include games and horseplay. Obvious, but often overlooked, is training in the company’s products, its markets, competition, technologies used, and so forth. Newly born ideas are extremely fragile, quite the opposite of the strong and almost unstoppable concepts that are 80 percent of the way through the process. By then, many ideas have picked up one or more powerful owners. So, if we give these people a hard time, show no appreciation for their ideas, offer no particular encouragement, they simply let the ideas slide by, vowing to “not waste my ‘genius babies’ on those idiots.” Figure 4.2 shows the kind of roadblocks that exist within firms and keep them from generating creative new ideas: not knowing the customer, not being empathetic to customers’ needs, preferring ideas that everyone agrees with rather than the truly creative ones, and so on. John Cleese, formerly of Monty Python, is now a training consultant. He jokes, “No more mistakes and you’re through!” This sparks a sense of excitement in creative people, and there’s nothing like excitement to get the innovative juices going. Michael Dell, CEO of Dell Computers, says that it is important to keep his employees unafraid of failure, as he believes that innovation involves learning from failure.8 Managements therefore have two packages of activity, one designed to encourage the creative function, and the other to remove roadblocks that thwart it. 7 Davenport et al., 2003, p. 171. “Inspiring Innovation,” op. cit. 8 Chapter Four Creativity and the Product Concept 101 Activities to Encourage Creativity Today’s managers recognize that innovators are apt to be different and need special treatment. “Accommodative” is the word. Innovators can’t be allowed to violate rules at will, but it’s good to recognize individuality, be tolerant of some aberrations, and be supportive under stress. Also, management should allow innovators freedom to associate with others in similar positions. This freedom extends to all functional areas and to outside the firm as well—no locked cells. Management should also permit innovators to help select projects for development, though this is often difficult. Job assignments should be challenging. Creative people don’t lack confidence and, in fact, often consider their present assignments a waste of time. This means they will determine whether an assignment is worthy—no one can tell them. Some firms deliberately create competitive teams and have them race to a deadline. Bell & Howell’s management once faked the news of an impending competitive breakthrough to urge a scientific group to speed up. A more common technique is free time. Google, for example, offers their researchers 20 percent free time to work on whatever creative projects they want to. Google’s Gmail service was one innovative product that sprang from a free time project. Another was Post-It Notes; 3M has been a longtime believer in free time for its employees.9 Flextime is a similar tool, but for creative types it means letting employees take work home or stay in their workplaces and work all night if they want. Transferring creative personnel also helps. Creative people like novelty and want to change situations occasionally. Then, of course, we see a wide range of unique techniques developed by individual firms, especially those known for their creative achievements. Texas Instruments, for example, had a program called IDEA (identify, develop, expose, and action). Sixty IDEA representatives throughout TI could dole out funds (without higher approval) for projects proposed by personnel who did not have enough influence to get funds through normal channels. Speak & Spell was a notable result. 3M also awards genesis grants of up to $30,000 to fund innovative new projects that don’t fit the business structure. Polaroid’s SX-70 system began this way too: The project was actually “special experiment number 70,” developed outside the normal structure at Polaroid. Sony and Toshiba will give teams a six-month project budget to take a new product concept all the way through development and out to market on a small scale. This investment not only gives the team development resources, but also allows the firm to establish a technology standard and identify the early adopters in the market.10 The 3M Company has a long history of innovation, so it is not surprising that a chairman once said: “We do expect mistakes as a normal part of running a business, but we expect our mistakes to have originality.”11 One very creative product 9 Ray Boyer and Rishu Mandolia, “Has the Recession Changed Innovation? Varied Perspectives from Those on the Front Lines of NPD,” Visions, 34(1), 2010, pp. 6–7. 10 Karen Anne Zien and Sheldon A. Buckler, “From Experience: Dreams to Market: Crafting a Culture of Innovation,” Journal of Product Innovation Management, 14(4), 1997, pp. 274–287. 11 L. W. Lehr, “The Role of Top Management,” Research Management, November 1979, pp. 23–25. 102 Part Two Concept Generation design firm, IDEO of Palo Alto, California, takes several specific steps to create a culture of creativity and innovation. They seek individuals who love product design, set up offices in cities like Chicago, San Francisco, Boston, and Tokyo that attract creative types, and permit employees to swap positions and locations frequently. To generate ideas, IDEO designers take objects apart and may visit places like airplane junkyards or the Barbie Hall of Fame. They collect assorted parts and pieces of items in Tech Boxes and index all pieces on the company intranet. Tech Box curators conduct weekly conference calls.12 Creative firms often use a computerized database, or “idea bank,” to store and document ideas from earlier, unused new product projects for reuse later. These ideas can come from market research or test market results, project audits, design plans, engineering notes, and elsewhere. To help transfer information, managers that worked on the earlier project can be assigned to the project where the idea might be reused.13 Guinness Breweries, for one, periodically reviews its idea bank, viewing it as an important component of the concept generation phase of its new products process. Oce, a computer peripheral manufacturer, calls this database their “refrigerator of ideas.”14 In general, creative operations should be in areas conducive to exchange of ideas; office arrangements should make people comfortable; and distractions should be held to a minimum. The offices of the Internet startup factory Idealab!, and also IDEO, are laid out such that employees can hear each other’s problems and interact with each other as much as possible.15 These firms actively encourage creativity among their employees and get the desired results. Design Continuum, for example, had to design a new pulsed lavage (a product that cleans wounds with saline solution); battery-powered squirt guns were the inspiration. This firm also created a new kitchen faucet design after considering valves used in toys, cars, and medical products. Earlier, Design Continuum had been responsible for the Reebok Pump shoe, having applied ideas from inflatable splints, intravenous bags, and diagnostic valves.16 Another highly creative firm, Qualcomm, uses an innovation engine technique to generate ideas. Senior managers carefully select employees who have proven to be highly creative and innovative and form several groups of 12. Each individual receives a homework assignment, which is to prepare six ideas they think Qualcomm should be working on. The groups are sent to offsite brainstorming sessions, where they discuss and extend the ideas brought in by each individual, the goal being to identify new, big ideas (not incremental new products). Each 12 Andrew Hargadon and Robert L. Sutton, “Building an Innovation Factory,” Harvard Business Review, May–June 2000, pp. 157–166. 13 Sarah J. Marsh and Gregory N. Stock, “Building Dynamic Capabilities in New Product Development through Intertemporal Integration,” Journal of Product Innovation Management, 20(2), March 2003, pp. 136–148. 14 Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, “Optimizing the Stage-Gate System: What Best-Practices Companies Do—I,” Research-Technology Management, September–October 2002, pp. 21–27; Tekla S. Perry, “Designing a Culture for Creativity,” Research-Technology Management, March–April 1995, pp. 14–17; and Zien and Buckler, op. cit. 15 Hargadon and Sutton, op. cit. 16 Hargadon and Sutton, op. cit. Chapter Four Creativity and the Product Concept 103 member receives an allocation of “Qualcomm Bucks,” which they can invest in what they think is the best new idea. This procedure might generate about 12 good new ideas, which are then narrowed down in successive rounds to about four. Qualcomm can turn these four “Big Bang” ideas into products within about three months. One of the more prominent Big Bang ideas: Flo TV, a service by which owners of iPhones and similar devices can watch TV, for a modest monthly fee, without experiencing the poor video quality and slow download times usually encountered with this kind of service.17 Special Rewards There is no question about the value of recognizing creative achievement. But creative people are usually unimpressed by group rewards. They believe group contributions are never equal, especially if the group is company employees, for many of whom creatives have great disdain. This is unfair; large portions of successful creativity are now set in groups, and we know more now about how to make group judgments work. But creatives do like personal accolades—preferably immediately. The famous Thomas Watson of IBM commonly carried spare cash in his pockets so he could reward persons with good ideas when he heard them. Campbell Soup has Presidential Awards for Excellence. Many firms have annual dinners to recognize employees who obtained patents during the year. At IDEO, there are no organization charts or job titles: Parties and trophies, rather than job promotions, are the rewards for a job well done. In one of the most dramatic reward systems, Toyota and Honda have their champions follow the new product out the door and take over its ongoing management.18 The Removal of Roadblocks As seen in Figure 4.2, some organizations set up roadblocks, perhaps unintentionally, that stop new product concept creativity. Managers will say that the concept “simply won’t work,” or “it’s against policy,” or “we don’t do things that way.” These statements are often well intentioned, and they may be accurate statements of the status quo. But they are extremely discouraging to fragile ideas, and only conscious effort by managers can help scare them away. Some organizations use a technique called itemized response. All client trainees must practice it personally. When an idea comes up, listeners must first cite all its advantages. Then they can address the negatives, but only in a positive mode. The recommended language for bringing up a negative is “OK. Now—let’s see what would be the best way to overcome such-and-such a problem.” Note that this constructive comment assumes the problem can be overcome, and the listener offers to help. To encourage creativity, some firms deliberately encourage conflict by putting certain employees together on the same team—for example, a blue-sky 17 The Flo TV example is documented from several news articles on the Qualcomm Web site, qualcomm.com. 18 These ideas and many more are discussed in Tekla Perry, “Designing a Culture for Creativity,” op. cit. 104 Part Two Concept Generation FIGURE 4.3 Barriers to Firm Creativity 1. Cross-functional diversity. A diverse team means a wide variety of perspectives and more creative stimulation, but also can lead to difficulties in problem solving and information overload. 2. Allegiance to functional areas. The team members need to have a sense of belonging and to feel they have a stake in the team’s success. Without this, they will be loyal to their functional area, not to the team. 3. Social cohesion. Perhaps a little unexpectedly, if the interpersonal ties between team members are too strong, candid debate might be replaced by friendly agreement, resulting in less innovative ideas. 4. The role of top management. If senior management stresses continuous improvement, the team might stick with familiar product development strategies and make only incremental changes. Top management should encourage the team to be adventurous and try newer ideas. Source: Reprinted by permission of Harvard Business Review. Exhibit from “How to Kill a Team’s Creativity,” by Rajesh Sethi, Daniel C. Smith, and C. Whan Park, August 2002. Copyright © 2002 by the Harvard Business School Publishing Corporation; All rights reserved. creative person and a practical type. This technique is sometimes called creative abrasion.19 The bottom line here is that managers need to be aware of the barriers to group creativity. New product teams are, by definition, cross-functional, which means a greater variety of perspectives but also potential difficulties in reaching a solution acceptable to all. Further, if the team members share strong interpersonal ties, the creative abrasion might be lacking: Team members may simply reach friendly agreements. Figure 4.3 describes these and other barriers to overcome in stimulating group creativity. The Product Concept Given creative and exciting people, just what is it we want them to produce? What is this thing called concept? How does it differ from a new product? When does it come about? Let’s start with the end point, the successful marketing of a new product, and back up. A new product only really comes into being when it is successful—that is, when it meets the goals/objectives assigned to the project in the PIC. When launched, it is still in tentative form, because changes are quite apt to be necessary to make it successful. Therefore we say it is still a concept, an idea that is not fulfilled. Back before technical work was finished, the product was even more of a concept. To understand this and see how it relates to the ideation process, we have to look at the three inputs required by the creation process. • Form: This is the physical thing created, or in the case of a service, it is the sequence of steps by which the service will be created. Thus with a new steel alloy, form is the actual bar or rod of material. On a new mobile phone service it includes the hardware, software, people, procedures, and so on, by which calls are made and received. 19 James Krohe Jr., “Managing Creativity,” Across the Board, September 1996, pp. 16–22. Chapter Four Creativity and the Product Concept 105 • Technology: This is the source by which the form was attained. Thus for the steel alloy it included, among others, the steel and other chemicals used for the alloy, the science of metallurgy, product forming machines, cutting machines, and more. Technology is defined in product innovation as the power to do work, as you will recall from Chapter 3. In most cases there is one clear technology that is at the base of the innovation, the one that served as the technical dimension of the focus-arena. Sometimes there are two. • Need/Benefit: The product has value only as it provides some benefit to the customer that the customer sees a need or desire for. We put these together this way: Technology permits us to develop a form that provides the benefit. If any of those three is missing, there cannot be product innovation, unless one buys a product ready-made and resells it without change. Even then, there would be some change in the service dimension—where it is sold, how it is serviced, and so on. Even clone makers add value, if nothing more than price; we even hear computer buffs say something like: “XYZ makes better clones than PDQ does!” Oddly, the innovation process can start with any one of the three dimensions and can vary in what happens second (see Figure 4.4). Here are the primary ways (which we will illustrate with the Designer Decaf example later in this chapter): Customer has a NEED, which a firm finds out about. It calls on its TECHNOLOGY to produce a FORM that is then sold to the customer. A firm has a TECHNOLOGY that it matches with a given market group, and then finds out a NEED that group has, which is then met by a particular FORM of product. A firm envisions a FORM of a product, which is then created by use of a TECHNOLOGY and then given to customers to see if it has any BENEFIT. Any of the three can start the process, and in each case either of the other two can come second. Now, you may say, so what is the difference? The difference is too often that between success and failure. Putting benefit last is very risky, since it comprises a solution trying to find a problem. DuPont, for example, spent several years finding applications where Kevlar could yield a profitable benefit. Apple’s experience with the Newton Message Pad personal digital assistant shows the risk firms take when they put benefits last. Technology developed by Apple’s R&D department allowed a user to enter handwritten inputs, eliminating the keyboard. From this technology, a form was conceived: a pen-based, digitized notepad designed to capture and process ideas and data. Apple didn’t check with customers, however, to see if this form actually satisfied customer needs or addressed customer problems. Customers were apparently satisfied with the triedand-true ways of capturing ideas and data: pen and paper, sticky notes, calendars, and electronic address books. The fact that it retailed at about $800, and that handwriting recognition did not work flawlessly, did not help Newton’s case either. The Newton never sold well.20 20 Abbie Griffin, “Obtaining Customer Needs for Product Development,” in The PDMA Handbook of New Product Development, M. Rosenau, A. Griffin, G. Castellion, and N. Anscheutz, eds. (New York: John Wiley, 1996), pp. 153–166. 106 Part Two Concept Generation FIGURE 4.4 The New Product Concept Form Need–form concepts Need New product Form–technology concepts Need–technology concepts Technology Concept: "A far better way of meeting the learning needs of computer users is to utilize modem-based online systems to let them see training videos on the leading software packages." (This has a well-known need/benefit and stipulates the several technologies that will be used; but exactly how this service will function is still to be worked out.) Another way of stating this concept would be: "XYZ Corporation has a national telecommunications network in place, and also owns a chain of video rental stores. Surely there is some way we can use these capabilities to help meet the training needs of homebased computer users." Again, this offers the market need and the technologies; it still lacks method/process, which is the service product's equivalent of form. (Note how close a new product concept can come to sounding like the focus/arena of a product innovation charter.) Here are two statements that may sound like new product concepts, but clearly aren't. In each case, to be worked-out concepts, another piece of the puzzle must be added. "Let's create a new way of solving the in-home training/educational needs of personal computer users." (Need, but no form and no technology. Just a wish, like a cure for cancer.) "I think we ought to develop a line of instructional videos." (No specific market need/benefit, and no form—just a technology.) Therefore we often put benefit first. Incidentally, even technology-driven scientists actually put benefit first in most cases because they have some idea of need that is leading them in their efforts. For example, a pharmaceutical chemist seeking a new compound for lowering blood pressure knows how widespread that problem is. Given the benefit, preferences vary. Some people like to visualize what type of finished product could meet the need, and then design that form. Others Chapter Four Creativity and the Product Concept 107 like to give technical people the basic benefit(s) and let them use their available technologies without restraint on form. This book follows the latter. Granting that in practice, all versions exist, and no one would throw out a good idea just because it came up in the wrong way, the fact remains that we are speaking about management. If one wants to design the best way to go about product innovation, then, in general, the best way is to have first the benefit, then the technology, and then the finished form. Think about toilet brushes. Old-fashioned brushes do the job, but one could imagine someone (a thoughtful customer, perhaps, or a research chemist at a detergent company) having the idea that a new, improved brush that somehow makes toilet cleaning easier would be a big seller. Note that we said “idea” here and not “concept” as all we have so far is a need: a new brush that offers convenience. What does that mean? Longer handle? Disposable bristles? Probably not. How about a brush that contains detergent, which is refillable and easy for the customer to attach to the brush? Now we have something resembling a simple product concept, as we have a need (convenient brush) based on use of a technology (detergent suitable for toilet scrubbing). Apparently at least three companies hit upon that concept at about the same time a few years back. But these companies developed and launched products whose forms are very different. Reckitt Benckiser produced the Lysol Ready Brush: An aerosol can of cleanser is mounted into the brush. The can is replaced when empty; the brush is not disposable. SC Johnson’s entry is the Scrubbing Bubbles Fresh Brush: Here, a disposable pad containing Scrubbing Bubbles cleaner is attached to the end of the brush. The Clorox Toilet Wand is much like the Fresh Brush, but with a round, disposable sponge instead of a pad. Think of this as a kind of translation process: In each case, the idea was developed into a concept, but the concept was translated three different ways, resulting in three different products (that offer pretty much the same benefit to customers). Which of these will do the best in the marketplace? Probably the one that offers the greatest benefit to customers. Chapter 5 will deal with how we go to customers and find out what their problems are. In Chapter 12 we will talk about a form of product description (called a protocol) that is written out prior to undertaking technical development; the description is primarily benefits. Features (form) are put into it only if they seem absolutely essential (e.g., required by law). Let’s put this all into a simple case, and maybe the issues will become clearer. The Designer Decaf Example Many years ago, coffee was, well, coffee. One went to a favorite restaurant, diner, or food truck in the morning or at lunch and ordered an inexpensive “regular.” Typically, coffee sold in North America contained a blend of cheaper coffee beans, and that was that. With the emergence of Starbucks and competitors, the North American coffee-drinking culture changed abruptly. Fancy coffee bars, based on the Italian coffee bar model, sprang up everywhere, and Italian-style espresso soared in popularity. Espresso-based concoctions like cappuccinos and lattes, often selling for three to four times the price of restaurant coffee, became big sellers overnight. Let’s imagine we worked at a major coffee roasting company at 108 Part Two Concept Generation about this time. Imagine also three different people walked into the new product office one week, at different times, each with an idea for a new product. Each was unaware the others were coming in. One person said, “Our most recent customer satisfaction report disclosed that customers would like a decaffeinated espresso coffee that tastes identical to regular espresso and can deliver a full-flavored cappuccino. No current decafs offer this benefit.” The second person was a product manager who said, “I was thinking last week about our coffees and our competitors, and noticed they were all about the same color and thickness. I wonder if we could mass-produce a darker espresso that actually pours out thicker, something like Turkish coffee” (form). The third person was a scientist who had just returned from a technical forum and said, “I heard discussion of a new chemical extraction process that can isolate and separate chemicals from foods cheaply and effectively; maybe it could be applied to taking caffeine out of coffee” (technology). Each of these people had a germ of an idea, but as a concept each suggestion wasn’t really very useful. The first person had something on a par with a cancer cure—benefit, but no way to supply it. The product manager had no idea whether consumers would like darker, thicker espresso or how it might be made. The scientist didn’t know whether the technology would work on coffee or even whether consumers wanted a change. A new product concept would result if the first person met with either the second or the third. If the second, they would ask the lab for a technology that would produce the sought form and benefit. If the third, they would undertake lab work to find the exact form of the new technology (for example, should all or only some of the caffeine be extracted; if darker or thicker in appearance, how dark or thick). What might best sum up the point that a concept is evolving from its creation until it metamorphoses into a new product is the saying of one manager: “Don’t waste your time trying to find a great new product idea; it’s our job to take a rather ordinary idea and make it into a successful new product.” The Concept Statement Ideas, concepts, new products, and so on are all words in common use. But, as in all disciplines, we have to clarify them for understanding. Medical books draw a sharp distinction between common cold, sinusitis, upper respiratory infection, and so on, even though as patients, we don’t care. Figure 4.4 showed that any two of the three (form, benefit, technology) can come together to make a concept, a potential product. All three together produce a new product that may or may not be successful. Often, there is little difference. For example, inventors frequently call on companies with a prototype in hand. This is a concept that is virtually finished—it has form, based on a technology, and you can be sure the inventor knows a benefit it provides. Of course, firms know from experience that the inventor usually overstates the benefit; the technology will have drawbacks that make it impractical to use in a plant; and the form is very tentative based primarily on tools and space in a crude workshop. At the other extreme the very first thought about a new product may be so incomplete that nothing can be done with it as is. For example, the scientist returning Chapter Four Creativity and the Product Concept 109 from the technical forum had only capability—nothing that had value to anyone in the coffee roasting company. Once a concept appears, with two of the three dimensions (technology, form, benefit), we have to screen it before undertaking development. That part of the process comes in Chapter 9, and it requires what we call a product concept statement. Technical people and intended customers must tell us the concept is worthy of development. Their review of the concept statement allows this, if the concept tells them what they need to know to make that judgment. A concept statement will usually do this if it has two of the three basic essentials (technology, form, benefit). If you were asked, “How would you like zero-calorie ice cream?” you could not really answer. You probably already find yourself thinking, what will it taste like, what is it made of, what’s the catch? To do concept testing, we need a concept statement that meets these information needs. What if you wanted to design a fork that would help people eat more slowly, which would reduce digestive problems, minimize gastric reflux, and (importantly) help people lose weight? What we have at this point is a concept, of course, since we have the form (a fork) and a customer benefit (weight loss and other positive health outcomes). Now, if there is customer interest in this concept, the firm can add the technology component: the R&D people can develop the technical specifications and begin making prototypes. In this case, it would require developing software that keeps track of a person’s eating habits (as well as other metrics such as time spent on exercise or sleep), as well as an indicator light that flashes if the person is eating too fast. The technical side is likely to be challenging and possibly costly, so it is good to know at this early stage if the concept is not likely to be popular and time and effort is not wasted.21 A concept, then, is a verbal and/or prototype expression that tells what is going to be changed and how the customer stands to gain (and lose). Early on, the information is quite incomplete, but when marketed the concept is (hopefully) complete. Anything that doesn’t communicate gain and loss to the intended buyer is still just an idea that needs work. An interesting demonstration of the three-facet concept source came when Eddy Goldfarb, a famous toy inventor, was asked how he did it. He replied, “Notice what things your child plays with, and try to spot what’s lacking.” He also said he likes to look for new processes and materials and “for holes—you know, a lack of a certain item on the market.” These statements cite benefit, technology, and form, in that order.22 The importance of these three dimensions varies by industry. In most industries, one of the three often needs no attention because of general knowledge within the industry. Pharmaceutical new products people do not have to check out the desirability of stopping body fluid buildup, or of eliminating cancer. Furthermore, pharmaceutical expertise is available to manufacture virtually any new drug, so technology is the only unknown and thus the focus of attention. On the other hand, the leading food companies presume the kitchens and factory can put 21 This product exists and is sold successfully. It’s the Hapifork, manufactured by Hapilabs (see www.hapifork.com for details). 22 Fran Carpentier, “Can You Invent a Toy?” Parade, December 1981, pp. 14–15. 110 Part Two Concept Generation together anything the customer wants, so benefit (ascertained through taste tests, for example) becomes the prime variable. In the automobile industry, car manufacturers so dominate the new products process that components suppliers are told what benefit is wanted and then work with either technology or form for its innovation. In these three different industry situations, discussion with new products people quickly indicates the critical avenue of innovation for their firm or industry. And the distinctions are not moot—they provide the direction for the idea stimulation process. Still, it takes all three. If a project aborts, it may be the fault of the department with the easy task. For example, a television manufacturer’s marketing research may show that consumers want a television set that will increase in volume as room noise picks up and decrease in volume as room noise subsides. This research engenders the idea for the new product, so the process would be demandinduced. But, in reality, the technical side of the business has the toughest task. Two Basic Approaches Now, given some agreement on language, we can go back to the original question: How should we go about generating new product concepts? The diagram given in the figure at the start of Part II showed five routes—technology, end user, team, other insiders, and other outsiders. Two of these involve receiving product ideas created by others, and three of them involve a managed process run by the team. This distinction is the one that makes a managerial difference, and it is the one we will use in this book: ready-made versus do-it-ourselves. Here we will discuss the ready-made source, and in Chapters 5 through 7 doing it ourselves. Of course, most firms use both ready-made and tailored. But in each industry it is common knowledge as to which has a better batting average. For example, food manufacturers usually will not even read new product suggestions sent in by consumers. They have more than enough concepts of their own; consumer suggestions are very repetitive or old ideas; and even just glancing at hundreds of thousands of ideas every year would be almost impossible. Yet, in some other industries (e.g., toys and tools) inventors thrive. There are even inventors’ fairs, where inventors are invited to display their creations. Haystack Toys Inc. periodically holds a Great American Toy Hunt, inviting toy inventors from all over the United States to present their prototype products. Each inventor gets 15 minutes to present his or her product before Haystack judges. About 100 make the final judging round, and of these, Haystack will select at most 10 to develop and market. Dan Lauer, cofounder of Haystack, believes that the biggest toy companies overlook the best and most innovative toy ideas, preferring to extend well-entrenched brands (like Barbie) or to get movie licensing rights. In fact, most big toymakers will routinely turn down walk-in-the-door ideas regardless of their potential value. Says Lauer of his experiences trying to sell one of his toy ideas (novel, fun tub toys) to the big companies, “I had to start a company to make what I want.”23 23 Samuel Fromartz, “Creation Theory,” Inc., March 2000, pp. 86–103. Chapter Four Creativity and the Product Concept 111 Some manufacturers have employee and customer idea contests. In the food industry, Pillsbury runs an annual Bake-Off Contest to capture thousands of new recipes for their possible use. Alan Klingerman, head of AkPharma Inc., has a knack for thinking up products that give people relief, having developed both Lactaid (lactose-reduced milk) and Beano antiflatulence pills, the success of which he attributes partly to the humorous-sounding name. Both Lactaid and Beano were eventually sold to big drug companies. Each also resulted in at least one spinoff product: CatSip (lactose-reduced milk for cats) and CurTail (a kind of Beano for dogs). Klingerman claims to keeps file drawers full of what he calls “ideas and stuff.”24 One thing we know for sure, concept generation should be an active, not reactive process. This is no time to be thinking like the Maytag repairman, waiting for something to happen. Important Sources of Ready-Made New Product Ideas Appendix A lists and discusses the most common sources of ideas already created. Experience in the field of product innovation has it that 40 to 50 percent of new product ideas are ready-made, coming from a wide variety of sources (see Figure 4.5). Many firms are turning to formal methods to tap into ideas from customers and external stakeholders such as suppliers. User toolkits, crowdsourcing, and lead user analysis are valuable ways to source customer ideas; many companies have adopted an open innovation framework to source from a wide range of external partners. Let’s examine each of these methods in turn. User Toolkits Some firms are turning to user toolkits, a method that formally turns the innovation task over, to some extent, to the users themselves.25 A toolkit is a user-friendly set of design tools that customers can use, together with their understanding of their own needs, to customize a product that would be best suited to them. The customer-designed product can then be directly transferred to manufacturing or production. Most everyone is familiar with product configurators, which are a simple kind of user toolkit. You can “build” your own car at www.fiat500.com or your own laptop at www.dell.com, or design your own running shoes at www.nikeid. com. These configurators allow the user to mix and match different components (for a car, this might be engine size, interior and exterior colors, hubcaps, audio system, and so forth) and see what the retail price would be. Nike even allows the user to select a message to appear on the sides or tongues of the shoes (though, not surprisingly, the user is not permitted to enter “Buy Adidas”). In essence, these firms are practicing mass customization, at least to some extent, in which the user is a “segment of one.” The Fiat webpage once boasted there were more 24 Robert Zausner, “An Inside Job,” Philadelphia Inquirer Magazine, February 25, 2001, pp. 16–25. Eric von Hippel, “Toolkits for User Innovation,” in P. Belliveau, A. Griffin, and S. M. Somermeyer, The PDMA Toolbook 2 for New Product Development (New York: John Wiley, 2004). 25 112 Part Two Concept Generation FIGURE 4.5 Sources of Ready-Made New Product Concepts Actively seek out ready-made new product concepts Inside sources (employees) New products people Technical: R&D, engineering, design Outside sources End-users Lead users Marketing and manufacturing Other stakeholders: resellers, vendors, etc. All others General public: idea people, inventors Idea exploration firms, consulting engineers Secondary sources than 500,000 different combinations, almost guaranteeing that a newly ordered car would be different from any other. (The cereal franchise Cereality claims to have a “gazillion” combinations of cereals and toppings on its Web site, www.cereality .com.) Coca-Cola has hundreds of soda machines with several flavors so that customers can mix them and try out new combinations. Not only is this a form of mass customization (customers can pick unique flavor combinations), but CocaCola tracks customer selections, observes which combinations seem popular, and gets ideas for future new products.26 26 Gary R. Schirr, “User Research for Product Innovation: Qualitative Methods,” in K. B. Kahn, S. E. Kay, R. J. Slotegraaf, and S. Uban (Eds.), The PDMA Handbook of New Product Development, Hoboken, NJ: Wiley, 2013, Ch. 14, pp. 239–240. Chapter Four Creativity and the Product Concept 113 FIGURE 4.6 Two Examples of Toolkits for User Innovation* International Flavors and Fragrances (IFF) This company makes specialty flavors added into processed foods. Typically, a customer would place a requirement, such as a “meaty flavor to add to a soy product,” and IFF starts working. A sample might be shipped back to the customer within a week. The trouble is that the customer firm may not be entirely happy, but finds it difficult to define exactly what it wants (e.g., “make it ‘gutsier’”). Several iterations might need to occur between IFF and the customer before the latter is satisfied. This is especially problematic to IFF since customers typically expect they will get it right the first time. To respond to this problem, IFF created an Internet-based user toolkit that provides a huge database of flavor profiles as well as design rules used in combining or modifying these. The actual chemical formulations are not provided in the toolkit, of course, in order to protect IFF’s intellectual property. The customer firm now can design its own flavor and send it directly to a machine that will make up a sample within a few minutes. The customer can easily make adjustments, using the easy-to-understand customer interface appearing on the computer screen, until the desired flavor is obtained. 3M Telecom Enclosure Division This 3M division makes enclosures for telecom firms like Verizon, used in mounting external equipment. In the past, a telecom customer firm would give 3M the equipment about to be installed in a customized enclosure, and 3M would design the appropriate enclosure using a CAD program. The customer checks the design, and may at that point rethink the required equipment or some other part of the specification. As in the above example, numerous iterations might occur. 3M’s solution is to provide the customer with a simple-to-use version of its own CAD program. (As above, intellectual property rights are protected by providing customers with only the customer-interface parts of the program.) The customer inputs the required equipment and other specifications, and allows the program to do its work. It can make whatever adjustments are necessary until satisfied, then sends the complete design back to 3M that can then put it right into production. *Examples are taken from Eric von Hippel, “Toolkits for User Innovation,” in P. Belliveau, A. Griffin, and S.M. Somermeyer, The PDMA Toolbook 2 for New Product Development, John Wiley & Sons, Inc., 2004. User toolkits are not only for getting final consumer ideas. Figure 4.6 shows how two companies use user toolkits in a business-to-business setting: International Flavors and Fragrances provides an Internet-based user toolkit to design specialty flavors for processed foods; 3M’s Telecom Enclosure division provides its customers with a computer-aided design (CAD) program that allows them to design their own telecom enclosure, which is then reworked with the assistance of 3M until an ideal design is achieved. Do consumers like being able to design their own products and services? And are they willing to pay a modest increase in price to do so? It seems the answer is yes, so long as the toolkit is fun to use and easy to learn, as is the case with most consumer product configurators. Product researcher Nik Franke and his colleagues found that customers were willing to pay more for user-designed watches (not necessarily better-designed ones) than for comparable off-the-shelf ones, because they are more suited to the customer’s unique preferences. Another study by Franke, which involved user-designed scarves, showed that the willingness to pay was higher if customers really enjoyed using the configurator. 114 Part Two Concept Generation In addition, the more customers thought the scarf design was a success, the more they felt a sense of accomplishment.27 But these are all simple consumer products; often, customizing business-to-business products with user toolkits is more challenging to the user. Learning to use the flavor toolkits in Figure 4.6, for example, is hard work, and if proper training is not provided, that positive sense of accomplishment (and higher willingness to pay) might be replaced with negative feelings of “too stressful” or “too difficult.” We don’t have all the answers yet, but manufacturers of business products can learn from the consumer product examples shown here. Crowdsourcing Many firms have recently gone online to obtain product ideas from their customers efficiently: this kind of open idea solicitation is known as crowdsourcing.28 Dell’s Idea Storm initiative encouraged customers to submit ideas for new products and improvements to existing products online. Over 10,000 ideas were obtained from sources around the world.29 It was reported that Apple made use of crowdsourcing in generating ideas for the iPad. Apple monitored reviews and customer blogs, and also obtained voice of the customer data, to understand the most pressing needs of potential users, not just of the iPad tablet itself but also of related devices such as the iPhone.30 As a classic example of crowdsourcing, Threadless invites users to submit designs for T-shirts to its Web site, www.threadless.com, and encourages users to vote on their favorite designs, which the company then produces and sells. Since awareness of new offerings is done totally on the Web site, the online community also has a big part in the marketing of the T-shirts as well. In another crowdsourcing example, the Dutch travel information Internet company 9292 saw an opportunity to develop a new application for coordinating public transport schedules. The firm decided to leverage the application development community (many of whom were college students who were already building their own semi-legal versions of this app). This was seen as a major step for a company that had done all development in-house up to that point. In 2009, the company held a contest among college students to see who could develop the best application, on any platform, for public transport. Winners would receive a modest monetary reward but would also be considered for future work. The winning developer, a college student, received a 27 The watch example is from N. Franke and F. Piller, “Value Creation by Toolkits for User Innovation and Design: The Case of the Watch Market,” Journal of Product Innovation Management, 21(6), 2004, pp. 401–415; the scarf example is from N. Franke and M. Schreier, “Why Customers Value Self-Designed Products: The Importance of Process Effort and Enjoyment,” Journal of Product Innovation Management, 27(7), 2010, pp. 1020–1031. 28 A source on crowdsourcing is Gary P. Pisano and Roberto Verganti, “Which Kind of Collaboration Is Right for You?” Harvard Business Review, 86(12), pp. 78–86. 29 See ideastorm.com. 30 Reena Jana, “Apple iPad Product Development Approach,” The Conversation blog, Harvard Business Review, January 27, 2010. Chapter Four Creativity and the Product Concept 115 share of revenues and a part-time job at 9292 in exchange for the right to use the new app and to brand it as a 9292 product.31 Crowdsourcing also occurs in business-to-business markets. Innocentive provides technical crowdsourcing for large pharmaceutical firms, including Eli Lilly, as well as many consumer-goods firms. Innocentive offers a fee to the technical expert that can identify the best solution. Netflix tried the crowdsourcing route by holding a contest when it needed a better algorithm to select movie recommendations based on previous viewing behavior.32 We will further explore the use of online communities in idea generation in the next chapter. Note that user toolkits and crowdsourcing are most likely to generate modest product improvements rather than new-to-the-world products. Also, the typical user is less likely to come up with ideas that are easily developed into real products: Product development professionals (or more experienced users) will have a more realistic view of what is and is not feasible.33 The role of the end user also depends on the industry. For example, manufacturers of scientific instruments and plant process equipment report the majority of their successful new products came originally from customers. In other industries, such as engineering polymers and chemical additives for plastics, customers may provide less help. Lead Users Many firms seek to elicit new product ideas from their lead users, that is, the customers associated with a significant current trend (for example, fiber optics in telecommunications).34 The lead user firms (or individuals) share several characteristics: they are at the front edge of the trend, have the best understanding of the problems faced, and expect to gain significantly from solutions to those problems.35 Although usually fairly easy to identify, they may also be outlanders, or not established members of that trade. And, if they are really leaders, they may think they have already solved their problems. But in an evolving trend, their solutions may not hold up; product developers can work with them to anticipate their next problem. 31 Gert Staal, “’Crowdsourcing’ is Used by Dutch Internet Company 9292 to Create a New Travel Application,” Visions, 34(2), 2010, pp. 25–27. 32 The Threadless, Innocentive, and Netflix examples are from Gary R. Schirr, “User Research for Product Innovation: Qualitative Methods,” op. cit., p. 237. 33 Per Kristensson, Anders Gustafsson, and Trevor Archer, “Harnessing the Creative Potential among Users,” Journal of Product Innovation Management, 21(1), January 2004, pp. 4–14. For an excellent resource on the importance of establishing a dialogue with customers, see C. K. Prahalad and Venkat Ramaswamy, The Future of Competition: Co-Creating Unique Value with Customers (Cambridge, MA: Harvard Business School, 2004). 34 The best summaries regarding lead user analysis are Eric von Hippel, The Sources of Innovation (New York: Oxford University Press, 1988); and Lee Meadows, “Lead User Research and Trend Mapping,” in P. Belliveau, A. Griffin, and S. M. Somermeyer, eds., The PDMA Toolbook for New Product Development (New York: John Wiley, 2002), pp. 243–265. 35 See discussion in Nikolaus Franke, “Lead Users,” in Jagdish N. Sheth and Naresh K. Malhotra, Wiley International Encyclopedia of Marketing, Volume 5, Product Innovation and Management (West Sussex, UK: John Wiley, 2011), p. 118. 116 Part Two Concept Generation Lead users are especially helpful in giving new product ideas because their work is of the problem-find-solve type, a method stressed in the next chapter. For example, suppose your firm makes snowboards for use by extreme athletes in competitions such as the X-Games. While there are always improvements in equipment for established sports such as football or golf, there are many, many more uncertainties in designing products such as high-performance snowboards. The top athletes are even today still creating new moves and pushing the boundaries of the sport. So what should your next generation of snowboard be like? Shorter? Longer? Lighter? Heavier? Wider? More aerodynamic? More flexible? How would you know, and whom should you ask? It is those very same top athletes who would know—they are your lead users. They probably care little about the appearance of the board, as they are most concerned about improving highlevel performance. By partnering with these athletes, your firm would be able to develop radical new snowboards that address these rapidly emerging needs. What is more, these same athletes are also quicker to adopt new products than ordinary users, and are therefore also influential in speeding adoption of your new product in the marketplace.36 As a general rule on lead user selection, one study on this topic found two characteristics of lead users who are more likely to come up with commercially attractive innovations: high expected benefits and being “ahead of the trend.”37 Regardless of whether the firm turns to typical end users or lead users, one important principle is to ask customers for outcomes —that is, what it is they would like the product, or service, to do for them. Too often, product developers ask customers what product improvements they want. Note the difference: Customers said they wanted low-fat fast food, low-salt canned soup, and nicotine-free cigarettes, but didn’t buy them! One product expert, Anthony Ulwick, suggested that product developers be “informed” by customers. Ulwick notes that Kawasaki, seeking to improve its original Jet Skis, asked current Jet Ski customers what they wanted. Most suggested adding padding, or some other features that would make it more comfortable for the standing rider. No one suggested adding a seat, which of course provides the desired outcome (increased comfort). By the time Kawasaki added seats to its Jet Skis, other competitors had already done so, reducing the one-time leader to a “me-too” competitor. Ironic, since Kawasaki could have looked to the motorcycles it produced and got the seat idea!38 For another example, see Figure 4.7. After a disappointing product launch with Vista, an operating system often described as not user-friendly, Microsoft’s goal with the Windows 7 operating 36 Martin Schreier and Reinhard Prügl, “Extending Lead-User Theory: Antecedents and Consequences of Consumers’ Lead Userness,” Journal of Product Innovation Management, 25(4), 2008, pp. 331–346. 37 Nikolaus Franke, Eric von Hippel, and Martin Schreier, “Finding Commercially Attractive User Innovations: A Test of Lead-User Theory,” Journal of Product Innovation Management, 23(4), July 2006, pp. 301–315. 38 Anthony W. Ulwick, “Turn Customer Input into Innovation,” Harvard Business Review, January–February 2002, pp. 91–98. Chapter Four Creativity and the Product Concept 117 FIGURE 4.7 How Cordis Corporation Turned Customer Input on Desired Outcomes into Innovation Cordis Corporation was a Florida-based medical device maker looking to improve angioplasty balloons. They carried out a series of interviews with cardiologists, nurses, hospital administrators, and other health professionals to determine what outcomes they wanted in an improved angioplasty product, before, during, and after surgery. The customers interviewed agreed that a major outcome to focus on was “minimizing the recurrence of blockages.” The stated outcomes led to the development of the angioplasty stent, which became at the time the fastest-growing medical device in history, generating about $1 billion in revenues in its first year. Other outcomes generated suggested possible market segments: Some physicians highly valued “precision placement of the device,” while others wanted “speed in completing the procedure.” Source: Anthony W. Ulwick, “Turn Customer Input into Innovation,” Harvard Business Review, January–February 2002, pp. 91–98. system was to “make your PC simpler.” Software engineers, working with typical users as well as partner firms, sought to make the new operating system simpler and better than previous ones, offering quick access to programs, greater compatibility, better remote streaming, and several other advantages. During development, Windows 7 was extensively beta-tested with potential users (a subject we will explore in Chapter 15). Many new ideas obtained from typical users were incorporated into the finished product. To emphasize they were listening to their customers, Windows 7 advertising featured typical computer users saying “I’m a PC and Windows 7 was my idea.”39 One way to determine whether a particular industry can benefit from working directly with users to gather concepts is to ask whether customers are tinkerers. For example, dentists are; so are medical technicians and farmers. In some of these industries, the participants not only have good ideas, but have prototypes as well, and may even have undertaken a form of manufacturing by making prototypes for their friends. In fact, some research suggests that it is especially important to turn to end users with a wide diversity of perspectives and experience in order to identify highly innovative product ideas.40 Reportedly, Chrysler got the idea for building 32-ounce cup holders into their Ram pickup trucks by observing that many pickup drivers had installed their own large-size cup holders themselves. Sometimes, observing the customer identifies the problem, leaving it to the firm to find a solution. A Chrysler engineer noticed that his wife struggled putting a child’s car seat into their minivan. He came up with the idea of integrating car seats into the van’s seating system; Chrysler added this feature, which turned out to be extremely popular. Being a customer of your competitors’ products as well as your own can also provide insights on customer problems and needs. A company that makes checkout scanner systems has its 39 Source: Microsoft News Center, www.microsoft.com, Oct. 9, 2009; Windows 7 page on en.wikipedia.org; and other public sources. 40 Joseph M. Bonner and Orville C. Walker, Jr., “Selecting Influential Business-to-Business Customers in New Product Development: Relational Embeddedness and Knowledge Heterogeneity Considerations,” Journal of Product Innovation Management, 21(3), May 2004, pp. 155–169. 118 Part Two Concept Generation employees work as checkout clerks a few days a year to get a sense of the product in use and the kind of problems that can crop up. GM requires its employees to rent GM cars when on business—thus passing up an opportunity to compare its cars with the competition in a real use situation.41 Many firms now get end-user input for new product ideas by involving end users effectively, from the earliest phases of the new products process, on their new product teams (see Chapter 14). This brings their needs and problems directly onto the table. But there will always be firms where they don’t wait to be asked— they go right ahead and prototype their ideas. A new example of this today is the information technology field, especially computers and telecommunications, where end users have become quite sophisticated. Open Innovation42 One of the most exciting new developments in new product development is the adoption by many firms of an open innovation model. Open innovation has been defined as “the process a company employs to externally search for. . . research, innovation, new technologies, and products.”43 The first advocate of open innovation was Henry Chesbrough, who viewed it as a new paradigm for innovation in which the firm makes a strategic commitment to use the knowledge in the external environment to improve innovation performance. Open innovation has, in fact, been described as the dominant model of innovation for the 21st century.44 For years, firms have sought to externally acquire technologies that they lack, but on an as-needed basis. Outsourcing is common, for example, in the pharmaceutical industry, where top firms such as Eli Lilly and GlaxoSmithKline outsource a substantial amount of their new product research due to the enormous costs involved in new drug product discovery, development, regulatory approval, and launch.45 Under an open innovation policy, firms start with the understanding that much, if not most, of the knowledge they could use resides outside the firm (that is, “not all the smart people work for us”). They systematically and intentionally set out to acquire knowledge from external resources to complement their own internal resources and accelerate innovation. Accessing this innovative pool is critical, even more so as global competition heats up. The result, ultimately, is 41 The examples from this paragraph are from A. Griffin, “Obtaining Customer Needs for Product Development,” op. cit. 42 Much of this section was derived from Henry Chesbrough, Open Innovation: The New Imperative for Creating and Profiting from Technology (Boston, MA: Harvard Business School Press, 2003); Henry Chesbrough, “Why Companies Should Have Open Business Models,” Sloan Management Review, 48(2), Winter 2007; and Henry Chesbrough and Melissa M. Appleyard, “Open Innovation and Strategy,” California Management Review, 50(1), Fall 2007. 43 R. M. (Skip) Davis, “How to Make Open Innovation Work in Your Company,” Visions, January 2006, pp. 10–13. 44 Larry Huston and Nabil Sakkab, “Connect and Develop: Inside Procter & Gamble’s New Model for Innovation,” Harvard Business Review, March 2006, pp. 58–66. 45 Roger J. Calantone and Michael A. Stanko, “Drivers of Outsourced Innovation: An Exploratory Study,” Journal of Product Innovation Management, 24(3), pp. 230–241. Chapter Four Creativity and the Product Concept 119 improved joint value for all partners. And open innovation does not stop with inflows of knowledge. Inevitably, a firm will have invested in innovations that they ultimately don’t use: They may no longer fit their business model, for example. With an open innovation policy, a firm could spin off this innovation (sell outright to a willing buyer), offer it under license, form a joint venture with a partner, or otherwise profit from it.46 Open innovation does not mean that the firm outsources its R&D. Rather, the firm’s goal is to reach out beyond its familiar research partners and to access R&D carried out globally, so that it will complement the knowhow it develops internally. By partnering with an outside firm, the innovating firm leverages and supports its own R&D and product development staff. In a sense, intellectual property (IP) in open innovation is like building blocks that allow the firm to build and execute its business model. A firm can acquire IP from a partner if it supports its business model; and it can profit from an unused IP building block if another firm has a use for it. Aside from these obvious leverage advantages, the firm benefits in other ways: It has a much larger pool of innovative ideas from which to draw; it speeds up its new products process by linking with partners that have required technology; and it obtains access to its partner’s IP with lower risk. A key to making open innovation work is to select the best partner or partners. Some researchers have suggested that the innovating firm evaluate prospective partners in terms of their technological, strategic, and relational characteristics. A high level of trust between the partners is also critical to success.47 Open innovation is seen as a valuable counterpoint to traditional closed innovation models. The closed innovation model allows for inputs to come from internal sources (marketing or strategic planning inputs) as well as external ones (such as customer inputs or market information). Under open innovation, firms at the fuzzy front end of product innovation are now no longer looking externally only for inputs such as unmet needs or unsolved problems. Now, inventors, startup companies, or various sources or technology (such as independent, government, or industry labs) are all actively sought out as possible joint venture partners, or as the basis for leveraging internal product development skills. An established firm with commercialized products can also benefit from open innovation by accessing technologies that allow it to more easily move up by emerging product generation. While open innovation may have originated in high-tech industries, it is increasingly used in lower technology environments. In fact, two of the foremost proponents of open innovation are Procter & Gamble and Kimberly-Clark.48 In 2000, P&G was going through a rough period: Innovative successes were scarce 46 Michael Docherty, “Primer on ‘Open Innovation,’ Principles, and Practice,” Visions, April 2006, pp. 13–17. 47 Zeynep Emden, Roger J. Calantone, and Cornelia Dröge, “Collaborating for New Product Development: Selecting the Partner with Maximum Potential to Create Value,” Journal of Product Innovation Management, 22(4), July 2006, pp. 330–341. 48 The P&G example is drawn from www.pgconnectdevelop.com, and the Kimberly-Clark example from Patrick Clusman and Amy Achter, “How Kimberly-Clark Uses Open Innovation to Enhance NPD Success: Interview with Cheryl Perkins,” Visions, 30(4), September 2006, pp. 10–11. 120 Part Two Concept Generation FIGURE 4.8 Open Innovation in Action: Two Success Stories Clorox and Procter & Gamble may be fierce competitors in the cleaning-products arena, but are also open innovation partners elsewhere. P&G had the intellectual property for plastic technology, in particular strong plastic film, which is the technology used in two Clorox products: Glad Press’n Seal and also Glad ForceFlex plastic garbage bags. P&G also brought its global marketing expertise to the table, while Clorox contributed the Glad brand equity, its R&D knowhow in plastics and resins, and its organizational structure suited to marketing plastic film products. Due to this open innovation partnership and the key contribution of P&G to the plastics technology, Glad sales doubled within four years, and Glad has become the second billion-dollar brand at Clorox. Kraft Foods sought open innovation partners for its planned Tassimo Beverage System. While they had the food knowhow, suppliers, and distribution channel, they needed assistance in the development and manufacture of the coffee maker. They assessed different home appliance manufacturers for manufacturing and R&D capabilities and competence in the appliance product category, and also for brand value compatibility, cultural fit, and compatibility of business strategies. In particular, they sought a manufacturer that shared Kraft’s attitudes toward quality, convenience, and responsibility. Ultimately, they selected the Bosch and Siemens Home Appliance Group. Source: Jacquelin Cooper, “How Industry Leaders Find, Evaluate and Choose the Most Promising Open Innovation Opportunities,” Visions, 36(1), 2012, pp. 20–23. and stock prices were low. The incoming CEO at that time, A. G. Lafley, felt that the problem was P&G’s closed innovation model. He instituted a now-famous open innovation model known as Connect and Develop, in which no fewer than 50 percent of new initiatives had to include at least one external partner. The results soon emerged: P&G worked with a French partner involved in wound-care R&D to jointly develop Olay Regenerist, an antiwrinkle cream. Pringles Stix originated with an innovation by a Japanese partner firm. P&G licenses the Mr. Clean trademark to partner firms that make cleaning gloves, mops, and car-cleaning kits. For two detailed open innovation success stories, please see Figure 4.8. P&G’s Connect and Develop has resulted in over 1,000 active agreements between P&G and external partners. Kimberly-Clark counts universities, entrepreneurial startups, health care companies, and packaged goods manufacturers among its open innovation partners. Indeed, Connect and Develop (and similar initiatives at Shell, Kraft, and elsewhere) is a hybrid of crowdsourcing and lead user analysis: consumers, inventors and entrepreneurs are invited to go online and send new idea submissions in for consideration.49 When forming a relationship with a new partner, Kimberly-Clark begins by carefully considering the partner’s strategic fit, vision, mission, and culture to make sure they are choosing the best partner for that situation. Kimberly-Clark also manages a venture capital fund so that it has the option of taking an equity stake in its partner. Among the ideas generated though open innovation at Kimberly-Clark: a free sample of UV sensors, produced by SunHealth Solutions, placed on packages of Huggies Little Swimmers® Swimpants so that parents can monitor the child’s exposure to ultraviolet B radiation. 49 Jacquelin Cooper, “How Industry Leaders Find, Evaluate and Choose the Most Promising Open Innovation Opportunities,” Visions, 36(1), 2012, pp. 20–23. Chapter Four Creativity and the Product Concept 121 Another application of open innovation is seen in the pharmaceutical industry. Merck, for example, instituted its Merck Gene Index, in which Merck funded university research on genetic markers. These markers would prove valuable in later pharmaceutical development. Merck published these markers in the Gene Index, accessible to any interested researchers. Why would Merck put the results of its heavily funded research initiative in the public domain? The reason was strategic: Though Merck lost exclusivity to the markers, they also blocked small biotech startups from patenting the markers themselves. Had a small firm patented a marker, it would have kept Merck from being able to develop it further into new, marketable drugs. By creating an open source of inputs (the markers), Merck hoped to capture value downstream in the compounds derived from the markers. Danish toymaker LEGO implemented an open innovation system to generate ideas from customers. Company management had identified a new robotics buildingblocks system as a high-potential new product. Relying on the high brand equity and reputation for reliability and quality associated with its name, LEGO was able to identify and attract knowledgeable lead users and offer them the opportunity to play the role of co-creators of the new robotics offering for little more than the cost of supplying them with early versions of the product. LEGO used inexpensive but effective ways to keep in touch with their lead user community: a closed Web forum, Web sites, and blogs, in which participants could share and improve ideas and even purchase them. The company also invited the participants to tour the actual production facilities, which raised their excitement level and stimulated very positive word-of-mouth. The result: a solid, highly engaged online community that helped LEGO make its robotics system its most successful product ever.50 Nike co-created Nike+ with customers, with the participation of Apple in an open innovation framework. Nike+ allows the runner to monitor performance, set objectives and workout targets, and challenge other runners, using expertise gained from Apple as an open innovation partner. Some Nike+ runners also acted as lead users by learning to track their runs using Google Maps; Nike added map tracking to its Nike+ offerings in response. Nike claims an increase in runningshoe market share of 10 points worth over $500 million.51 One interesting approach to open innovation is taken by the Dutch electronics company Philips, which created a specialized facility in Singapore known as the InnoHub.52 This facility provides several realistic environments simulating an apartment, a fashion store, and a hospital ward, as well as office and workshop areas. In these environments, end users, product developers, and other partners work together to develop new ideas for breakthrough innovations. As an example, a mirror display in the fashion store triggered a couple of ideas: shoppers viewing videoclips of products at home via the Internet, then ordering online; or sending images of themselves wearing different outfits to their friends via multimedia 50 Jennifer Dominiquini, “Dispelling the Myths About Product Innovation,” at www.prophet.com (May 28, 2009). 51 Mark Deck, “Co-Creation: A Big Idea with Major Implications,” Visions, 35(2), 2011, pp. 32–35. 52 Elke den Ouden, Darren Ee, and Nicky Goh, “The Philips InnoHub–Generating Breakthrough Innovation in an Open Innovation Setting,” Visions, Vol. 32, No. 1, March 2008, pp. 20–21. 122 Part Two Concept Generation FIGURE 4.9 Advantages and Risks of Open Innovation • Importing new ideas multiplies innovation building blocks—ideas and expertise, resulting in more total sales generated from new products. • Exporting ideas raises cash (IBM gets about $2 billion per year in patent royalties), and improves employee retention, since creative types know that good ideas will be exported and not buried. • Exporting signals the true worth of an innovation. Eli Lilly offers pharmaceutical licenses, but if outsiders don’t bite it suggests the value of the new drug is perceived to be low. • Exporting clarifies core business: Boeing sticks with design and systems integration, and often finds partners for manufacturing. • Risk: the deal is not structured in a way that captures the financial value of your innovation—ask Xerox! • Proprietary secrets can be lost to a partner, even inadvertently. • Theft of technology, or poaching of top researchers, is a concern. Source: Darrell Rigby and Chris Zook (2002), “Open-Market Innovation,” Harvard Business Review, 80(10), 2002, pp. 80–89; and Mariann Jelinek, “Open Innovation,” in V. K. Narayanan and Gina C. O’Connor (eds.), Encyclopedia of Technology & Innovation Management, Chichester, UK: John Wiley, 2010, Chapter 18. messaging. In its first four years, over 4,000 people involved in innovation visited the InnoHub; visitors interact spontaneously to the concepts they see and often generate even more ideas. Finally, another form of open innovation is the completely online system. One of these is InnoCreative, which describes itself on its Web site as a Web community that exists to match scientists to the research challenges of global firms.53 One of the complicated issues a firm must manage in an open innovation policy is intellectual property protection. Without careful partner selection, the firm opens itself up to the possibility that intellectual property could be accidentally disclosed by a partner, or worse, deliberately used illegally or given to competitors. Leading product consultants suggest that it is up to the firm to do its due diligence on prospective partners early and to make sure that all of the legalities are handled correctly, including letters of intent, memoranda of understanding, and detailed contracts.54 Some general advantages and risks of open innovation are found in Figure 4.9. Firms such as P&G and Kimberly-Clark that have committed to open innovation have adjusted their new products process accordingly. In short, the new products process must be able to incorporate externally developed ideas, intellectual property, technology, and/or commercialized products. To accomplish this, changes in the new products process can be made at any or all of the phases. In the concept generation and evaluation phases, these firms actively seek inventors, new startups, entrepreneurial firms, and other possible open innovation partners and assess 53 See www.innocentive.com; see also description in Mariann Jelinek, “Open Innovation,” in V. K. Narayanan and Gina C. O’Connor (eds.), Encyclopedia of Technology & Innovation Management (Chichester, UK: John Wiley, 2010), Chapter 18. 54 Robert Cooper, “What Leading Companies Are Doing to Reinvent Their NPD Processes,” Visions, Vol. 32, No. 3, September 2008, pp. 6–10. Chapter Four Creativity and the Product Concept 123 the potential value of joint product development. During the development phase, firms may be looking for technical assistance from scientists and other individuals outside the firm, or may seek to acquire externally developed innovations or intellectual property that can push the project along. As would be expected, this is also an opportune time for the firm to find a licensee for intellectual property not currently being used. Finally, at the time of launch or commercialization, firms may look to sell or license newly commercialized products if this provides good value, or may acquire products already launched elsewhere to obtain immediate growth potential.55 Summary Chapter 4 has introduced concept generation for new products. First we noted that management has the task of preparing an organization for concept generation. This includes applying the strategic guidance of a product innovation charter, finding and training creative people, and then creating an environment for them to work in where they can be motivated to produce. Next came a look at the concept itself, what it is, what it isn’t, and how it comes into existence. The concept is built around ideas of technology, form, and benefit and is tested by whether it can communicate to an intended buyer what the proposed product is all about and whether it appears useful. After noting that there are two broad categories of approaches to getting good new concepts, we explored the one that involves looking for ready-made concepts. Many firms use this approach heavily, and all should make at least some use of it. There are legal problems here, of course, and the chapter concluded by outlining the steps to follow in handling ideas that come from end users, lead users, employees outside the new products loop, and so on. This prepares us to look at the most difficult, but by far the best method for creating new product concepts: problem-based ideation. This is the subject of Chapter 5. Applications 1. “Lots of our people try to get good new product ideas from outsiders, but they are careful to keep it legal. I wonder, though, about something I ran into on a trip to Australia last fall. I met what our company people there called a professional espionage agent. He employs a network of flight attendants to gather tidbits of information overheard in the first-class compartments of international flights and sells this information for over a million dollars a year! I wonder what suggestions I should put in a memo for employees to minimize the chances that our key new product information will be stolen by competitors.” 55 Robert G. Cooper, “Perspective: The Stage-Gate® Idea-to-Launch Process–Update, What’s New, and NexGen Systems,” Journal of Product Innovation Management, 25(3), May 2008, pp. 213–232. 124 Part Two Concept Generation 2. “In-house inventors are tough to deal with. Right now we have this PhD in physics, a really great person, bright as they come, and terribly creative. She has had no less than 11 ideas go to market since she joined the firm four years ago. But she feels we don’t reward her properly, even though she is on a good salary, shares an annual bonus with all the other persons in research, and even got a special bonus of $5,000 last year. Frankly, I think she will leave us if I don’t find some way to let her have an equity position in some of her ideas. What do you think of her argument, and how might I arrange something if I wanted to?” 3. “In these days of intensive ideation, it sure surprised me to read that a man named Reuben Ware, a retired furniture upholstery restorer in Savannah, Georgia, had to reactivate a successful business he had shut down—producing and selling a special formulation of carpet shampoo. Seems that he had invented a formulation that removed almost anything (blood, lipstick, doggie stains, whatever) from your carpets, your laundry, or even your windshield. Sold it for a while, then dropped it. People clamored for it, so Rich’s department store bankrolled him for more product. He says he chose the product’s name, Aunt Grace’s, because he was paying the trademark attorney by the hour, so he accepted the first name that got through. When people ask him about his not being a chemist, he answers, ‘Was Edison an electrician?’ Seriously, how in the world, in these days of expensive R&D laboratories, could someone out there come up with a formulation that seems to be better than anything industry can make? And after first marketing his formulation in 1965, how was he able to keep his lead?” Case: Pillsbury Grands! Biscuit Sandwiches56 General Mills is one of the largest and most recognizable food companies worldwide. Global net sales in 2012 totaled $16.7 billion. The U.S. retail business segment is comprised of seven divisions, responsible for familiar brands such as Cheerios, Yoplait, Betty Crocker, and Pillsbury. In addition, the company has an international business segment as well as one responsible for bakeries and food service. Jeff Bellairs, director of the Worldwide Innovation Network at General Mills, says, “When you look at open innovation in particular, I believe that one personality type becomes critically important to the innovation process: that of a connector or a ‘connected innovator.’” He defines a connected innovator as a person who builds relationships and finds solutions easily, whether from the person in the next cubicle, a senior person within the firm, or an external partner who has needed expertise. The connected innovator also is skillful at taking an initial challenge or vision and directing it toward a real outcome. In recent years, General Mills new 56 This case is adapted from Jeff Bellairs, “DNA of a Connected Innovator,” Visions, 37(1), 2013, pp. 6–9. Chapter Four Creativity and the Product Concept 125 product teams have developed many new food products in combination with open innovation partners, including Yoplait Smoothies, Fiber One 90-Calorie Brownies, and Pillsbury Grands! Biscuit Sandwiches. General Mills seeks new open innovation partners through its G-WIN open innovation platform, and at the same time also reviews past relationships to find a potential partner having the requisite skills for a new project. The Grands! Biscuit Sandwiches product development and launch can be used as an illustrative example of open innovation in action at General Mills. Grands! was a concept for a new hot breakfast biscuit that would be quick and easy to prepare, and the new product team in charge began working in earnest in June 2010. While it was possible for the Pillsbury division of General Mills to develop the product entirely in-house, there were certain considerations that made it particularly suited to an open innovation partnership. First, at that time, General Mills baking plants did not have enough extra baking capacity to add a new product without major disruption. Second, the biscuit toppings were to be an important part of the product, and General Mills realized that developing this expertise inhouse would be time-consuming and costly. Further, since this would be General Mills’ first frozen breakfast food product, management saw it as a potentially risky undertaking if done internally. It was decided to search for a partner with the requisite expertise in baking and, in particular, in getting the topping up to speed quickly. Better Baked Foods, a Pennsylvania-based baking company, was a General Mills partner on a couple of earlier projects, notably a new pizza to be launched under the Totino’s name. While the Totino’s project never was launched, General Mills maintained the relationship with Better Baked Foods and agreed that they would look for joint development opportunities in the future. Research scientist Peeyush Maheshwari was working for the Totino’s brand at the time, but since had moved to the Pillsbury division. Because of his prior experience in partnering with Better Baked Foods, he thought immediately of them as a suitable partner for the Pillsbury new product team working on the Grands! project. Better Baked Foods was approached by the Grands! new product team in September 2010 and was able to get started on the project within 10 days. Better Baked Foods had technical capabilities that perfectly complemented the internal expertise possessed by General Mills, and ultimately the new biscuit sandwiches were brought to market successfully on an aggressive time deadline and without overspending on new product development. What was the role of the connected innovator in the Grands! Biscuit Sandwiches case? What, in general, do you think the role of the connected innovator should be in open innovation? How would this role change in a different industry, for example B2B products, medical services, or automobiles? Jeff Bellairs says that one of the most important personal traits of a connected innovator is perseverance. In pursuing a feasible solution for a tough product problem, the connected innovator will inevitably hear “we’ve tried that before” several times, but he or she must not give up. What other personal traits do you think would be important for connected innovators, and of these, which are the two or three most important? 126 Part Two Concept Generation Case: P&G CarpetFlick57 In 1999, Procter & Gamble launched the Swiffer, a floor sweeper with a disposable cloth that fits over a rectangular head. The cloth actually trapped dust and thus was a big improvement over regular sweepers and brooms. But one thing the Swiffer could not do was clean a carpeted floor. Since about three-quarters of the floors in U.S. homes are carpeted, this posed a unique challenge and an opportunity for P&G. In late 2003, the company decided to do something about it and set a target launch date of mid-2005. P&G had a long-standing relationship with IDEO, the Palo Alto, California, design consulting firm. Over the years, IDEO had worked on several “one-off” projects with P&G, such as on the redesign of a toothpaste tube. Recently, the two firms moved closer together with the intention of collaborating on innovative products. By 2003, the collaboration was already showing a promising track record: Pringles Prints (potato chips with trivia questions written on them) and Mr. Clean Magic Reach (a bathroom-cleaning wand with a removable scouring pad). While successful in their own right, these products didn’t really break into any new markets. P&G was turning to IDEO for help in the design of a new Swiffer product for use on carpets, which would be the most ambitious joint project so far, but one that potentially opens up a whole new market for P&G. The P&G chemist assigned to the project was Bob Godfroid, and his first day at IDEO headquarters was an interesting one, to say the least. IDEO had started by visiting homeowners, asking them questions about existing sweepers and taking pictures. They discovered that there was a real need for an effective sweeper that didn’t make a lot of noise and could pick up just about anything. Then, in November 2003, IDEO went into “deep dive” mode. About 15 IDEO designers went to the local hardware store and bought all sorts of random items that might be even remotely handy in carpet cleaning. Then they placed several carpets all over their facility and got them as filthy as possible. When Bob walked into the session, he noticed one designer sucking up dirt with a suction gun, while several others were busy trying to pick up dirt by rolling balls of Play-Doh around on it. He observed that the room “looked like a bomb went off. . . . I don’t know if we’re going to come out of here with anything other than a bunch of pictures of a trashed room.” On the morning of the second day, Bob took a squeegee blade and scraped a dirty carpet with it. To his surprise, dirt and paper confetti particles popped up in the air, as if they were Tiddly-Winks. He angled the blade differently, and the pieces popped up higher. Someone else thought of suspending a balloon overhead, which trapped the pieces of paper with static electricity. Soon enough, Bob and IDEO’s Mike Strasser had built a prototype—a plain box, really, with a slit in the bottom to mimic the action of the squeegee blade—which they immediately named the Shagilator. By the end of the two-day period, IDEO had several crude 57 Information for this case was obtained from Sarah Lacy, “How P&G Conquered Carpet,” BusinessWeek Online, businessweek.com, September 23, 2005; Beth Belton (ed.), “Lafley on P&G’s Gadget ‘Evolution,’” BusinessWeek Online, businessweek.com, January 28, 2005; and other public sources. Chapter Four Creativity and the Product Concept 127 but working prototypes, which used either suction, glue, or scraping to get the dirt up, and the Shagilator was judged to be the best. Bob returned to P&G’s Ohio head offices and demonstrated the Shagilator to Gilbert Cloyd, P&G’s chief technology officer, by spilling crushed Froot Loops on the carpet and successfully sweeping them up. Impressed, Cloyd wrote out a check for several hundred thousand dollars in seed money to keep the project rolling. IDEO’s “deep dive” model resembled a kind of lickety-stick approach (see discussion in Chapter 3), except it was company designers and engineers, not customers, who were trying out and evaluating the crude prototypes. IDEO employees were trying out all kinds of variations on the box-and-slit idea. Thinking P&G all the way, one even made a Shagilator out of a Pringle’s can, crushed and spilled the Pringles, and made the can “eat” them back up! By early 2004, a more refined version of the Shagilator had been designed. IDEO tried another novel touch, a disposable strip that ejected out of the box, but decided against it for manufacturing cost reasons. By late September 2004, the design was fixed, and the prototype was being beta tested in 350 homes. The beta test households noticed one drawback—the sweeper couldn’t pick up hair or lint. P&G was planning an August 2005 launch, but were reluctant to go without addressing the hair and lint problem. IDEO staffers took another “deep dive,” buying lint rollers, Brillo pads, glue, and anything else they could find. The P&G lab engineers did the same. The P&G engineers thought adhesive paper was the way to go, but it kept sticking to the carpet. Finally, one of them tried gluing a chopstick down the center of the adhesive paper (to keep the latter from touching the carpet). This solution worked: The paper was high enough not to stick to the carpet, but was low enough to trap hair and lint. A few additional tweaks were made. The sweeper’s color was changed from “Swiffer green” to a new bright orange, to emphasize that this product was to clean a whole new kind of surface. A more appropriate name, CarpetFlick, was chosen. It was shipped, first to Europe, then on to other parts of the world, by the end of July 2005, exactly on time. What was IDEO’s contribution in the development of the CarpetFlick? What was unusual about it, and in what unusual ways did P&G gain from this contribution? How else might P&G have generated a concept or concepts that would address this market opportunity? Suppose you are called in as a creativity consultant to assist in further development of this product. How could new product concepts that would further satisfy P&G’s wishes be generated? Case: Aquafresh White Trays58 This case details how GlaxoSmithKline (GSK) entered into an open innovation relationship with a small manufacturing company, Oratech LLC, to get into the teeth whitening market with Aquafresh White Trays. In one sense, the partnership 58 This case is drawn from Scot Andersen, Kevin Foley, and Lee Shorter, “A Story of What Happens When Opposites Attract—Hint: It’s Something to Smile About,” Visions, 31(4), December 2007, pp. 16–17. 128 Part Two Concept Generation seems to be a perfect match, a textbook example of open innovation in action. GSK noted the rapid growth in teeth whitening products and was experienced in product marketing, sales, and distribution. Also aware of the potential in this category, Oratech had already developed the product, owned the patents, and could handle manufacturing. But the road was not as smooth as expected. The first successful teeth whitening product was P&G’s Crest Whitestrips, launched in 2001, followed by several competitors. While the market showed great interest in these products, there were frequent customer complaints. Most notably, customers found the first few products difficult to use, foul-tasting, and messy. This suggested to GSK a market opportunity based on improved customer value. By offering better teeth whitening properties, and at the same time delivering a product that was tastier and easier to use, GSK could capture a share of this market. GSK was already in the oral care business with its Aquafresh toothpaste line; GSK management felt that only a product that offered superior customer value would be worthy of carrying the Aquafresh name. They also recognized that the most efficient way to enter this market was with a partner that had the required technology knowhow. Oratech, a small private-label manufacturer, was already making a teeth whitening tray and selling it to dentists and other professionals. They too recognized the growth potential in the consumer market, but needed a partner who owned the requisite marketing skills and brand equity. Oratech identified a small number of potential partners and soon chose GSK due to their marketing and R&D capabilities, in addition to competency in working with regulatory agencies. Some problems arose in the early going, due to differences in corporate culture. Oratech was initially surprised by the complexity of development and regulatory standards, which were second nature to a huge global corporation like GSK. Perhaps more unexpectedly, the new products processes employed by the two firms were somewhat different as well, with Oratech’s version of the process being a little simpler and somewhat more streamlined, typical of a smaller manufacturing firm. The partnership went well, taking only about 18 months to get the new product ready for launch. GSK ran into some development challenges, which required that they make some decisions typical of late-phase product development: Do we sacrifice quality, or slow down time-to-market? Despite this slight setback, there was really never any doubt at GSK: The commitment to bring the best-quality product to the consumer under the Aquafresh name was the number-one priority. Oratech managers were quite impressed with how seriously this commitment was taken by GSK, who brought in consultants to try to fix the development problems and not slow down development time too much. Scot Andersen, VP of marketing and sales at Oratech, said that the “level of sophistication with which GSK treats its own brands resulted in an improvement in our own processes.” Aquafresh White Trays were launched in early 2007, beating all sales forecasts and going on to be a top player in the teeth whitener category. Executives from both companies agreed that a key factor leading to this success was open communication throughout the new products process. As it turned out, if one partner ran into a manufacturing problem, the other was able to find a solution. A good Chapter Four Creativity and the Product Concept 129 example of this was the manufacturing process for the trays themselves. GSK preferred individual molding of the trays, but knew that this would run up production time and cost; the alternative was to vacuum-form and cut them, which led to imperfections at the edges. With its technical and manufacturing expertise, Oratech figured out a way to trim the edges, resulting in a desirable finished product. In turn, Oratech was very surprised to see how accessible GSK employees (and even senior management) were throughout the new products process; they were not expecting such a close relationship given GSK’s size. What accounts for the market success of the Aquafresh product? Keep in mind: As large and knowledgeable as GSK is, both P&G and Colgate already had similar products on the market, and both could easily defend themselves against the competitive launch of Aquafresh. More generally, what can be learned from GSK’s perspective, and also from Oratech’s perspective, about making open innovation work? C H A P T E R F I V E Finding and Solving Customers’ Problems Setting Chapter 5 will be devoted to the most productive concept-generating system that we know—the problem-based approach of finding and solving customers’ problems. It seems obvious and easy: Ask customers what their problems are and have a scientist put together the solution! But it’s not always so simple. Just getting customers involved is often difficult. Learning their toughest problems is more difficult, partly because they often don’t know their problems very well. Many departments of a firm may be involved, not just the technical ones. You might want to glance back at Figure II.1 in the introduction to Part II, which briefly depicts the problem-based approach to generating concepts, and see how problem-based ideation fits in with other methods for gathering new product concepts. But ask product managers, and you’ll find that they are passionate about identifying customer problems and figuring out how to best solve them—for them, this is fun and exciting work! Think about toy companies. The most innovative ones recognize that one cannot just ask young children what problems they experience with existing toys. But watch them playing in a room with a variety of toys and observe what appears to be missing to them and what they do about it (for example, using the box a toy car came in as a garage), and you may be on to something! The Overall System of Internal Concept Generation Every ideation situation is different and varies by the urgency, the skills of the firm and its customers, the product, the resources available, and so on. But one general approach, that of problem-based ideation, works best and can be modified to fit virtually every situation. The steps are diagrammed in Figure 5.1. The flow essentially is from the study of the situation, to use of various techniques of problem identification, to screening of the resulting problems, and to development of concept statements that will then go into the evaluation phase. The whole system is based on close involvement with parties who have information to help us, primarily stakeholders, which include end users, of course, but also advisors, financiers, consultants, maybe architects, physicians, or other professional Chapter Five Finding and Solving Customers’ Problems 131 FIGURE 5.1 ProblemBased Concept Generation Determine category of interest (PIC) and make thorough analysis of that situation—company, customers, resellers, etc. Team members gather needs and problems of stakeholders Sources of stakeholder needs: Search of internal records, direct input from technical and marketing, problem analysis, scenario analysis Gather information from stakeholder contacts through interviews, focus groups, role playing, observation Pool of problems Screen problem pool to acceptable problem set Undertake problem-solving efforts (by new product team members, and/or through group creativity techniques such as brainstorming or disciplines panels) Choose acceptable solution(s) and prepare concept statements groups, possibly resellers—even current nonusers certainly have information that may be useful to us! Recall from Chapter 2 that the leading cause of new product failure is the absence of a perceived need by the intended end user. If our development process begins with a problem/need the end user has and agrees is important, then we have answered the toughest question. Fortunately, organizations today are getting close to their stakeholders. But stakeholder integration is especially tough on high-security new product matters. So we figure out how to do it, just as customer satisfaction managers have. Gathering the Problems Figure 5.1 showed four sources for needs and problems of stakeholders: internal records, direct inputs from technical and marketing departments, problem analysis, and scenario analysis. Let’s explore each of these. 132 Part Two Concept Generation Internal Records The most common source of needs and problems comes from an organization’s routine contacts with customers and others in the marketplace. Daily or weekly sales call reports, findings from customer or technical service departments, and tips from resellers are examples. Sales files are peppered with customer (and reseller) suggestions and criticisms. Warranty files will show where problems are. In addition to these routine contacts, a firm may conduct formal marketing research to gather information on customer satisfaction. Studies of this type are useful, as are the files of the groups working on total quality management. Industrial and household consumers sometimes misunderstand products and erroneously project into their use of products what they are seeking. A complaint file thus becomes a psychological projective technique. One approach to handling user complaints is the toll-free complaints number or complaints Web site. It helps defuse criticism and can lead to new products. Engineers or other employees may be collocated (sent to work at customer sites) to observe customer problems firsthand. Information gained through routine market contacts can be profitably combined with other methods, such as the problem-solving technique or customer surveys. A consumer study commissioned by the SC Johnson Company in 2006 found that about one-third of homeowners cleaned the shower only once a month or less, and that a common reason was that they thought this job was difficult and took a long time to do. Over half of the respondents said that they waited until there was visible scum or dirt on the shower before they attempted to clean it! A couple of years later, another survey commissioned by the Soap and Detergent Association found that having a “sparkling shower” was one of the most satisfying cleanup jobs in the house. Since most people in the same survey said they would not employ a housekeeper or cleaning service, this job would have to be done by the homeowner him- or herself. Putting the results of the internal consumer study and the industry association study together, management identified a potentially huge unmet need: a shower cleaner that made the job easy. The result was the Scrubbing Bubbles Automatic Shower Cleaner: attached to the shower head; it sprays cleaning solution throughout the tub/shower area effortlessly by pushbutton. The product was a natural extension of SC Johnson’s popular Scrubbing Bubbles bathroom cleaners and sprays and became quite popular.1 Direct Inputs from Technical and Marketing Departments Understanding about end users and other stakeholders also lies in the minds of marketing and technical people.2 Most of them have spent time with customers and end users, sometimes many years of it. Team representatives from these two functions should canvass their colleagues, seeking out every piece of evidence on 1 From an SC Johnson press release dated March 16, 2006. A good reference on using firm employees as sources of new product ideas is Christine Gorski and Eric J. Heinekamp, “Capturing Employee Ideas for New Products,” in P. Belliveau, A. Griffin, and S. Somermeyer (eds.), The PDMA Toolbook for New Product Development (New York: John Wiley, 2002), pp. 219–241. 2 Chapter Five Finding and Solving Customers’ Problems 133 problems. They have to take the initiative on this because most of these people are busy; it’s strictly “you call me.” It’s good to remember that technical people may be found anywhere in the business, not just in R&D or engineering—especially in manufacturing, technical service, and regulatory affairs. Salespeople may not be considered in marketing, and thus are sometimes overlooked. The only real problems with using in-house people to report on customer problems are (1) each suggestion is usually someone’s perception of what the customer problem is, and (2) there is usually a solution given with each suggestion. In fact, sometimes we have to ask what new product customers are asking for and then ask why; the why is what we want to know at this time. These problems, including the time and difficulty of actually gathering memories, lead us to depend more on active searching for stakeholder problems. That is, making direct contact with all relevant stakeholders, asking them what their problems and needs are. And, although all of the above market contacts and searches around the firm help us compile useful problems, the methods of direct user contact are what we usually mean when we say problem analysis. Problem Analysis It seems that every history of an industry, a business firm, or a famous businessperson cites some key time when a new good or service capitalized on a problem that others didn’t sense or appreciate. But problem analysis is much more than a simple compilation of user problems. Although the term problem inventory is sometimes used to describe this category of techniques, taking the inventory is only the beginning—analysis is the key. As an advertising agency executive once said: If you ask people what they want in a new house and also ask them what are their problems with their current house, you will get distinctly different subject matter on each list. If you then observe their subsequent behavior, it becomes clear their problem list is a far better predictor than the want list. Users verbalize their wants in terms of current products, whereas problems are not product specific. Thus, if you ask what a person needs or wants from a shampoo, the answers will be clean hair, manageable hair, and so on—replies reflecting recent promotions of product benefits. But if you ask, “What problems do you have with your hair?” the answers may range into areas (for example, style or color) unrelated to shampoo. See Figure 5.2 for an example of what we are looking for in problem analysis, as applied to smartphones. Several recent award-winning product designs have resulted from the application of problem analysis. In one case, homeowners reported several problems with smoke and carbon monoxide detectors: ugly designs, too hard to shut off (without climbing up on a chair), nuisance alarms, poor instructions on what to do in case of an emergency. Coleman developed its line of Safe Keep Monitors to be aesthetically pleasing in appearance and added a broom button for easy reach. The carbon monoxide monitor comes with a door that opens to reveal instructions when activated (thus eliminating the need to hunt for a manual during an emergency). By being designed to solve real customer problems, the Safe Keep line has 134 Part Two Concept Generation FIGURE 5.2 Here are several smartphone problems that came up in a consumer study. See if you can generalize to a smaller number of problems. Then select the one big problem that sounds most productive for smartphone new products people. Problem Analysis Applied to the Smartphone Keeping the phone clean. Breaks when I drop it. Battery doesn’t stay charged long enough. Hard to surf the Web. Printing on Web pages is too small. Keyboard “buttons” too small so it’s hard to text. Internet connection is slow. Hard to download apps. Hard to find apps on the screen. Finding it in the dark. Not enough choice of colors or styles. Battery dies when I am in the middle of a conversation. Who “out there” can hear me? Dropped calls (line goes dead for no reason). Difficulty in looking up numbers. Other party’s voice fades in and out. I’ve heard about health risks—are they true? Can’t cradle it between ear and shoulder. My arm and ear get tired. Ringing is usually too loud, but sometimes I can’t hear it. It is a very disruptive instrument. I can’t see facial or body language. Getting flustered making emergency calls. People who call the wrong number in the middle of the night. The call doesn’t go through. Fear of what the ringing might be for. Avoiding “If you want sales, push 1,” etc. Knowing when is the best time to call people. done well in terms of sales.3 In a business-to-business application, Cemex (a large Mexican cement company) conducted customer research and discovered a previously hidden problem: Customers were unhappy with late supply arrivals at the project site. Cemex seized the opportunity and repositioned itself as the on-time supplier—a virtual “Domino’s Pizza” of the cement industry!4 Problem analysis was, at least informally, used by James Dyson in the development of the Dual Cyclone bagless vacuum cleaner (which you saw at the beginning of Chapter 1). Existing vacuum cleaners were unsatisfactory in terms of performance, maneuverability, and ease of disposing of dirt, and Dyson set out 3 Examples are from Bruce Nussbaum and contributing writers, “Winners: The Best Product Designs of the Year,” BusinessWeek, June 2, 1997, pp. 94–111. 4 Erika B. Seamon, “Achieving Growth through an Innovative Culture,” in P. Belliveau, A. Griffin, and S. M. Somermeyer (eds.), The PDMA Toolbook 2 for New Product Development (New York: John Wiley, 2004). Chapter Five Finding and Solving Customers’ Problems 135 to create a better vacuum. In later years, Dyson produced a powerful hand dryer, the Airblade, sold to the business market, and by 2009 adapted the technology behind the Airblade to create a better fan. Like vacuums, regular household fans have remained quite unchanged in design for decades. Dyson’s innovation was guided by a quick but thorough problem analysis that identified several points of improvement. As Dyson said, conventional fans have “spinning blades [that] chop up airflow, causing annoying buffeting. They’re hard to clean, and children always want to poke their fingers through the grille.”5 One could add a few more problems: Fans can tip over and are not very energy efficient. The Air Multiplier, as it was called, was purported to address many of the problems. It was bladeless, increasing safety and ease of cleaning, as well as creating a smooth stream of air. The Airblade technology provided for effective and efficient cooling, and its low center of gravity prevented dangerous tipping. It featured functional and “cool” design elements associated with the other Dyson products. The product was successfully developed and launched in late 2009, at a price point significantly above conventional fans (about $300), but within reach of customers who appreciate good design and substantially improved performance. Note that in this and the earlier examples, it is up to senior management to encourage new product teams to look beyond their normal boundaries when they explore customer problems. Problem Analysis Procedure There are several variations in problem analysis. But one commonly used procedure is reverse brainstorming. In this procedure, participants generate a list of key problems with the product currently in use, then group and prioritize these such that product development can focus on addressing the most important problems.6 The general approach is the following: Step One Determine the appropriate product or activity category for exploration. This has already been done if the product innovation charter has a use, user, or product category dimension in the focus statement. Step Two Identify a group of heavy product users or activity participants within that category. Heavy users are apt to have a better understanding of the problems, and they represent the bulk of the sales potential in most markets. A variation is to study non users to see if a solvable problem is keeping them out of the market. Step Three Gather from these heavy users or participants a set of problems associated with the category. Study the entire system of product use or activity. This is the inventory phase mentioned earlier, but far more is involved than just asking respondents to list their problems. A good method of doing this is asking 5 Rebecca Smithers, “Latest for the Dyson Touch: The Fan Without Blades,” The Guardian, October 13, 2009. 6 Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, “Optimizing the Stage-Gate Process: What Best-Practice Companies Do—I,” Research-Technology Management, September–October 2002, pp. 21–27. 136 Part Two FIGURE 5.3 The Bothersomeness Technique of Scoring Problems Concept Generation The following is an abbreviated list of pet owners’ problems found by manufacturers of pet products. Need constant feeding Get fleas Shed hairs Make noise Have unwanted babies A Problem Occurs Frequently B Problem Is Bothersome C A3B 98% 78% 70% 66% 44% 21% 53% 46% 25% 48% .21 .41 .32 .17 .21 Source: From Burton H. Marcus and Edward M. Tauber, Market Analysis and Decision Making, Little, Brown, 1979, p. 225. Reprinted with permission. respondents to rate (1) the benefits they want from a set of products and (2) the benefits they are getting. The differences indicate problems. Complaints are common and often taken as requests for new products. But they are apt to be just the result of omniscient proximity, meaning that users face a minor problem frequently, so it is the first one mentioned. Some firms have had success observing consumers or business firms actually using products in a given category; for example, observing skiers as they shoot down a hill or office workers handling a mailing operation. Step Four Sort and rank the problems according to their severity or importance. Various methods can be used for this, but a common one is shown in Figure 5.3. It uses (1) the extent of the problem, and (2) the frequency of its occurrence. This bothersomeness index is then adjusted by users’ awareness of currently available solutions to the problem. This step identifies problems that are important to the user and for which the user sees no current solutions. Methodologies to Use The generalized structure of problem analysis still contains the question of how to gather the list of customer problems. Many methods have been used, but the task is difficult. The customer or user often does not perceive problems well enough to verbalize them. And, if the problems are known, the user may not agree to verbalize them (for many reasons, including being embarrassed). Much of the sophistication in newer technologies was developed specifically to deal with these problems and will be discussed in Chapter 6. Experts We have already mentioned going to the experts—using them as surrogates for end users based on their experience in the category under study. Such experts can be found in the sales force, among retail and wholesale distribution personnel, and in professionals who support an industry—architects, doctors, accountants, and the staffs of government bureaus and trade associations. Zoo experts first publicized the problem of elephant keepers being killed when trying to cut the big animals’ toenails. Today an Elephant Hugger grabs an elephant, rolls it over on its side and holds it there, while the keeper cuts away. Later, the inventor turned his attention to a giraffe-restraining device.7 In another example, Nokia of Chapter Five Finding and Solving Customers’ Problems 137 Finland has an R&D team of 8,000 scientists and managers who collect environmental information on wireless communications and identify the main challenges. This has helped Nokia sustain growth in this market through the introduction of innovative, successful new products.8 Published Sources Also as mentioned earlier, published sources are frequently useful—industry studies, the firm’s own past studies on allied subjects, government reports, investigations by social critics, scientific studies in universities, and many others. Stakeholder Contacts The third, and most productive, is to seek out the voice of the customer (VOC)—that is, we will ask household or business/industry customers directly, via interviewing, focus groups, direct observation, or role playing. • Interviewing The most common method by far is direct, one-on-one interviewing. Sometimes this is a full-scale, very formal, and scientific survey. Other times the discussion is with lead users, an idea-generating method discussed in Chapter 4; lead users often are the first to sense a problem, and some go on to respond to it themselves. Still other times, it may be no more than conversations with some key customer friends at a trade show, because a problem statement may come from only one person and yet be very significant for us. Phone interviews have been shown to be a quick and effective way to get useful new product ideas and help to ensure that the targeted respondent (for example, a professional or a senior manager at a customer firm) actually responds, rather than a last-minute fill-in.9 Because many end users don’t think that much about the products they use and often just accept them as parts of living, even very informal discussions with individuals at a trade show or over the telephone can reopen thinking, bringing to mind things forgotten. • Focus Groups The focus group is designed to yield the exploratory and depth-probing type of discussion required, and it can be easy and inexpensive to set up and use. If done wrong, it only appears that way. Granted, in this case we are not seeking facts or conclusions, just genuine problems, and the focus group method works well by stimulating people to speak out about things they are reluctant to mention when in one-on-one interview situations. It’s much easier to talk about one’s problems when others in the group have already admitted they have problems, too. But, even in a single focus group, the costs are deceptive. Such sessions can cost from $3,000 to $10,000 in normal usage. Even at $3,000, a two-hour meeting of 10 people will yield about 10 minutes talk per participant. Since the cost is $300 7 Laura E. Keeton, “Marketers Debate the Best Way to Trim an Elephant’s Toenails,” The Wall Street Journal, February 25, 1995, p. B1. 8 Muammer Ozer, “A Survey of New Product Evaluation Models,” Journal of Product Innovation Management, 16(1), January 1999, pp. 77–94. 9 For more on telephone interviews and qualitative interviewing in general, see George Castellion, “Telephoning Your Way to Compelling Value Propositions,” in P. Belliveau, A. Griffin, and S. Somermeyer (eds.), The PDMA Toolbook for New Product Development (New York: John Wiley, 2002), pp. 63–86. 138 Part Two Concept Generation per participant, that’s talk at the rate of $30 per minute, or $1,800 an hour! It had better be very good indeed. Although the focus group technique is common, the outcome is not always, or even usually, successful. The focus group is a qualitative research technique. Unlike the traditional survey, it depends on in-depth discussions rather than the power of numbers. A problem analysis focus group should be asked: What is the real problem here—that is, what if the product category did not exist? What are the current attitudes and behaviors of the focus group members toward the product category? What product attributes and benefits do the members of the focus group want? What are their dissatisfactions, problems, and unfilled needs? What changes occurring in their lifestyles are relevant to the product category?10 In a typical example, Nissan conducted focus groups of American children between the ages of 8 and 15 to get ideas for storage, cup holders, and other features as part of the design of its full-size minivan.11 Other suggestions for helping guarantee the usefulness of focus group findings are to invite scientists and top executives to the sessions and to avoid what some people call prayer groups: Managers sit behind the mirror and pray for the comments wanted rather than really listening to what users are saying. Be sure the focus groups are large enough for the interactions and synergy that make them successful, and don’t expect focus group members to like your products! Focus group moderators know not to begin the session cold, but instead to let people get comfortable and introduce themselves—a rule of thumb is to treat participants as one would treat strangers at a party. The best moderators genuinely like people and generate openness and trust by asking ice-breaker questions and by contributing personal experiences and practices.12 • Observation Observation methods are rooted in sociological studies, and involve watching customers (or noncustomers) using products in their own environments. Video cameras or photos are sometimes used to record observational data. The new product team observes the data carefully for actions, body language, and so on and tries to identify customer needs and wants, and new product ideas that might satisfy these needs.13 10 “When Using Qualitative Research to Generate New Product Ideas, Ask These Five Questions,” Marketing News, May 14, 1982, p. 15. 11 Norihiko Shirouzu, “Tailoring World’s Cars to U.S. Tastes,” The Wall Street Journal, January 15, 2001, pp. B1, B6. 12 Joseph Rydholm, “Respondent Collages Help Agency Develop Ads for New Pontiac,” Quick’s Marketing Research Review, March 1995, p. 7; and Tim Huberty, “Sharing Inside Information,” Quick’s Marketing Research Review, March 1995, p. 10. 13 Dorothy Leonard and Jeffrey F. Rayport, “Spark Innovation through Emphatic Design,” Harvard Business Review, 75(6), November–December 1997, pp. 102–113. For a look at how the design firm IDEO uses observation, brainstorming, and rapid prototyping to identify and refine product concepts, see Bruce Nussbaum, “The Power of Design,” BusinessWeek, May 17, 2004, pp. 86–94, or check the IDEO Web site, www.ideo.com. (We explore prototyping issues in Chapter 13 of this book.) Chapter Five Finding and Solving Customers’ Problems 139 In developing a revolutionary new hand-held instrument for the chemical industry, Fluke Corporation visited chemical industry trade shows and customer plants, talking informally with end users (the instrument engineers). Internally this technique was known as fly on the wall or day in the life research.14 Nokia has sent teams of employees to developing nations like Uganda for up to 12 days at a time to understand phone usage better. They learned that phone sharing is more common in these nations and sought to make their mobile phones more amenable to sharing.15 When redesigning its popular Explorer sport-utility vehicle, Ford sent a team of designers out to parking lots in order to watch how people used their cars. The researchers’ duties were not unlike those of zoologists watching animals in their natural habitat—in fact, the work was colloquially known internally as “gorilla research.” Among other ideas, the research suggested ways the Explorer could be made easier to get into.16 Similarly, Honda engineers and executives visited the homes of U.S. families that owned Ford SUVs and noted, to their surprise, how many parents put their children and their neighbors’ children in the first two rows and the dogs in the third row. Had the research been conducted only in Japan, the researchers would have entirely missed the American love affair with dogs and might consequently have made the passenger compartment too small.17 Role Playing Though role playing has long been used in psychology to enhance creativity, there is little evidence of its successful use in generating ideas for new products. Presumably, it would be valuable in instances where product users are unable to visualize or verbalize their reactions. It should also be valuable where consumers are emotionally unable or unwilling to express their views—for example, in areas of personal hygiene. Unfortunately, though users are the best place to begin the ideation, and problem analysis is widely used in one form or another, most firms still do not have organized systems to exploit this source. Considering that Levi Strauss got the idea for steel-riveted jeans from a Nevada user in 1873, one must wonder why not. An alternative way to generate concepts is based on product function analysis. A product can be expressed in two words, a verb and an object (for example, toothpaste “cleans teeth”). Thinking of new combinations of verbs and objects can suggest new product functions. In this method, hundreds of these two-word miniconcepts can be generated and shown via computer to respondents, who rate them in terms of likely interest. The highest scoring concepts are identified and in-depth interviews are conducted to explore feelings and ideas further. In an application 14 Robert G. Cooper, “From Experience: The Invisible Success Factors in Product Innovation,” Journal of Product Innovation Management, 16(2), March 1999, pp. 115–133; see also Cooper, Edgett, and Kleinschmidt, op. cit. 15 Anonymous, “Nokia’s Design Research for Everyone,” www.businessweek.com, March 14, 2007. 16 Al Haas, “Spying Helps to Improve Explorer,” Philadelphia Inquirer, December 24, 2000, p. G1. 17 Norihiko Shirouzu, op. cit. 140 Part Two Concept Generation in the food processing industry, several novel mini-concepts emerged (have fun with food, touch food), while several others fared poorly (sponge food, vaporize food). To develop these concepts further, one would need to examine why these mini-concepts were liked.18 Problem Analysis in Action One unmet need that had existed for years was the noisy candy wrapper in the theater. TV personality Gene Shalit complained one morning on the air about crackling candy bar wrappers. An expressway-commuting executive from Hercules Inc. overheard his comment and asked the laboratory for a silent candy wrapper. Polypropylene provided the answer, though not without tricky effort on heating, waterproofing, and airproofing. Toyota, Mitsubishi, and other carmakers redesigned their sport-utility vehicles to appeal more to the U.S. marketplace demand. Often, these changes come about after disappointing sales with early SUV versions. The Toyota T100 pickup had disappointing sales in the United States; consumer research suggested that the reason was that it was viewed as too small. The full-size Tundra comes complete with a V8 engine and a passenger compartment reportedly large enough for “a passenger wearing a ten-gallon cowboy hat.”19 Finally, ongoing problem analysis is critical to identification of newly emerging problems and continued improvement. Consider Domino’s Pizza. Decades ago, Domino’s founders identified a real unmet need in the market: quick, reliable pizza delivery service. Late-night customers, in fact, were satisfied with an average-quality pizza, as long as it was delivered fast and hot. Generations of customers knew Domino’s promise: “thirty minutes or it’s free.” But by 2009, competition in the pizza business had heated up; major delivery competitors such as Papa John’s had achieved immense success and even the traditional-restaurant Pizza Hut chain was getting into the delivery business. Fast and hot was no longer enough. Domino’s focus groups found that customers had lots to say about the taste, most of it negative. Company president Dan Boyle decided to respond to the threat by assigning a product team to develop a new, better-tasting pizza. Marketing employees used focus groups and other research methods to capture the voice of the customer; the food engineers developed a totally new recipe to meet the specifications. Over a dozen different sauces and crusts were tried, as well as dozens of types of cheese. Despite the risks of such a dramatic strategy (what if it were New Coke all over again and customers demanded the old product back?), the new pizza was just what the market ordered. We will revisit the Domino’s case in Chapter 7.20 18 Jeffrey F. Durgee, Gina Colarelli O’Connor, and Robert W. Veryzer, “Using Mini-Concepts to Identify Opportunities for Really New Product Functions,” Journal of Consumer Marketing, 15(6), 1998, pp. 525–541. 19 Norihiko Shirouzu, “Tailoring World’s Cars to U.S. Tastes,” The Wall Street Journal, January 15, 2001, pp. B1, B6. 20 Anonymous, “New Domino’s Pizza Recipe Doubles Quarterly Profits,” nydailynews.com, March 2, 2010; Domino’s Pizza 2009 Financial Results. Chapter Five Finding and Solving Customers’ Problems 141 Scenario Analysis So far, we have talked about going to technical and marketing people within the firm for ideas on customer problems, about searching the many files and recordkeeping places where customer concerns can be found, and about problem analysis. The fourth general source of stakeholder needs shown in Figure 5.1— scenario analysis —comes into play because the ideal problem for us to find is one that customers or end users don’t know they have at this time. As hockey star Wayne Gretzky said, “I don’t skate to where the puck is. I skate to where it’s going to be.” Similarly, we have to stay one step ahead of the customers by anticipating their problems.21 A future problem is a good problem because most problems we find in interviews and focus groups have already been told to competitors and anyone else who will listen. Providers of the goods and services have been working on them for many years, for example, flimsy music stands and steam on bathroom mirrors. We have time to solve a future problem and have that solution ready to market when the time comes. Unfortunately, end users usually don’t know what their future problems will be. And they often don’t really care, at least not right now. So they are not much help in interviews. This is where scenario analysis becomes valuable. Here’s how it works. If we were to describe apartment life 20 years from now, we would probably see lots of windows and sunlight coming in. If a furniture manufacturer were doing this scenario analysis, an analyst could immediately see problems, such as: Those apartment dwellers will need (1) new types of upholstery that are more resistant to the sun, and (2) new types of chairs that will let them continue such activities as conversing and eating but also let them gain exposure to all that sunlight. The scenario analysis procedure is evident: First, paint a scenario; second, study it for problems and needs; third, evaluate those problems and begin trying to solve the most important ones. The ideal scenario is a “stylized narrative”—that is, it should be like a story: painting a clear picture of the future state, containing a “plot” or sequence of believable events. Painting a scenario does not yield a new product concept directly; it is only a source of problems, which still must be solved. In fact, it is often valuable for concept generation if several future states are described. Creative people can then choose to focus on the most likely scenario, or possibly attempt a multiple coverage strategy in which a separate strategy is pursued for each of several possible scenarios. A carmaker might develop several different alternative engine technologies (gas/electric hybrid, hydrogen cell, etc.) in parallel if it is unclear which of these will be dominant in the future.22 Scenarios take several different forms. First, we distinguish between (1) extending the present to see what it will look like in the future, and (2) leaping into the future to pick a period that is then described. Both use current trends to some extent, of course, but the leap method is not constrained by these trends. For 21 Mark Henry Sebell, “Staying Ahead of Customers,” U.S. Banker, October 1997, p. 88. For more on using scenarios, see Steven Schnaars and Paschalina (Lilia) Ziamou, “The Essentials of Scenario Writing,” Business Horizons, July–August 2001, pp. 25–31. 22 142 Part Two Concept Generation FIGURE 5.4 The Relevance Tree Form of Dynamic Leap Scenario The analysis begins at the top of the chart (the ideal future condition that is the expected end). Working down the page, each level shows the necessary conditions for the item above it. All branches of the Relevance Tree are worked down to the conditions that already exist. Somewhere in the analysis, on one of the branches, a condition that does not exist offers someone today an opportunity for product innovation. In this diagram, with only a few of the branches completed, there appears to be an opportunity for some firm to develop better, cheaper diagnostic systems for dealers to use. (The analysis is for demonstration only.) Start Here Auto dealer service problems essentially solved Work Down Human element eliminated in dealership service departments Car components are entirely replaceable modules Better, cheaper systems available Dealers adopt advanced electronic diagnostic tools Dealers have access to capital Car nearly free of service needs Cars extensively redesigned to eliminate problems Car manufacturers' quality programs are successful Dealer managements acquire technical skills Opportunity Identified example (hypothetically) an extend study might be: Currently, homeowners are converting from individual housing to condominium housing at an annual rate of 0.9 percent. If this keeps up for 20 years, there will be 7 million condominium units in use, which will present a need for 250,000 visitors’ motel units in major condominium areas to house visitors who cannot stay in the smaller units with their hosts. The thinking of the utopian school is sometimes used. By contrast, a leap study might be: Describe life in the year 2030 in a major urban area of Germany contrasted with life in a similar setting in France. Leap studies can be static or dynamic. In dynamic leap studies, the focus is on what changes must be made between now and then if the leap scenario is to come about—the interim time period is the meaningful focus. In static leaps, there is no concern about how we get there. Figure 5.4 shows a dynamic leap period in which the auto dealer service problem no longer exists. The time between now and then is broken down to yield the technical breakthroughs needed soon to reach that ideal condition. As another illustration, one professional forecaster made several rather bold predictions regarding technologies and our lifestyles in the future (see Figure 5.5). Any of these could be viewed as a leap scenario into some time in Chapter Five Finding and Solving Customers’ Problems 143 FIGURE 5.5 One Professional Forecaster’s View of the Future Graham Molitor is a professional forecaster who relies on a variety of sources to develop his forecasts: census documents, government statistics, trade journals and similar publications, weekly newsmagazines, and his own 40 years of experience. Here are a few of the trends and forecasts he envisions for the 21st century: 1. Investment in communication will allow more people to work at home; by 2020, telecommuting and videoconferencing will have largely replaced in-person business meetings. 2. Internet use will continue to increase rapidly, and Americans will spend more on computers than on televisions. Handheld videophones will be a commercial hit by 2025. 3. Medical technology breakthroughs will continue to happen: Improved cloning technology will extend human life, and computerized health monitors will be of wearable size. 4. Ethical and social issues related to health and lifestyle will continue to be prevalent: These will include euthanasia, cloning, genetic manipulation, and biological engineering. 5. The traditional “nuclear family” will continue to become a thing of the past; by 2020 the average household size will be down to 2.35 persons. 6. By 2050, over a quarter of the U.S. population will be over 65. Large-type and recorded books, and cars that can be operated by people with reduced dexterity, will become popular. 7. By 2100, Americans of European descent will be in the minority (that is, less than 50 percent of the population). Continued diversity and multiculturalism will be stimulated by increased immigration. 8. In the far future (2200 to 2500), biotechnology and related life-science industries will have replaced tourism as a key employer in the U.S. 9. Greater globalization of manufacturing industries, more outsourcing of capital-intensive functions, and more electronic commerce, will mark business and industry. 10. Supplies of petroleum will shrink and prices will rise; by 2050, electricity demand will multiply by a factor of four. What new products do each of these forecasts suggest? Do any of them suggest any changes in the new product development process? Do any of them seem too far-fetched to believe? Source: Graham T. T. Molitor, “Trends and Forecasts for the New Millennium,” The Futurist, July–August 1998, pp. 53–59. Reprinted with permission. the not-too-distant future: These scenarios (if not too farfetched) might suggest opportunities for several new products. (Which do they suggest to you?) Another variant is a study of wild cards—high-impact, low-probability events (see Figure 5.6 for a set of wild cards identified recently by the Arlington Institute). In a wild card study, one assesses the likelihoods of occurrence of the identified events and investigates the threats or new product opportunities they suggest. While any one of them might be rather unlikely, it does not mean that one should not develop a contingency plan, especially if it may set off a chain of events that can have an impact on innovation. For example, a natural disaster may result in an epidemic, triggering border closings and quarantines and threatening the airline industry. A key here is to try to recognize the early warning signs of the wild cards, as often they will exist (possibly outside one’s own discipline).23 23 John L. Petersen, “The ‘Wild Cards’ in Our Future: Preparing for the Improbable,” The Futurist, July–August 1997, pp. 43–47. 144 Part Two Concept Generation FIGURE 5.6 Human Cloning: Raises ethical issues, as well as the possibility of extending human life thanks to organ and tissue cloning. Wild Card Events and Their Consequences No-Carbon Policy: Global warming may cause governments to put high taxes on fossil fuels, shifting demand to alternative sources of energy. This changes the allocation of R&D investment toward alternative energy, possibly causes new “energy-rich” nations to emerge, and ultimately may lead to a cleaner environment for everyone. Altruism Outbreak: This is the “random acts of kindness” movement—solve social problems rather than leaving it up to the government. Schools and other institutions will revive due to community actions, and perhaps inner cities would be revitalized. Cold Fusion: If a developing country perfects free energy, it becomes prosperous overnight. It gains further advantages by becoming an energy exporter. Other wild cards identified in the study: Civil war in the U.S., revolt in the inner cities, computer hackers blackmail the Federal Reserve. Secession of a Western state, collapse of the U.N. . . . not a bright picture! Luckily, none of these might happen, though their possible occurrence should not be ignored. Source: John I. Petersen, “The ‘Wild Cards’ in Our Future: Preparing for the Improbable,” The Futurist, July–August 1997, pp. 43–47. Reprinted with permission. Scenario analyses lead to great learning and insights, but are hard to do well. Several guidelines have been suggested for conducting a good scenario analysis: 1. Know the now. The participants must have a good understanding of the current situation and its dynamics, otherwise the future they envision will not be realistic or useful for idea generation. 2. Keep it simple. Participants will likely have difficulty understanding really complex scenarios. 3. Be careful with selecting group members. A group of about six, with contrasting or complementary viewpoints and prior experiences, works best. 4. Do an 8- to 10-year projection. Too far out, and the participants are guessing. Not far enough out, and the respondents will just extend whatever is going on now. 5. Periodically summarize progress. This keeps the group on track and avoids contradictions. 6. Combine the factors causing changes. Scenarios should not be determined by just one factor. 7. Check fit or consistency at the end. 8. Once you have done the scenario analysis, plan to use it several times. These can be expensive. 9. Reuse the group. The more scenario analyses they do, the more they enjoy the task, and the better they get at it.24 24 These points are from David Mercer, “Scenarios Made Easy,” Long Range Planning, August 1995, pp. 81–86; and Audrey Schreifer, “Getting the Most out of Scenarios,” Planning Review, September–October 1995, pp. 33–35. See also Schnaars and Ziamou, op. cit. Chapter Five Finding and Solving Customers’ Problems 145 Solving the Problems Once an important user problem has been identified, we can begin solving it. Most problem solving is probably done by members of the new products group that has been leading the concept generation work so far. They do it instinctively, from the moment they hear of a problem. There is no way we can quantify or describe the methods they use, most of it being intuitive. It is probably best for the group to attempt to solve one problem at a time, however—taking on too much in the real world can be confusing and may foster communication difficulties. Many problems are sent into the technical areas for more systematic attempt at solution. Here science and intuition rule, side by side. Some firms have it as strategy that problem solutions must come from R&D or engineering, with the solution itself being found in the application of some specific technology. A bus line wants travel problems solved by buses, and a bank probably wants problems solved by borrowing money. Besides technical people, the creative talents of marketing people are often used as well. Note that the problem has to be carefully specified in order to find a good, creative solution. P&G product developers reportedly spent months trying to solve the problem, “How can we make a green striped soap that will draw sales from Irish Spring?” It was only when they focused their attention on a new problem, “How can we make a soap that connotes freshness in its appearance, shape, and color better than Irish Spring?” that Coast (a soap with blue and white swirl patterns and a more oval shape) was developed and was ultimately successful.25 Group Creativity New products people use individual problem-solving effort, but many think that group creativity is more effective. Some scientists protest loudly that this is not true, that the synergism of groups is way overplayed. Generally, individuals can handle really new ideas and find radical solutions to problems better than groups can. Some feel that one reason small firms are more innovative than large firms is that they do not often use group creativity. Back in 1938, advertising executive Alex Osborn wrote a book about a technique he called brainstorming. All of the group ideation techniques developed since that time are spin-offs of his process and embody one idea: One person presents a thought, another person reacts to it, another person reacts to the reaction, and so on. This presenting/reacting sequence gives group creativity its meaning, and the various techniques developed simply alter how ideas are presented or how reactions take place. Brainstorming Because brainstorming techniques have been around so long, they are widely abused and misused. It is good to be able to recognize bad brainstorming, because bad brainstorming just does not work. Thomas Kelley of the design firm IDEO laid 25 Peter Wilson, “Simplex Creative Problem Solving,” Creativity and Innovation Management, 6(3), September 1997, pp. 161–167. 146 Part Two Concept Generation out several rules for making brainstorming sessions more effective. These include: mind the rules (go for a large quantity of ideas, defer judgment, no snickering allowed); number the ideas (can you hit 100 ideas per hour?); jump and build (when the group hits a plateau, the facilitator suggests a new direction); and get physical (as in the Carpet Flick case, by using odds and ends to build models and prototypes).26 The biggest change in the practice of problem solving over the past 20 years is to use brainstorming combined with other tools of creativity. We still try to avoid the bazooka effect (state an idea only to have someone shoot it down), but also to avoid the scores of easel sheets with hundreds of ideas scribbled on them. Instead, we aim for group deliberations that are exploratory, evaluative in a constructive way, hours long (versus the 20-minute brainstorming session), and built toward a few specific solutions that appear operational. IDEO uses brainstorming in combination with “lickety-stick” prototype development (see Chapter 2) to speed up innovation.27 There have been many attempts to stick with the basic idea of brainstorming, but to tweak it in some way to overcome the problems. In brainsketching, participants draw their ideas rather than expressing them in words. Some evidence shows that brainsketching helps participants draw more connections with earlier ideas when coming up with new ideas.28 Another emerging technique is called speedstorming. It is described as a round-robin format, similar to speed-dating, in which participants pair off (at random, or with some pattern in mind such as that the two participants must be from different functional areas) and discuss a topic for a three- to five-minute round. The goal of each round is to come up with ideas that can be pursued by the new product team. After each round, partners switch around and another round begins. At the end of the session, numerous new ideas have been generated, and participants have identified which partners they seem to collaborate with the best. For this reason, the proponents of speedstorming claim that it is particularly good at overcoming the communication difficulties typical of cross-functional teams.29 Some other common techniques are described in Appendix B. Electronic Brainstorming and Computer-Assisted Creativity Techniques Despite its popularity, brainstorming has several drawbacks. Only one person can talk at a time, and social loafing may occur (average work intensity may be lower in a group setting). Further, some individuals may still fear being criticized for having unpopular ideas. Electronic brainstorming, a form of brainstorming assisted by group support systems (or GSS) software, is said to overcome these limitations of traditional brainstorming, as it allows participants to all answer at once, and also to answer anonymously. 26 Tom Kelley, The Ten Faces of Innovation (New York: Currency Books, 2005). Bruce Nussbaum, op. cit. 28 Remko Van Der Legt, “Brainsketching and How It Differs from Brainstorming,” Creativity and Innovation Management, 11(1), 2002, pp. 43–54. 29 Caneel K. Joyce, Kyle E. Jennings, Jonathan Hey, Jeffrey C. Grossman, and Thomas Kalil, “Getting Down to Business: Using Speedstorming to Initiate Creative Cross-Disciplinary Collaboration,” Creativity and Innovation Management, 19(1), 2010, pp. 57–67. 27 Chapter Five Finding and Solving Customers’ Problems 147 A GSS-assisted brainstorming session may take place in a room set up with a network of computer terminals. Participants sit at the terminals and respond to questions provided by the moderator, who runs the GSS software. The GSS software gathers the participants’ responses and projects them onto a large screen at the front of the room or on the participants’ monitors. Seeing the responses stimulates even more ideas and encourages follow-up discussion. The GSS also automatically takes electronic notes of all the proceedings, so nothing is lost or erroneously transcribed.30 One is not restricted to a single location, either. GSSs can facilitate activity at many sites simultaneously (through computer linkups or videoconferencing) and handle group sizes into the hundreds. GSSs are becoming much more popular in facilitating meetings, and there is increasing evidence that electronic brainstorming outperforms traditional brainstorming in terms of productivity and output of unique ideas.31 An increasing number of firms are using computer programs such as Mindlink, Mindfisher, and NamePro to assist their creative efforts in idea generation and management, and also to help out in other creative tasks such as brand name generation and selection. While they come in many forms, many of these work by drawing from large databases of words, phrases, or even pictures, encouraging the user to think laterally (gather unrelated thoughts, then try to associate them with the problem at hand). Most are straightforward and stimulating to use.32 Also, many are adaptable to use in a GSS setting. Online Communities33 Online communities (or virtual communities) have revolutionized customer information gathering. An online community can be defined as any group that interacts using a communications medium such as online social networking. Numerous firms, including P&G, Kraft, Dell, and Hewlett-Packard, use online communities as a key part of their voice of the customer efforts and, indeed, throughout their new products process. Familiar online communities such as Facebook, Twitter, or LinkedIn are open to everyone and widely popular. But there are alternatives, some of which are much less well known. Some online communities such as 30 An assessment of GSSs is found in Robert O. Briggs and Gert-Jan De Vreede, “Meetings of the Future: Enhancing Group Collaboration with Group Support Systems,” Creativity and Innovation Management, 6(3), June 1997, pp. 106–116. 31 Keng L. Siau, “Group Creativity and Technology,” Journal of Creative Behavior, Third Quarter 1995, pp. 201–217. 32 Tony Proctor, “New Developments in Computer Assisted Creative Problem Solving,” Creativity and Innovation Management, 6(2), June 1997, pp. 94–98; and Mark Turrell, “Technology Spotlight: Unfuzzing the Front-End with Web-Based Idea Management,” Visions, 27(1), January 2003, pp. 18–21. For a critique of several of these computer programs, see Arvind Rangaswamy and Gary L. Lilien, “Software Tools for New Product Development,” Journal of Marketing Research 34, February 1997, pp. 177–184. 33 Much of this section derives from Claire-Juliette Beale, “How Online Communities Are Changing the NPD Landscape—An Introduction to the Value of This New Tool,” Visions, 32(4), December 2008, pp. 14–18. 148 Part Two Concept Generation tivocommunity.com are set up by lead users with an interest in a particular product or service; some such as Johnson & Johnson’s babycenter.com are launched by firms. In addition, service providers like MarketTools or Vision Critical obtain rich customer insights by setting up private online communities of 500 or fewer carefully selected members. Firms can also access proprietary online panels (POPs), which may contain hundreds of thousands of individuals who are statistically representative of a target market. These panels can be used to supplement online communities in a number of ways. For example, POPs can validate promising ideas or insights generated from a private online community. Firms may have a range of objectives when initiating an online community. As a VOC technique, online communities provide a listening function: They allow firms to obtain new ideas from customers and get feedback on new concepts. Working with a service provider like MarketTools, firms can monitor public communities and blogs to spot new customer trends and emerging opportunities. Online communities are also a way to establish rapport with customers, enable customer support, and build emotional bonds between customers and the firm. Del Monte Foods (makers of many pet food brands such as Kibbles ‘n Bits, Milk-Bone, Meow Mix, and 9 Lives) used online communities extensively to get a better understanding of changes in its consumer market, identifying opportunities early and quickly developing products. In 2006 they joined forces with MarketTools and brand monitoring agency Umbria to start the I Love My Dog initiative. By analyzing data from millions of blogs, user forums, and message boards, Del Monte was able to identify the things that pet owners cared about and wrote about the most. In fact, a new customer segment of dog lovers (named the “Dogs Are People, Too” segment) was identified. Next, an online community was created to encourage innovative solutions from consumers within this segment. A community of 500 consumers was contacted and asked to enter a by-invitation-only, password-protected site that encouraged interaction and mutual understanding among participants. The community generated and refined ideas for a new breakfast product, which was immediately put into development by Del Monte. During the development process, Del Monte contacted community members, either individually or in group format, about a dozen times. By summer 2007, the new product, Snausage Breakfast Bites, was launched. The process from idea to launch took only six months, half the normal time for a product in this category. Since then, Del Monte has continued to explore ways to exploit online communities. In 2008, the firm partnered again with MarketTools, this time to tap into the latter’s Moms Insight Network and quickly identify cat owners. The newly created cat owner community, named Meow Mixer, is used by Del Monte to generate ideas, develop concepts, sample new products, and obtain packaging and marketing suggestions. Like anything else, online communities take work, and the firm seeking to institute an online community must be aware of the drawbacks.34 Building and managing an online community requires hiring moderators and facilitators, and can take time—a good-sized community may take more than a year to mature. 34 Claire-Juliette Beale, “Creating Your Own Online Community—How to Avoid the Pitfalls,” Visions, 33(1), March 2009, pp. 15–19. Chapter Five Finding and Solving Customers’ Problems 149 Also, the longer the community is in operation, the more difficult it becomes to organize the content and make it easy for participants to find their way around. There are also legal issues, such as member privacy, confidentiality of statements, and content ownership, that would need to be considered. Usually, participants would be expected to sign a service agreement so that the sponsoring firm could avoid legal problems down the road. Despite the drawbacks, it is likely that online communities will be a major source of customer input for years to come. Disciplines Panel Several of today’s leading new products consulting firms believe creativity groups should actually work on a problem, not just talk about it, particularly in situations calling for significant innovation. Their approach is to assemble experts from all relevant disciplines and have them discuss the problem as a disciplines panel. A panel on new methods of packaging fresh vegetables might include representatives from home economics, physics, nutrition, medicine, ecology, canning technology, marketing, plastics, chemistry, biology, industrial engineering, agriculture, botany, and agronomy. The panel may also include outside experts. One panel working in the shampoo industry was focusing on a consumer need: to put on hair conditioner that actually sought out split ends and went to work there. An R&D person on the panel noted that the products available at the time already did that! This surprising comment led to a new product that made the claim others had overlooked, and which turned out to be very successful. Concept Generation Techniques in Action This chapter provided several creativity-stimulating techniques that can be used to generate concepts; Appendix B provides many more. Throughout the chapter, we have provided examples of firms that have successfully applied these techniques. Here are a few additional recent examples that illustrate the successful use of some other, perhaps less common, techniques. 1. Using Props. Life Savers Company wanted to develop new flavors. They hired a consultant who filled a room with samples of fruits, varieties of perfumes, and lists of dozens of ice cream flavors. Life Savers’ Fruit Juicers line came out of the session. P&G’s Duncan Hines Pantastic party cakes came from an idea stimulation session where greeting cards were among the props used. 2. Role Playing. Bausch and Lomb’s Polymer Technologies Division came up with the idea of cushioning material bonded to the lens surface by getting pairs of executives to play the roles of eyeball and contact lens. The actors had to think of ways the lens could stop hurting the eyeball while role playing. 3. Imitating Nature. Goats eat waste and emit it in the form of small pellets. This idea inspired Whirlpool in its development of the Trash Smasher compactor.35 35 Bryan Mattimore, “Eureka: How to Invent a New Product,” The Futurist, March–April 1995, pp. 34–38. 150 Part Two Concept Generation Summary Chapter 5 began our study of the many specific techniques developed by concept creators to aid them in their work. The most common approach is based on the paradigm of “find problem, solve problem,” requiring participation by many people in the firm, plus stakeholders and others outside the firm. Then, we looked at the many techniques developed to spot problems. These included (1) inputs from technical and marketing departments; (2) search of internal records from sales calls, product complaints, customer satisfaction studies, and more; (3) problem analysis as a way of involving end users and other stakeholders; and (4) scenario analysis as a way of learning about future problems. Once problems are discovered, efforts at solution can begin; most efforts are individual thinking and analysis, whether in the office or in the lab. One major group of techniques uses the label of group creativity; it includes a great variety of approaches, but most are variations of brainstorming. Next we will turn to some methods called analytical attributes, created over the years to aid marketing managers in seeking improvements while they are waiting for the approach of problem-find-solve to bear fruit. This is the approach where we start with form, then see if there is a need, and if so, then develop the necessary technology. Applications 1. “I recently met the president of a Florida university who had previously researched the new products operation in Silicon Valley firms. He wasn’t impressed. Said that sales reps told over and over about getting suggestions and tips from their customers and sending them in on call reports, but nothing ever happened. Apparently, upper-level sales and marketing executives only rarely have much customer contact, yet don’t capitalize on the contacts of salespeople. You have any ideas on how I might go about being sure this condition doesn’t exist in our various divisions?” 2. “I believe in problem analysis—that’s at the heart of things. But I sure don’t like those focus groups. I sat in on a couple last year, and all the people did was chat. And the chatting never seemed to lead to anything. After the second one was over, I quizzed the moderator, and she agreed that there had been a lot of rambling. She kept talking about the gems of knowledge we found—common threads, I believe she said. Now, honestly, isn’t that pure bunk? However, she did say she thought focus groups would be especially useful in Eastern Europe, where businesses have so many needs, and we have to be sure to cull down to the most critical ones. I wonder, suppose our Swiss trucking division could use focus groups to help them develop new services for Eastern European businesses?” 3. “You know a lot about smartphones, I imagine. Can you take me through a problem analysis using the smartphone market as an example? We are getting into the smartphone business in our electronics division, with the idea of having a direct competitor to Apple and Samsung, and I’m curious to see what problems you come up with that we haven’t solved yet.” Chapter Five Finding and Solving Customers’ Problems 151 Case: Campbell’s IQ Meals36 In 1990, Campbell Soup was the undisputed leader among U.S. soup manufacturers, with a market share of over 75 percent. Soup consumption, however, was leveling off, and top management was looking for opportunities for growth in related markets. Competitors such as ConAgra (Healthy Choice brand) and H. J. Heinz (Weight Watchers brand) were making sizeable sales and profit gains in their frozen foods lines, stressing their dietary benefits, and this seemed like a good place for Campbell to begin generating new product ideas. At the time, the U.S. public was becoming more interested in the relationship between diet and disease prevention. It seemed that, every day, health benefits were turning up in one food or another, causing fads such as oat bran to sweep the country. Campbell’s R&D department soon turned to investigating the dietdisease relationship, focusing on foods that could be used to prevent illnesses such as diabetes or cardiovascular disease (including high blood pressure). Given that 58 million Americans have some form of cardiovascular disease and another 16 million have diabetes, this focus seemed very reasonable. Soon enough, the rough idea had been generated: a line of foods with medical benefits. The rough idea now needed to be further developed. The challenge was to develop a food line that not only played a role in the prevention of these diseases, but also would be accepted and adopted by the U.S. population. Dr. R. David C. Macnair, Campbell’s chief technical officer, built an advisory board consisting of leading nutrition, heart disease, and diabetes specialists, who would scientifically analyze the new products. Campbell’s CEO at the time, David W. Johnson, was 100 percent behind the food-with-medical-benefits idea, saying that it had “explosive potential.” Soon, he was attending the advisory board meetings as well. Mr. Johnson said, “Wouldn’t you be dumbfounded by the opportunity to take a quantum leap and develop a product that could help improve the health and nutrition of the world?” With the backing of the Campbell CEO, the project was underway with a clear goal: to make the concept of healthy, vitamin-and-mineral-rich meals a reality. The Campbell food technologists found this a challenging task—one of the early prototype fiber-enriched rolls “could have been marketed as a hockey puck,” according to Macnair. By fall 1994, however, about 24 meals that passed early taste tests were ready for clinical trials to determine health benefits. Over 500 subjects ate the meals for 10 weeks, and most reported improvements in cholesterol, blood pressure, and blood sugar levels. None experienced side effects, and many reported they liked the taste. Meanwhile, Mr. Johnson created Campbell’s Center for Nutrition and Wellness, based in the Camden, New Jersey, head office and employing 30 nutrition scientists and dietitians. Next came the market test. Campbell marketing staff selected the name “Intelligent Quisine” (or IQ Meals), and a blue box or can for packaging. The plan was for UPS drivers to deliver 21 meals (mostly frozen, a few in cans) each week to test 36 This case is largely based on Vanessa O’Connell, “Food for Thought: How Campbell Saw a Breakthrough Menu Turn into Leftovers,” The Wall Street Journal, October 6, 1998, pp. A1, A12. 152 Part Two Concept Generation subjects’ doors. By January 1997, the product was being test marketed in Ohio, backed up with a print ad campaign and a 10-minute infomercial designed to stimulate toll-free calls to Campbell’s information line. Campbell also hired parttime pharmaceutical sales reps to pitch IQ Meals to doctors, and contacted leading hospitals such as the Cleveland Clinic to distribute IQ Meals and promotional material. Things were looking up! The first sign of trouble was at the phone bank. Callers found out that the oneweek sample pack cost $80, and the recommended plan (10 weeks) cost $700, and promptly hung up. Fixed-income households found the price especially steep. At the American Heart Association’s Columbus office, Campbell sponsored a lunch to promote IQ Meals’ benefits, but failed to impress many of the dietitians present. Further, Wall Street analysts had their doubts as well: One of them wrote a report titled, “UPS T.V. Dinners Drive Top Line?” Soon, Campbell executives were doubting the IQ Meals as well. Consultants were called in to assess the project’s viability, and Dale Morrison, head of international and specialty foods, cut IQ’s budget drastically. By May 1997, sales in the Ohio market test were dismal, and another problem was arising. Those that had stuck with the program since January were showing health benefits, but now many of them were reporting that they were getting tired of the same nine meals over and over again. The fate of IQ Meals was sealed in a corporate shakeup at Campbell in July 1997. Mr. Johnson, its biggest supporter, gave up his CEO position (and became Campbell’s chairman). Mr. Morrison rose to president and CEO, with a plan to expand international sales and to focus on key brands. Swanson, Vlasic, and other Campbell brands were spun off—and the marketing and promotion for IQ was terminated (though clinical trials were continued). The Center for Nutrition and Wellness researchers were reassigned. By fall 1997, Campbell announced plans to sell IQ Meals. IQ Meals seemed to be a classic case history. The idea that was generated seemed foolproof with respect to the marketplace opportunity and the associated demographic trends. Campbell would appear to be the perfect company to pull it off, given its core competencies and its willingness to expand into growth areas. The line even did well in both clinical trials and early consumer tests. But somehow, something got lost in the translation. And clearly, this is not an isolated incident. What went wrong? And what might Campbell product developers or executives have done differently? Or was this one just doomed from the start? Case: Earning Organizational Respect In this case, you and your classmates play the role of the marketing department in a firm involved in new product development. Your firm is struggling with instituting team-based product development, and over the last several months several of the marketing staff have been placed on product teams with personnel from engineering, design, and manufacturing. The experience so far has not been positive for you and your marketing colleagues. You feel that marketing is routinely left out of key team decisions and that top management seems more sympathetic to Chapter Five Finding and Solving Customers’ Problems 153 the engineers when conflicts arise within the team. You suspect that part of the reason is that most top management personnel in your firm come from an engineering background and just understand the perspectives and the decision-making style of the engineers better. You also feel that marketing has a lot to contribute to product development. There is an excellent marketing research department that can provide quick feedback on customer behavior using state-of-the-art equipment, and the sales force is second to none in the industry and routinely gathers key market information and intelligence. There are several very good creative people on staff responsible for generating high-potential ideas, which your firm has developed into many successful new product launches. One of your creative colleagues in product development suggests using a problem-based ideation approach, commonly used to generate new product ideas, to try to find a way to get top management to respect the marketing department more. Ideally, you would like them to recognize your skills, training, and experience and to appreciate and use the unique information you can bring to the new product process. You succinctly state your problem as follows: “How can we communicate the value and potential contributions of the marketing department effectively to top management, so that they will respect us more?” Using the ideation techniques given in Appendix B (or any others you prefer), develop creative solutions to this problem. First, generate at least half a dozen ideas individually. Keep a basic rule in mind: There are no bad ideas—the more, the merrier. Then, with your instructor working as a group facilitator, boil these down to the four or five best ideas and, as a group, discuss and refine these. Your goal is to arrive collectively at one or more clear, well-thought-out programs that you could realistically begin implementing soon. One other rule: Use your imagination! This is an exercise where you can really stretch. Though you can try any of the techniques given in Appendix B, some you might find particularly useful are the following: Scenario Analysis: Identify a set of trends (fashions, hot places to live/work, celebrities, exciting new products, etc.). Think about what might be suggested by or associated with any of these. Creative Stimuli: Look at the set of stimulus words provided in Appendix B and select a few of these at random. Ask yourself how each of your words suggests something that helps you solve your problem. Be creative. Forced Relationships: Forget about your problem altogether for a little while. Select a magazine. Turn randomly to a page and look at the picture on that page. (If none, leaf through the magazine until you get to one.) What does the picture suggest to you? Jot down at least half a dozen thoughts. Now, return to your problem and use the thoughts you came up with to help you think creatively about possible solutions. For a variation, use a dictionary, encyclopedia, or the Yellow Pages instead and find a random word on a random page. Use of the Ridiculous: Think of the most ridiculous idea you can. Then ask yourself if it suggests to you a not-so-ridiculous new idea. C H A P T E R S I X Analytical Attribute Approaches: Introduction and Perceptual Mapping Setting In Chapter 5, we studied an approach to concept generation that involves identifying users’ problems and finding solutions to them. The problem-based approach is very useful because product concepts found by this route are most likely to have value for the user. This chapter introduces a different set of techniques that are commonly used in the problem-solving phase (see Figure II.1 at the beginning of Part II). Everyone involved with the creation and sale of goods and services can make use of these techniques, including some who don’t even know they are doing formal concept generation. What these techniques do is create views of a product different from the usual ones—they can seem almost magic, but are quite deliberate. They can appear to be strictly fortuitous or lucky when they work, and they have indeed worked—many times, as with adding a third stocking to a package, quick-drying inks, and smartphones that search the Internet. But actually, they are quite deliberate and purposeful, allowing discovery—serendipitous findings that come to people who know what they are looking for. We refer to these techniques as analytical attribute techniques, and they are our concern in Chapters 6 and 7. Understanding Why Customers Buy a Product Products Are Groups of Attributes What is a product attribute? Figure 6.1 shows the set of them. A product is really nothing but attributes, and any product (good or service) can be described by Chapter Six FIGURE 6.1 Analytical Attribute Approaches: Introduction and Perceptual Mapping 155 A. Product attributes (for our purposes) are of three types: A Typology of Attributes Features Features can be many things: Dimensions Source ingredients Services Structures Functions Benefits Esthetic characteristics Manufacturing process Performance Trademarks Components Materials Price And many more Benefits can be many things: Uses Sensory enjoyments Economic gains Savings (time, effort) Nonmaterial well-being And many more Benefits are either direct (e.g., clean teeth) or indirect (e.g., romance following from clean teeth). Functions are how products work (e.g., a pen that sprays ink onto the paper). They are unlimited in variety, but are not used nearly as often as benefits and features. B. Analytical attribute approaches use different attributes: Dimensional analysis uses features Checklists use all attributes Trade-off analysis also uses determinant attributes Several methods in Appendix B use functions and benefits citing its attributes.1 Attributes are of three types: features (what the product consists of), functions (what the product does and how it works), and benefits (how the product provides satisfaction to the user). Benefits can be broken down in an almost endless variety—uses, users, used with, used where, and so on. Concept generation is a creative task, so great liberty has been taken with definitions in its activity. The classification system used in this book is an attempt, and no more than that, to arrange them for study. It is important here to recognize that it makes sense for us to define attributes broadly. A pair of shoes can be thought of as a group of attributes; a person may buy a given pair because she likes the appearance of the leather (feature), because they are excellent walking shoes (function), or because they are very comfortable (benefit). (And if you disagree with the classification of these attributes as features, functions, or benefits, that’s OK too!) A spoon is a small shallow bowl (feature) with a handle (another feature) on it. The bowl enables the spoon to function as a holder and carrier of liquids. The benefits include economy and neatness of consuming liquid materials. Of course, the spoon has many other features (including shape, material, reflection, and pattern), not to mention other functions (it can pry, poke, project, and so on, as school cafeteria managers know all too well) and benefits (such as pride of ownership, status, or table orderliness). 1 For a useful perspective on how to conduct research on identifying what attributes are most valued by customers, see Charles Miller and David C. Swaddling, “Focusing NPD Research on Customer-Perceived Value,” in P. Belliveau, A. Griffin, and S. Somermeyer (eds.), The PDMA Toolbook for New Product Development (New York: John Wiley, 2002), pp. 87–114. 156 Part Two Concept Generation Theoretically, the three basic types of attributes occur in sequence. A feature permits a certain function, which in turn leads to a benefit. A shampoo may contain certain proteins (feature) that coat the hair during shampooing (function), which leads to more shine on the hair (benefit). Analyzing Product Attributes for Concept Generation and Evaluation Analytical attribute techniques allow us to create new product concepts by changing one or more of its current attributes or by adding attributes, and to assess the desirability of these concepts if they were to be developed into products. That is, these techniques can be used in concept generation (which will be approached in this chapter), but also in concept evaluation and even further along in the new products process, as you will see in succeeding chapters. If we were to change current product attributes in all the ways they could be changed, or to think of many additional attributes that could be built into the product, we would eventually discover every change that could ever come about in that product. Other techniques capitalize on relating one attribute with another attribute (or to something else in the environment), forcing these relationships, whether normal and logical or strange and unanticipated. They all can work, as you will see. And they have been used in all product categories from new car lines at Ford or Toyota to new eyeglass styles and cereal brands. Analytical attribute techniques are felt to be more useful in Western culture than in Eastern. Western (particularly European and North American) thought goes heavily toward rearranging things, while Eastern (Asian) thought tends to start work anew.2 Commodity-type products are a major focus because slight rearrangements can differentiate one item from its competitors, thus allowing it to carry a higher price. There are a variety of quantitative and qualitative attribute analysis techniques available. In this chapter, we explore one common quantitative technique: perceptual gap analysis. After an introduction to determinant gap maps, we will show how perceptual mapping techniques such as factor analysis and multidimensional scaling (MDS) can be used to generate perceptual gap maps. These techniques are frequently used in concept generation, and indeed, throughout new product development, during launch, and even beyond. We shall be returning to them from time to time as we proceed through the new products process. Chapter 7 examines a second common quantitative technique, conjoint analysis, and several qualitative techniques such as dimensional analysis, checklists, relationships analysis, and analogy. Many more techniques are also given in Appendix B. Gap Analysis Gap analysis is a statistical technique with immense power under certain circumstances. Its maps of the market are used to determine how various products are perceived by how they are positioned on the market map. On a geographical map, New York City is much closer to Pittsburgh than it is to Los Angeles. But on a 2 Jacquelyn Wonder and Jeffrey Blake, “Creativity East and West: Intuition versus Logic,” Journal of Creative Behavior, Third Quarter 1992, pp. 172–185. Chapter Six Analytical Attribute Approaches: Introduction and Perceptual Mapping 157 nearness-to-the-sea map, New York City would be right next to Los Angeles. On any map, the items plotted tend to cluster here and there, with open space between them. These open spaces are gaps, and a map that shows gaps is, not surprisingly, called a gap map. Several levels of sophistication will be cited, because many firms prefer to use the technique in a simple form, while others have achieved their greatest success with the more complex versions. Gap maps are made in three ways: (1) Managerial expertise and judgment is used to plot products on a map and make a determinant gap map; (2) a manager uses customer attribute ratings to get data from users for an AR perceptual gap map; and (3) a manager uses overall similarities to get data from users for an OS perceptual gap map. Determinant Gap Maps Figure 6.2 shows a map of snacks prepared by members of a new products team seeking to enter the snack market. The map consisted of two dimensions (they personally thought crunchiness and nutritional value were important in snacks). Scales ran from low to high on both factors. Each brand then in the market was scored by the managers on each of the two factors. While the scoring may seem arbitrary and subject to managerial error, determinant gap maps are often a good place to start. Remember, concept generation takes place after strategy (the PIC) has targeted a market or user group on which to focus. Either the firm had experience in this market (a strength) or the market was researched. Each brand was then entered on the diagram (Figure 6.2) according to FIGURE 6.2 Gap Map for Snack Products Crunchiness Pretzels Peanuts High Fritos Nachos Potato chips Popcorn Corn curls Beef jerky Candy bar Pizza Pie Cake Apple $? Soda crackers Cookies Low Raw vegetables Granola bar Bagels Raisins Nutritional Value High $? Donut Banana Cheese Yogurt Ice cream $? Soda pop Fruit juice Low 158 Part Two Concept Generation its scores. The result was a map of the brands, each in relationship to all others, on these two factors. Many maps could have been prepared, each with a different pair of attributes. They can also be three-dimensional. But the managers providing inputs to the determinant gap map are not new to this industry; they would have valuable and useful beliefs and judgments that may be very helpful in guiding concept generation. (Of course, they can still be wrong. Do you disagree with any of the assigned positions in Figure 6.2? Look closely!) Attributes used in gap analysis should normally be differentiating and important. Consumers differentiate snacks on their crunchiness and on their nutritional value. And these attributes are important in buying snacks. Snacks also have different shape esthetics, but these are not often used to differentiate one from another. Even if they were, most people would probably not think them important. Attributes that both differentiate and are important are called determinant attributes, because they help determine what snacks are bought. In an industrial study of vinyl siding, some of the determinant attributes identified were appearance/status, maintenance/weathering, application/economy, and dent resistance.3 The reason it is important to use determinant attributes in making the maps is that our purpose in this method is to find a spot on the map where a gap offers potential as a new item, one that people might find different and interesting. For example, on the snacks map in Figure 6.2 the circles marked “$ ?” are gaps, and thus offer new product possibilities. Note that the large number of snacks makes our gaps few and small—for example, the gap of semi-high crunchy and semi-high nutritional is close to the granola bar, the apple, beef jerky, and soda crackers. Determinant gap maps are speedy and cost-efficient, but have the weakness of being driven by only managerial judgment. Customer perceptions may indeed be quite different. Plus, brand perceptions might be more difficult for managers to judge correctly. In Figure 6.2, we might all agree that potato chips have lower nutritional value than granola bars, but how do customers perceive different brands of granola bars? Do they really think Nature Valley bars are the most nutritious, best tasting, or lowest in calories? And how important are each of these attributes to customers when they form preferences? Techniques that gather customer perceptions use them to develop gap maps that can provide important (and perhaps surprising) insights to the manager. We now explore two commonly used types of perceptual gap maps. Perceptual Gap Maps Based on Attribute Ratings (AR) Unlike the determinant gap map method, attribute ratings (AR) perceptual gap mapping asks market participants (buyers and users of the product) to tell what attributes they believe products have. For example, product users may think candy bars are high in nutrition—doubtful, but if this were so, then any map putting 3 Steven A. Sinclair and Edward C. Stalling, “Perceptual Mapping: A Tool for Industrial Marketing: A Case Study,” Journal of Business and Industrial Marketing, Winter/Spring 1990, pp. 55–66. Chapter Six Analytical Attribute Approaches: Introduction and Perceptual Mapping 159 FIGURE 6.3 ts en 700 d on • • sp 2 • e R 1 1 2 1 2 • • • • • • • 15 3 . . . Choices . . . X Ideal Attributes A Data Cube candy bars low in nutrition is incorrect for seeing perceptual gaps. Determinant maps are based on reality as viewed by the new products manager (or, perhaps, by the firm’s R&D personnel). Perceptual maps, as the name implies, are based on marketplace perceptions of reality, which may or may not be accurate. They can complement each other, and both have a place in our work. In AR perceptual gap mapping, we begin with a set of attributes (again, these can be features, benefits, or functions) that describe the product category being considered. We gather customers’ perceptions of the available choices (brands, manufacturers, etc.) on each of these attributes. Typically, this is done using 1-to-5 or 1-to-7 scales (commonly called “Likert-type” scales), where the endpoints are “strongly disagree” and “strongly agree” with each attribute statement provided. We also ask customers which attributes are important in their purchases of products in this category. This procedure results in a formidable data cube (Figure 6.3), which while perhaps impressive in size, is not very helpful to managers. In Figure 6.3, the perceptions of the available choices on each attribute would appear under Brands 1 through X, while the importances of the attributes would be given in the Ideal column. The challenge is then to reduce the data cube into something more manageable— namely, a perceptual map. Factor analysis, a statistical technique available in computer packages, is typically used to reduce the large number of attributes to a small number of underlying dimensions (also called factors), which can then serve as the axes of the perceptual map. Other techniques beyond the scope of this book, such as multiple discriminant analysis, can also be used. Cluster analysis (to be presented in a later chapter) can then be used to group individual respondents together into benefit segments based on their preferences. Suppose, for example, that you are a product manager at a firm that makes women’s swimsuits. Based on your industry experience and your knowledge of the market, you have developed a set of attributes that customers use in evaluating and comparing swimsuits. You have commissioned a research study in which female respondents were asked to identify all the brands of swimsuits they are familiar with and to rate them on each of the attributes on 1-to-5 Likert-type scales (see Figure 6.4). They are also asked to state how important each of these attributes 160 Part Two Concept Generation FIGURE 6.4 Attribute Perceptions Questionnaire Rate each brand you are familiar with on each of the following: Disagree 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Attractive design Stylish Comfortable to wear Fashionable I feel good when I wear it Is ideal for swimming Looks like a designer label Easy to swim in In style Great appearance Comfortable to swim in This is a desirable label Gives me the look I like I like the colors it comes in Is functional for swimming Agree 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 1..2..3..4..5 is when deciding which brand of swimsuit to buy, again using 1-to-5 Likert type scales. The average ratings of each brand on each attribute are presented in the snake plot of Figure 6.5. The snake plot (the name refers to the snakelike shape of the lines that join the points) reveals some useful information. For example, respondents tend to think FIGURE 6.5 Snake Plot of Brand Ratings 1 2 Average Rating 3 4 1. Design 2. Stylish 3. Comfortable to wear 4. Fashionable 5. I feel good Attribute 6. Ideal swimming 7. Designer label 8. Easy to swim 9. In style 10. Appearance 11. Comfortable to swim 12. Desirable 13. Look I like 14. Like colors 15. Functional Sunflare Islands Aqualine 5 Chapter Six Analytical Attribute Approaches: Introduction and Perceptual Mapping 161 that Aqualine is more comfortable to wear and easier to swim in than Sunflare (attributes 3 and 8), while Sunflare has a more attractive design and is more stylish than Aqualine (attributes 1 and 2). But there is just too much in Figure 6.5 for it to be of much help in identifying a lucrative perceptual gap, and we still seem to be far away from a simple pictorial representation as seen in Figure 6.2. Closer inspection of Figure 6.5 suggests that there may be underlying patterns in the data. We notice, for example, that choices that are rated high on “attractive design” tend also to be perceived as “fashionable,” “designer label,” and so on. We might say that these attributes seem to hang together. Similarly, other attributes (“comfortable to wear,” “easy to swim in,” and “comfortable to swim in”) also seem to hang together. There may be a small number of such underlying dimensions or factors that explain most of the variation in perceptions presented in Figure 6.5. If we could identify these factors, then we would no longer need all the attributes: We could present most of what we know about customer perceptions using just the factors. We apply a factor analysis computer program to the customer perception data to identify these factors. The first challenge we face is to determine how many underlying factors to retain in the model, as this is seldom clear-cut. One rule of thumb is to plot incremental percent variance, explained as shown in Figure 6.6. As this figure shows, FIGURE 6.6 Scree Plot and Eigenvalue Test Factor 1 2 3 4 5 6 7 8 9 Eigenvalue 6.04 3.34 0.88 0.74 0.62 0.54 0.52 0.44 0.40 Percent Variance Explained 40.3 22.3 5.9 4.9 4.2 3.6 3.5 3.0 2.7 Percent variance explained 45 40 35 30 25 The scree 20 15 10 5 0 1 2 3 4 5 6 Factors 7 8 9 162 Part Two Concept Generation FIGURE 6.7 Attribute Factor Loading Matrix for Swimsuit Data 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Attractive design Stylish Comfortable to wear Fashionable I feel good when I wear it Is ideal for swimming Looks like a designer label Easy to swim in In style Great appearance Comfortable to swim in This is a desirable label Gives me the look I like I like the colors it comes in Is functional for swimming Factor 1 “Fashion” Factor 2 “Comfort” .796 .791 .108 .803 .039 .102 .754 .093 .762 .758 .043 .807 .810 .800 .106 .061 .029 .782 .077 .729 .833 .059 .793 .123 .208 .756 .082 .055 .061 .798 Factors 1 and 2 both explain a lot of variance, but going from two to three factors does not add very much to the model. This provides some evidence that the first two factors should be retained. This procedure is called the scree test. (A scree is a pile of rocks at the foot of a mountain. Figure 6.6 resembles the side of a mountain, and the cutoff is made at the scree.) The factor analysis procedure also provides a useful statistic (called the eigenvalue) for each factor, which is mathematically related to the amount of variance explained. A second rule of thumb is to keep only those factors whose eigenvalues are greater than 1. Figure 6.6 shows that the first two eigenvalues pass this hurdle (they are 6.04 and 3.34, respectively). Thus, both the scree test and the eigenvalue rule suggest that the two-factor solution is satisfactory. The factor analysis program then calculates a factor loading (or factor pattern) matrix, showing the correlation of the original set of attributes to their underlying factors. Figure 6.7 shows the rotated factor loading matrix obtained for the swimsuit data.4 Attribute 1 (“attractive design”) clearly loads on the first factor much more than on the second factor (the loadings are 0.796 and 0.061, respectively; in Figure 6.7, the large loadings are underlined and boldfaced for clarity). As the table shows, Attributes 2, 4, 7, and five others also load on the first factor in addition to Attribute 1. Similarly, a different set of attributes (3, 5, 6, and so on) load on the second factor. So what should we call the two factors? Again, there is no right answer; this is part of the analyst’s art. But, glancing at the attributes that loaded onto Factor 1 (“attractive design,” “stylish,” “fashionable,” “looks like a designer label,” 4 The factor loading matrix of Figure 6.7 has been varimax rotated. This procedure rotates the axes to aid in interpretation of the resulting factors by forcing the column entries to be close to 0 or 1. For details, see Gilbert A. Churchill Jr. and Dawn Iacobucci, Marketing Research: Methodological Foundations, 8th ed. (Fort Worth, TX: Dryden, 2002). Chapter Six FIGURE 6.8 Factor-Score Coefficient Matrix Analytical Attribute Approaches: Introduction and Perceptual Mapping 163 Attribute 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Attractive design Stylish Comfortable to wear Fashionable I feel good when I wear it Is ideal for swimming Looks like a designer label Easy to swim in In style Great appearance Comfortable to swim in This is a desirable label Gives me the look I like I like the colors it comes in Is functional for swimming Factor 1 “Fashion” 0.145 0.146 20.018 0.146 20.028 20.021 0.138 0.131 20.021 0.146 20.029 0.146 0.148 0.146 20.019 Factor 2 “Comfort” 20.022 20.030 0.213 20.017 0.201 0.227 20.020 0.216 20.003 0.021 0.208 20.016 20.024 20.022 0.217 Sample calculation of factor scores: From the snake plot, the mean ratings of Aqualine on attributes 1 through 15 are 2.15, 2.40, 3.48, . . . , 3.77. Multiply each of these mean ratings by the corresponding coefficient in the factor-score coefficient matrix to get Aqualine’s factor scores. For example, on Factor 1, Aqualine’s score (2.15 3 0.145) 1 (2.40 3 0.146) 1 (3.48 3 20.018) 1 · · · 1 (3.77 3 20.019) 5 2.48. Similarly, its score on Factor 2 can be calculated as 4.36. All other brands’ factor scores are calculated the same way. “in style,” and so on), we see a common thread. We might call this factor “fashion.” The second factor might be called “comfort” as its attributes all seem to have to do with comfort or ease of wear. Figure 6.7 shows the factor names at the top of each column. Incidentally, the fact that these two factors are easily interpretable is further proof that the two-factor solution is good. Several apparently unrelated attributes can sometimes get forced together onto a single factor: This may be a sign that too few factors were chosen. The program also calculates the matrix of factor-score coefficients (see Figure 6.8). These are regression weights that relate the attribute scales to the factor scores. Thus, since we know how each choice is rated on each individual attribute (this information is in the snake plot), we can use the factor-score coefficient matrix to estimate how they would have been rated relative to the underlying factors. These estimates, called factor scores, can be used to draw the perceptual map, which appears in Figure 6.9. The perceptual map shows that Aqualine is perceived as the most comfortable brand (score on Factor 2 is 4.36), but is low in fashion (score on Factor 1 is 2.48). Sunflare is the most fashionable swimsuit but is also perceived to be the most uncomfortable; Splash is rated quite low on both factors; and the other two choices occupy intermediate positions in perceptual space. Recall that this is information on how customers perceive the products; it may be quite unlike what management had previously believed. So we have gone from the messy snake plot of Figure 6.5 to the perceptual map of Figure 6.9. Granted, the perceptual map does not have all the information contained in the snake plot. But we have retained the two most important factors (in 164 Part Two Concept Generation FIGURE 6.9 AR Perceptual Map of Swimsuit Brands Comfort 5 Aqualine Gap 1 4 Islands Molokai 2 3 Splash 4 5 Fashion 1 Sunflare 2 Gap 2 1 Numbers along the axes represent factor scores. terms of variance explained) underlying customer perceptions. Thus we have a simple visual representation that is easily used and understood by managers, and that contains most of the information we started with. The perceptual map we just built resembles the snack map of Figure 6.2, and the search for gaps can proceed as before. Since the perceptual map was built using actual customer perceptions, any gaps found are more likely to interest the potential users.5 For example, the perceptual map suggests that customers perceive some swimsuits to be comfortable and others to be fashionable, but none offers both high comfort and high fashion (Gap 1 in Figure 6.9). Congratulations—you’ve just uncovered a gap! Perceptual Gap Maps Based on Overall Similarities (OS) AR perceptual maps suffered a criticism early on that led to a variation preferred by some product innovators. The problem was that users sometimes make purchase decisions using attributes they cannot identify. These phantom attributes 5 For more information on the use of factor analysis in new products, see Uwe Hentschel, “On the Search for New Products,” European Journal of Marketing 5, 1976, pp. 203–217. Chapter Six Analytical Attribute Approaches: Introduction and Perceptual Mapping 165 don’t show up on the lists, are not included as map dimensions, and by their absence distort the analysis. Also, some users have difficulty scoring attributes, even when they are aware of them, because they are simply unable to, or are unwilling to do so. In a focus group setting, some participants may not want to reveal something they feel is socially undesirable, for example. AR methods essentially view products as bundles of attributes. For AR to be effective, then, the attribute set needs to be complete. (If we had forgotten to include comfort-related attributes in the above analysis, our results would have been very different, and very misleading!) Also, customers should, by and large, make their purchase decisions according to these attributes. In a product category like cologne, for example, the customer’s decision may be driven more by brand image, aesthetics, or other attributes that are notoriously difficult for them to verbalize. DuPont offered an early example of the phantom problem. The company sold filler material for pillows and wanted to find the best type and form of filler to enhance its sales to pillow manufacturers. But DuPont market analysts found that consumers could not clearly describe the attributes of pillows and could not communicate the attributes they wanted in pillows. So the firm created many different types of pillows and then gave them to consumers three at a time, along with the question, “Which two are most similar, or which one is least like the other two?” DuPont’s research was much more complex than this question implies; but, in essence, the firm was now able to use a computer algorithm to convert the similarities data into a map showing closeness of products, without knowing a priori which attributes created that closeness. OS techniques do not require customers to rate choices on individual attributes. Rather, these techniques run on perceptions of overall similarities between pairs of brands. If there are five choices (as in the swimsuit example), there are 10 possible pairs. There are a couple of ways the data can be collected. Respondents could rank the pairs from most similar to most dissimilar, or rate pairs on, say, a 1-to-9 Likert-type scale where 1 is “very similar” and 9 is “very dissimilar.” If we had gathered similarities data on swimsuits using Likert-type similarity scales, we might have ended up with average similarity ratings as shown in Figure 6.10. This figure shows that customers tend to see Sunflare and Molokai as relatively similar (recall that lower ratings mean greater similarity), and Aqualine and Sunflare as very dissimilar. The next step is to convert the customer data (similarity ratings or rankings, depending on what data were gathered) into a perceptual map. In a very simple example, if you think Coke and Pepsi are very similar, and both are very different FIGURE 6.10 Dissimilarity Matrix Aqualine Islands Sunflare Molokai Splash Aqualine Islands Sunflare Molokai Splash X 3 X 9 8 X 5 3 5 X 7 4 7 6 X 166 Part Two Concept Generation FIGURE 6.11 OS Perceptual Map of Swimsuit Brands m Co t r fo Aqualine Islands Molokai n hio Fas Splash Sunflare from Dr Pepper, you could easily draw a map of your perceptions on a single line: Put Coke and Pepsi on the left and put Dr Pepper on the right. In the same way, we could eyeball the ratings in Figure 6.10, though the task would admittedly be difficult. Alternatively, we could use a computer program such as multidimensional scaling (MDS) to develop a perceptual map from the similarities data. MDS attempts to plot the choices on a map such that the similarities are best preserved (i.e., swimsuits that should be together, are together). Figure 6.11 shows the perceptual map obtained from the similarity ratings. The MDS-based perceptual map seems very similar to Figure 6.9, which was derived from factor analysis. In fact, the relative positions of the swimsuits are not that different. There is one important difference, however: The axes are not defined! MDS provides the relative positions only; follow-up analysis must be done to define the axes (there may be more than two) and determine what the relative positions mean. A manager knowledgeable about the industry might be able to infer the meanings by examining the points. For example, since Aqualine is known to be the most comfortable brand, and Splash and Sunflare are generally viewed as less comfortable, the north-south direction might represent comfort (north being more comfortable). In the same way, the more fashionable swimsuits seem to be toward the right, suggesting that the east-west direction represents fashion. Alternatively, computer programs can be used to assist in naming the axes if measures on specific attributes had been obtained from the respondents. One of the most common of these is PROFIT (for PROperty FITting), which fits vectors to the map that best correspond to the swimsuits’ positions. If, for example, customers were asked to rate each brand on comfort and fashion, PROFIT might have fit Chapter Six FIGURE 6.12 Comparing AR to OS Perceptual Mapping Analytical Attribute Approaches: Introduction and Perceptual Mapping 167 AR Methods OS Methods Input Required Ratings on specific attributes Attributes must be prespecified Overall similarity ratings Respondent uses own judgment of similarity Analytic Procedures Commonly Used Factor analysis Multidimensional scaling (MDS) Graphical Output Shows product or brand positions on axes Axes interpretable as underlying dimensions (factors) Shows product positions relative to each other Axes obtained through follow-up analysis or must be interpreted by the researcher Where Used Situations where attributes are easily articulated or visualized Situations where it may be difficult for the respondent to articulate or visualize attributes Source: Adapted from Robert Dolan, Managing New Product Development Process, 1st Edition, Copyright © 1993. Reprinted by permission of Pearson Education, Inc., Upper Saddle River, New Jersey. vectors corresponding to these attributes as shown in Figure 6.11. The most fashionable swimsuits tend to be in the direction of the fashion vector. Figure 6.12 compares the advantages and disadvantages of AR and OS perceptual mapping methods. Both methods are readily available as parts of easy-to-use commercial software packages and can generate detailed results at quite a low cost. Comments on Gap Analysis All gap mapping is controversial, but perceptual maps especially so. The input data come entirely from responses to questions about how the choices differ. Nuances and shadings are necessarily ignored, as are interrelationships and synergies. Creations requiring a conceptual leap are missed. In the early 1900s, for example, gap analysis might have led to breeding faster horses or to wagons with larger wheels, but it probably would not have suggested the automobile. The most troublesome aspect is that gap analysis discovers gaps, not demand. Gaps often exist for good reasons (e.g., fish-aroma air freshener or aspirin-flavored ice cream). New products people still have to go to the marketplace to see if the gaps they discovered represent things people want. Returning to Gap 1 in Figure 6.9, we do not know yet whether the market wants a very comfortable, very fashionable swimsuit. Inspection of Figure 6.9 also suggests there is a gap at medium fashion, low comfort (Gap 2). Maybe this is a better bet for a new concept. To answer this question, we need to turn to the importance data, which, as you will recall, we collected at the same time as we gathered the perceptual data. We will continue this example in Chapter 9 when we analyze customer preferences and identify benefit segments. 168 Part Two Concept Generation And, as in all ideation of new products, people must avoid being bound by what is now impossible. For example, for years gap maps on analgesics showed a big hole where strength was paired with gentleness. The strong/gentle part of the map was always empty, and everyone knew why—an over-the-counter analgesic that was potent yet didn’t irritate the stomach could not be made yet. It made sense for pharmaceutical companies to do research to develop analgesics that had both desired attributes. Of course, Extra-Strength Tylenol and later products such as Aleve were eventually developed that filled this gap, offering both strength and gentleness. As another example, for many years, most popular brands of soap were positioned as either deodorants (like Dial) or moisturizers (like Dove), and a perceptual gap (a brand that offered both attributes) existed. When new brands of soap such as Lever 2000 were launched that offered both deodorizing and moisturizing abilities, they very successfully filled that gap.6 Summary In this chapter, we have examined the use of gap maps in identifying potential product concepts. As we have noted, this technique will come in handy in later phases of the new products process, when we begin working with customer preferences and positioning (and repositioning) our products. We have looked in depth at both attribute-based and overall similarity–based perceptual mapping, each technique having its own advantages and disadvantages. Other attribute-based approaches are also available to us at the concept generation stage. Chapter 7 will introduce us to conjoint (trade-off) analysis and several other less quantitative approaches, which can also be useful in generating potentially lucrative product concepts. Applications 1. “One method you say you studied is of great interest to me, for reasons I’ll not go into. It’s gap analysis, especially the idea of maps. Several of our best divisions produce and sell services. Is the gap map method applicable to services? Could you please take, say, the college education market and draw up a product map for it? I understand it can be done by a manager at a desk, although, of course, it wouldn’t be nearly as accurate as if we had all the technical data, and so on. But could you try?” 2. “OK, you’ve identified a gap in the swimsuit market in your little example. Some customers like fashionable swimsuits, some like comfortable ones. We already know that. So isn’t it obvious you should design a swimsuit that’s both fashionable and comfortable? What insights did you get from the gap analysis that you couldn’t have figured out on your own?” 6 See Robert M. McMath and Thom Forbes, What Were They Thinking? (New York: Times Business Books, 1998), pp. 184–185. Chapter Six Analytical Attribute Approaches: Introduction and Perceptual Mapping 169 3. “A few years ago, most peanut butters were sold on the basis of their perceived quality (well-known brands vs. store brands) and crunchiness. There wasn’t a lot of difference among the competitors. Then Skippy hits the market, claiming to be healthier than other brands because it uses less salt. They didn’t even position themselves on the traditional attributes in their market. How does your gap map account for that?” Case: Comparing Smartphones (A)7 The smartphone, or mobile phone with advanced computing capability and connectivity features, is a part of daily life for an increasing number of people, and needs no introduction. Early smartphones combined the features of a cell phone and personal digital assistant (such as a Palm Pilot), while later models added portable media players, digital cameras and video cameras, and GPS navigation capability. Today’s smartphones offer touchscreen, Web browsing, mobile apps, and access to mobile commerce, all powered by Wi-Fi and mobile broadband. Of the hundreds of different smartphone models available on the market, five are commonly listed as the most popular or most feature-rich. These are the topof-the-line models produced by five leading manufacturers of smartphones. In no particular order, these are: the Nokia Lumia 920, the Sony Xperia ZL, the Samsung Galaxy S4, the iPhone 5, and the HTC One. Each has its particular strengths, and purchasing the best smartphone is always a trade-off between various features and price. The following table provides a direct comparison of all of these phones on common attributes: weight, price, screen size (measured diagonally), pixel density (for improved picture), talk time before battery needs recharging, amount of internal storage and RAM, CPU speed, and quality of the front and rear cameras (measured in megapixels). Nokia Lumia 920 Price ($) Weight (grams) Screen (inches) Pixel Density (pixels/inch) Talk Time (hours) Internal Storage (GB) RAM (MB) CPU Speed (GHz) Front Camera (Megapixels) Rear Camera (Megapixels) 7 449 185 4.5 332 17 32 1024 1.5 1.3 8.7 Sony Xperia ZL Samsung Galaxy S4 549 579 151 130 5 4.99 441 441 10 14 16 1 64 GB card 16 1 64 GB card 2048 2048 1.5 1.6 2 2 13 13 iPhone 5 HTC One 649 112 4 326 8 16 1024 1.2 1.2 8 579 143 4.7 468 19 32 2048 1.7 2.1 4 Sources: Definition of smartphones is from wikipedia.com; ratings are from smartphones. findthebest.com. The case is meant to provide a simple illustration of positioning maps and includes phones that are commonly rated as superior quality. 170 Part Two Concept Generation Pick two of the nonprice attributes that you consider to be most important in choosing a smartphone. Using these two attributes, construct a positioning map for this industry using the information presented in the case. (Alternatively, you can try a “per-dollar map”: divide the ratings on each attribute by price before constructing the map. Per-dollar maps give you an idea of how much of the attribute you get per dollar spent and improve the relative position of lower-price products.) Then, pick two other attributes and do the same (you will have constructed two positioning maps). Discuss the relative positions of the iPhone 5 and its major competitors on the selected attributes. Do you think the iPhone is well positioned with respect to its competitors? Which competitor(s) should the iPhone be the most concerned about? Why? What additional information might you want to have about the competitors and/or about the marketplace at this point? What factors accounting for the iPhone’s continued success are not considered in a positioning map such as this one? How might a seemingly “weaker” competitor (i.e., outpositioned by the iPhone) on key attributes make a dent in the iPhone’s sales share? C H A P T E R S E V E N Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques Setting The previous chapter presented market research techniques that are very frequently used to analyze customer perceptions and trade-offs and to generate promising product concepts. We begin this chapter with another useful and common quantitative technique: trade-off (or conjoint) analysis. These techniques will be encountered in subsequent phases of the new products process and, in a way, serve to provide continuity and guidance to the process (i.e., customer perceptual and preference data generated here can be used as inputs into protocol specification). We will then encounter several analytical attribute approaches that are more qualitative in nature. While less numbers-oriented, they are very helpful in getting customers and managers to think in creative ways to generate new product concepts. These and the quantitative approaches complement each other well in concept generation and development. For example, dimensional or relationships analysis could be used to help identify determinant attributes for subsequent use in an AR gap analysis; or any of the qualitative approaches could aid in interpreting a perceptual map produced by AR or OS methods. Trade-Off Analysis Trade-off analysis (often called conjoint analysis) is a technique that is more commonly used in concept evaluation, so we will meet it again in Chapter 9; but it can be used in generating high-potential concepts for future evaluation, and so it is introduced here. You will likely encounter both terms, though they are not 172 Part Two Concept Generation interchangeable. Trade-off analysis refers to the analysis of the process by which customers compare and evaluate brands based on their attributes or features. Conjoint analysis is the name of one of the most common analytical tools used to assess trade-offs (much like factor analysis is a tool that is used to develop perceptual maps). Trade-off analysis is thus the broader term. In this text, we will use “conjoint analysis” when we are specifically referring to that technique for assessing trade-offs. Recall that after finding the determinant attributes (important attributes on which the available products differ), gap analysis plots them on maps. In using conjoint analysis, we assume we can represent a product as a set or bundle of attributes. Conjoint analysis puts all of the determinant attributes together in new sets and identifies which sets of attributes would be most liked or preferred by customers. In fact, AR gap analysis output can be used to select the attributes used in conjoint analysis. Using Trade-Off Analysis to Generate Concepts Let’s say coffee has three determinant attributes: flavor, strength, and intensity of aroma. As Figure 7.1 shows, there are several different levels available for each of these attributes. If somehow we could get customer preferences (or utilities) for each attribute separately, we could combine the best level of each attribute into an overall favorite product. As shown in Figure 7.1, customers prefer medium strength, no (added) flavor, and regular aroma. Unless this particular combination was already on the market, we would have our new product concept. Other highpotential concepts are also suggested in the figure: For example, a strong hazelnut coffee might not be a bad idea. FIGURE 7.1 Factor Utility Scores–Coffee Example Strength Flavors Utility value 10 10 5 Utility 5 value 0 0 None Hazelnut Caramel Walnut Hickory Aroma 10 Utility value 5 0 Very weak Suggestive Regular Aromatic Mild Medium Strong Explanation: The scales are a statistical "utility" value, from 0 to 10. The judgments are by consumers, in trade-off tests. They tell us that consumers very much prefer no flavor, medium strength, and regular amount of aroma. To get a new coffee, try making it of hazelnut flavor, strong, and aromatic. If that combo couldn't be done with the hazelnut flavor, what would the next best combination be? Chapter Seven Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques 173 Trade-off analysis was used by the Sunbeam Corporation when it wanted to expand its kitchen mixing appliances sales in various countries around the world. The company identified three types of attributes—silhouette, features, and benefits. The determinant attributes for each appliance were identified, and the range for each selected. For instance, silhouettes had about 10 combinations—low versus high, strong versus stylized, and so on. Cards representing new products that combined specific silhouettes, features, and benefits were prepared. Consumers in the various countries were asked to sort the cards by preference from top to bottom. If a person wanted a low, strong silhouette, a large number of variable speeds, a very quiet motor, and the ability to use on semiliquids, one card may have had the right silhouette, speed, and noise, but couldn’t be used on liquids. Another could be used on liquids and had the right silhouette and noise but had only three speeds. To choose one, the consumer would have to trade off speed variety against use on liquids. With hundreds of consumers doing this, a good picture for each attribute can be obtained and the optimization process begun. Though our first couple of examples have centered on consumer goods, keep in mind that trade-off techniques are very versatile and can be used in many different situations. In fact, because business buyers tend to make a more rational analysis of product features, trade-off analysis has become quite valuable in industrial product innovation. Applications include snowmobiles, health care systems, aircraft, lift trucks, hotel rooms, and computer software, as well as business services of all kinds.1 Conjoint analysis was also used by Marriott Corp. when designing and developing the Courtyard chain to build in the most desired needs and wants of both business and leisure customers. A Conjoint Analysis Application We begin by illustrating a full-profile conjoint analysis, that is, one for which we obtain information on all possible levels of all the product’s attributes. We will see alternatives to this method later. Suppose you are managing a line of prepared salsas for a food company. You are looking to add to your product line and need to generate some new concepts that can be evaluated and brought to product development. Based on your understanding of the market and recent consumer research, you have found that three attributes are uppermost in customers’ minds when they choose a brand of salsa: (1) spiciness (mild, medium-hot, or extra-hot), (2) color (green or red), and (3) thickness (regular, thick, or extra-thick). There are 3 3 2 3 3 5 18 different types of salsa that can be made by combining the levels of these attributes in all possible ways (a mild, green, thick salsa is one way). 1 For details on usage of the trade-off technique, see Dick R. Wittink and Philippe Cattin, “Commercial Use of Conjoint Analysis: An Update,” Journal of Marketing, July 1989, pp. 91–96; Dick R. Wittink, Marco Vriens, and Wim Burhenne, “Commercial Use of Conjoint Analysis in Europe: Results and Critical Reflections,” International Journal of Research in Marketing, January 1994, pp. 41–52; and Gary L. Lilien, Arvind Rangaswamy, and Timothy Matanovich, “The Age of Marketing Engineering,” Marketing Management, Spring 1998, pp. 48–50. 174 Part Two Concept Generation FIGURE 7.2 Preference Rankings of One Respondent Thickness Spiciness Color Regular Regular Regular Regular Regular Regular Thick Thick Thick Thick Thick Thick Extra-Thick Extra-Thick Extra-Thick Extra-Thick Mild Mild Medium-Hot Medium-Hot Extra-Hot Extra-Hot Mild Mild Medium-Hot Medium-Hot Extra-Hot Extra-Hot Mild Mild Medium-Hot Medium-Hot Red Green Red Green Red Green Red Green Red Green Red Green Red Green Red Green Actual Ranking* 4 3 10 6 15 16 2 1 8 5 13 11 7 9 14 12 Ranking as Estimated by Model 4 3 10 8 16 15 2 1 6 5 13 11 7 9 14 12 *1 = Most preferred, 18 = Least preferred. We begin by designing 18 cards, each with a picture and/or verbal description of one of the combinations.2 Each respondent customer is then asked to rank the cards from 1 to 18, where 1 is “like most” and 18 is “like least.”3 As this task may be challenging, we can suggest that the respondent make three piles of cards (“like,” “neutral” and “don’t like”). The six or so cards within each pile can more easily be sorted, then the piles can be combined and final adjustments to rank order can be made. The rankings provided by one respondent are given in Figure 7.2. (Figure 7.2 also shows rankings as estimated by the model, which can be ignored for now.) Suppose a given customer likes extra-hot salsa. There are several cards in the stack (six, to be exact) that depict extra-hot salsa, in combination with other attributes. He or she would tend to rank these cards favorably (that is, give them a low rank number—the lower the rank number, the more the concept is liked). If he or she really likes extra-hot, we might expect almost all of the extra-hot cards to be assigned low rank numbers—that is, a pattern would be evident in the rank orderings. If the customer couldn’t care less about whether the salsa was green or red, 2 Some research suggests that verbal representations were good for facilitating judgment, while pictorial representations are good for improving respondents’ understanding of design attributes. See Marco Vriens, Gerard H. Loosschilder, Edward Rosbergen, and Dick R. Wittink, “Verbal versus Realistic Pictorial Representations in Conjoint Analysis with Design Attributes,” Journal of Product Innovation Management, 15(5), September 1998, pp. 455–467. 3 In addition to ranking, other types of responses can also be gathered. For example, respondents can be presented with pairs of cards and asked to state which they prefer. The different techniques lead to similar results. See Gilbert A. Churchill, Jr., Marketing Research: Methodological Foundations, 6th ed. (Fort Worth, TX: Dryden, 1995). Chapter Seven Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques 175 FIGURE 7.3 Conjoint Analysis Output—Salsa Data (a) Graphs Spiciness Thickness Color 2 1 Utility 0 ⴚ1 ⴚ2 Regular 0.161 Thick 0.913 Ex-Thick ⴚ1.074 (b) Importance of Attributes 0 Spiciness Thickness Color 20 Mild 1.667 Medium-Hot 0.105 40 Ex-Hot Red ⴚ1.774 ⴚ0.161 60 80 Green 0.161 100% 59.8% 34.6% 5.6% we would expect the rank numbers assigned to red salsa to be not that different from those assigned to green—no patterns would emerge. Conjoint analysis uses monotone analysis of variance (MONANOVA), a data analysis technique, to find these patterns within the rank order data. That is to say, we identify the customer’s underlying value system: which attributes are important, and which levels of the important attributes are favored. To do this, we use the rank orderings to estimate the utilities (sometimes called part-worths) of each level of each attribute for each customer. The graphical conjoint analysis output for the data of Figure 7.2 is given in Figure 7.3(a). The graphs in Figure 7.3(a) provide a visual representation of the relative importances of the attributes. The largest range in utilities is found for spiciness—the utilities assigned to mild and extra-hot are 1.667 and 1.774, for a range of 3.441— thus, spiciness is this individual’s most important attribute influencing likes and dislikes in salsa (using the same logic, color is the least important). The graphical output also indicates which levels of each attribute are preferred. As shown, mild salsa is favored over medium-hot or extra-hot, other things being equal. This particular customer also prefers medium-thick salsa to both regular and extra-thick, and likes green salsa very slightly more than red salsa. Figure 7.3(b) expresses the relative importances of the three attributes as percentages.4 As can be seen, the relative importance of spiciness to this individual 4 These are estimated by looking at the ranges of utilities of the three attributes (that is, the gap between the highest and lowest utilities). As seen, the range for spiciness is 3.441. The ranges for thickness and color can be calculated as 1.987 and 0.322. Summing the three ranges yields a total of 5.750, and each range is divided by this amount to get its relative importance. For spiciness, 3.441/5.750 5 59.84 percent. 176 Part Two Concept Generation is almost 60 percent. Thickness is also relatively important (about 34.5 percent), while this respondent seemed to be almost indifferent to color. One should keep in mind that these results are very dependent on the levels actually selected for the conjoint task. The customer might not have been indifferent to color if the options were red, green, and shocking pink. How well does our model predict this customer’s choice patterns? We can check predictive accuracy by adding the utilities comprising each of the 18 choices to get estimates of overall preference. For example, the estimate we would obtain for the extra-thick, mild, red salsa would be 21.074 1 1.667 2 0.161 5 10.432. We can then rank order these preferences as predicted by the model and compare these to the rankings actually made by the respondent. As shown in the last column of Figure 7.2, the estimates predict the actual preference order almost perfectly for this customer and identify the most-favored and least-favored combinations easily. In a typical application, trade-off analysis allows the manager to identify which attributes are the most important overall and also which levels of these attributes are the most popular. Assume that a large sample of individuals performed the trade-off task, and that the rankings shown in Figure 7.2 were average rankings over the sample rather than one individual’s responses. In this case, we might concentrate on developing a medium-thick, green, mild salsa. (Note that these are the levels of each attribute with the highest utilities in Figure 7.3(a).) Clearly, there may be segments within the market. It may be that about half the market likes mild salsa and about half likes extra-hot. If we only examined the average, we might conclude that medium-hot is best, though in reality nobody may like it! Thus, the next analytical step is to identify benefit segments based on the utilities. This will be discussed later in Chapter 9. The ranking task was relatively easy in this simple case, as there were only 18 cards to rank order. What if you had to consider many more attributes and/or levels? For example, in addition to the three attributes given above, you might need to consider type of container (glass jar vs. plastic bucket), size of container (10 ounce vs. 16 ounce), type of ingredients (organic vs. not organic), and three different potential brand names. That’s 3 3 2 3 3 3 2 3 2 3 2 3 3 5 432 different cards—equivalent to eight decks of playing cards stacked together, including jokers! No respondent, however well intentioned, will have the patience for this task. Fortunately, the full set of cards need not be rank ordered. By using a fractional factorial design, we can still estimate the relative preferences of all possible products using only a small subset of the cards.5 With a reduced set of cards, most respondents won’t be able to find the exact combination they want. So they must choose the combination that most closely meets their desires by trading off attributes wanted more against those wanted less. We all do this when our favorite brand of something is not in the store and we have to find a close substitute. 5 See discussion in David R. Rink, “An Improved Preference Data Collection Method: Balanced Incomplete Block Designs,” Journal of the Academy of Marketing Science 15, Spring 1987, pp. 54–61; also Joel H. Steckel, Wayne S. DeSarbo, and Vijay Mahajan, “On the Creation of Acceptable Conjoint Analysis Experimental Designs,” Decision Sciences 22, Spring 1991, pp. 435–442. Chapter Seven Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques 177 Is Conjoint the Right Method?6 While conjoint analysis is a very widely used technique in new product research, it can be misused if one is not careful. First, one should be able to break down the product into discrete attributes (such as spiciness, thickness, and color in the salsa example). Conjoint analysis assumes that customers combine these features rationally when evaluating brands, and there are cases where this is not a realistic assumption. A young consumer may reveal in conjoint analysis that she prefers thick, mild, green salsa, but will actually choose an extra-thick, spicy brand because that’s the brand her mom used to buy. Or a car buyer might say he likes large trunk space and a high-powered engine, yet will buy a Smart Car because he thinks it’s cute or he will look cool driving it. A further problem encountered with complex products such as cars is that there are so many hundreds or even thousands of attributes that one might consider, but the average customer can only handle about 10 of these in a given conjoint study, at most. The researcher must be certain that all the important ones are included. For example, if the researcher did not include “ease of parking” in a conjoint analysis for cars, the popularity of the tiny Smart Car might have been underestimated, especially among city residents. Other issues are just not handled by conjoint analysis. Purchase occasion is not included (the young consumer may buy extra-thick, spicy red salsa for herself but thick, mild green salsa for her children or her guests); variety-seeking behavior is not considered (every once in a while, she will buy the medium-thick, mild red salsa just to try something different); and decisions made jointly may pose difficulties (the husband may care about legroom while the wife cares about trunk space, and it is unknown which one has the ultimate decision or how they resolve differences in their car purchase decision). Additional guidelines for conjoint analysis would include the following: 1. One should know what the determinant attributes are before doing the conjoint analysis. AR gap mapping or one of the qualitative techniques can be helpful in this regard. 2. Respondents should be familiar enough with the product category and the attributes to be able to provide meaningful data on preference or purchase likelihood. Conjoint may be less useful for new to-the-world products. 3. The firm should be able to act on the results; in other words, actually develop a product that delivers the combinations of attributes preferred in the conjoint analysis.7 Finally, we should reiterate that trade-off (conjoint) analysis is commonly used in concept evaluation, and we will pick up the discussion of this technique in Chapter 9. 6 Steve Gaskin, “Navigating the Conjoint Analysis Minefield,” Visions, 37(1), 2013, pp. 22–25. Adapted from Robert J. Dolan, Managing the New Product Development Process: Cases and Notes (Reading, MA: Addison-Wesley, 1993), p. 125. 7 178 Part Two Concept Generation Alternatives to Full-Profile Conjoint Analysis Sometimes the decision problem just has too many attributes, and it cannot be easily solved using full-profile conjoint analysis. Another weakness of the full-profile approach is that it does not measure interactions among attributes. As an example of an interaction effect, a customer may like mild salsa, and may like mediumthick salsa, but may really like mild, medium-thick salsa—even more than you might have imagined from the conjoint analysis! There are adaptations of the fullprofile conjoint analysis that can make up for these shortcomings.8 Adaptive conjoint analysis, developed by Sawtooth Software, shows only a few attributes at a time to the respondent and adapts to the respondent as the conjoint exercise goes on. In the adaptive technique, the respondent is first asked which attributes are most important and which levels are most liked or disliked, then pairs of options are shown to the respondent that focus only on the most important attributes and levels that are most liked or disliked. In a car design application, if the respondent says that the number of car doors and country of manufacture are highly important, one question might ask if he or she prefers a four-door car made in the United States or a two-door car made in Japan. After a series of questions, the respondent is presented with “calibration concepts”—combinations of several attributes to which he or she states likelihood of purchase, given as a number between 0 and 100. (Example: On a scale of 100 where 100 is “definitely buy,” how likely would you be to buy a red, two-door, rear-wheel-drive U.S.-made car that costs $16,000?) Another alternative sometimes used is choice-based conjoint analysis, in which the respondent is shown several alternative product choices and is asked which he or she would prefer (if none, then “none of the above” is a possible response). As an example, the respondent may be asked if he or she would prefer: • • • • A red, two-door, rear-wheel-drive U.S.-made car that costs $16,000. A blue, four-door, front-wheel-drive Japanese-made car that costs $18,000. A green, two-door, all-wheel-drive German-made car that costs $20,000. None of the above. Both of these procedures minimize the number of attributes and levels any one respondent has to be exposed to. Note that there are also other simple tradeoff techniques available for concept development, similar to the conjoint analyses described above in terms of required data but not requiring any specialized software.9 8 The techniques and examples are adapted from the Sawtooth Software Web site, www.sawtoothsoftware.com. Sawtooth Software is one of the leading providers of conjoint analysis software. 9 For a good, fully worked-out example, see Nelson Whipple, Thomas Adler, and Stephan McCurdy, “Applying Tradeoff Analysis to Get the Most from Customer Needs,” in A. Griffin and S. M. Somermeyer, The PDMA Toolbook 3 for New Product Development (New York: John Wiley, 2007), Chapter 3. Chapter Seven Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques 179 Recent Modifications in Conjoint Analysis Other techniques are sometimes used to handle large numbers of determinant attributes and levels. In a property insurance example, analysts had to restructure a traditional form of conjoint measurement called SIMALTO by adding cost and savings to each of the attribute trade-off utilities. They then gave consumers budgets to spend on their choices, and thus captured a lot of variables in a willingness to pay. They kept the power of the original set of trade-off attributes without having to use the data-losing method of conjoint calculations.10 Recent work has examined some of the practical difficulties and concerns encountered in concept testing. Since concept testing is done so early in the new product process, often before a prototype is available for customer trial, can the respondent conceptualize the product and its uses well enough? If not, does the method still produce valid results? A study of a simple product line extension (a baking-soda toothpaste) suggested that conjoint results obtained when customers were only exposed to the product concept were very similar to those obtained if customers were actually allowed to try the product.11 Thus, conjoint results are a valid early indicator of ultimate product success, at least for product line extensions. Of course, conjoint analysis and perceptual mapping, as well as product trial, will be rich sources of customer information later in the new products process. In the case of major innovations (such as a new computer or telecommunications technology), customers without a high level of expertise in the product category may be unable to assess the innovation’s benefits, and concept test results may not validly predict how well the actual product will be received. Some have advocated using only customers with at least moderate levels of expertise, even for minor innovations.12 Virtual Prototypes in Concept Testing Virtual prototypes can be used in concept testing as well. These are either static pictures of the prototypes, or video clips that simulate the product in action, which can be presented to respondents via the Internet. These virtual prototypes are, of course, far less costly to produce and test than actual physical prototypes, allowing the firm to test a wide range of concepts quickly and cheaply.13 Improvements in virtual reality computer and video technology are providing marketers with many exciting new ways to test concepts with customers. One new measurement method, called information acceleration (IA), has recently 10 Peter D. Morton and Crispian Tarrant, “A New Dimension to Financial Product Innovation Research,” Marketing and Research Today, August 1994, pp. 173–179. 11 John R. Dickinson and Carolyn P. Wilby, “Concept Testing With and Without Product Trial,” Journal of Product Innovation Management, 14(2), March 1997, pp. 117–125. 12 Jan P. L. Schoormans, Roland J. Ortt, and Cees J. P. M. de Bont, “Enhancing Concept Test Validity by Using Expert Consumers,” Journal of Product Innovation Management, 12(2), March 1995, pp. 153–162. 13 Ely Dahan and V. Srinivasan, “The Predictive Power of Internet-Based Product Concept Testing Using Visual Depiction and Animation,” Journal of Product Innovation Management, 17(2), March 2000, pp. 99–109. 180 Part Two Concept Generation been developed and was first applied by General Motors in testing new electric car concepts.14 The unique feature of IA is that respondents are brought into a virtual buying environment that simulates the information typically available in a realistic purchase situation. Through the use of a video monitor and laser videodisk player, the respondent can see ads, read car magazines, and hear statements from salespeople and word-of-mouth comments from customers. Using surrogate travel technology, customers can virtually walk around a dealer showroom and look at computer-generated car prototypes. While expensive (an application might cost $100,000 to $300,000), IA is a potentially valuable complement to existing concept testing methods. Simple concept pictures or descriptions may not provide enough information to customers to enable them to make a realistic purchase decision, especially in the case of a very complex product like a new electric automobile. IA also allows testing of many virtual variations of the same basic concept, so that preferences can be observed. As video technology improves, IA will become less expensive and further extensions to it will be made (for example, the respondent may be able to virtually drive the car).15 Qualitative Techniques We have seen several quantitative techniques that can be used to incorporate customer input into concept generation. As noted at the beginning of this chapter, however, these techniques have natural complements: namely, a collection of qualitative techniques, which we will now explore. It is tempting to be dazzled by the fancy outputs generated by MDS or factor analysis, especially if one is not that familiar with them. Managers, however, should resist this temptation and not take the results at face value. The qualitative techniques presented below are useful ways to challenge the assumptions (for example, about what attributes are really determinant) that underlie the sophisticated approaches and can very frequently bring the manager perspectives that may go overlooked otherwise. Although our discussion of these is briefer, they are by no means less important or useful in concept generation. Dimensional Analysis Dimensional analysis uses any and all features, not just measurements of dimensions (such as spatial—length, width, and so on). The task involves listing all of the physical features of a product type. Product concept creativity is triggered by the mere listing of every such feature, because we instinctively think about how that 14 For detailed information on information acceleration, see Glen L. Urban, Bruce D. Weinberg, and John R. Hauser, “Premarket Forecasting of Really New Products,” Journal of Marketing, January 1996, pp. 47–60. See also Phillip J. Rosenberger III and Leslie de Chernatony, “Virtual Reality Techniques in NPD Research,” Journal of the Market Research Society, October 1995, pp. 345–355. 15 Reportedly, Caterpillar lets its customers virtually test drive tractors under different driving conditions using a similar virtual reality technique. See Brian Silverman, “Get ’Em While They’re Hot,” Sales and Marketing Management, February 1997, pp. 47–52. Chapter Seven Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques 181 feature could be changed. Rarely is anything worthwhile found in dimensional analysis until the list is long. It takes a lot of work to push beyond the ordinary and to visualize dimensions that others don’t see. Some of the most interesting features are those that a product doesn’t seem to have. For example, a spoon may be described in terms of its aroma, sound, resilience, bendability, and so on. Granted, the aroma may be hard to detect, the sound (at the moment) may be zero, and the resilience may be only when pushed by a vise. But each feature offers something to change. How about spoons that play musical notes as children move them to the mouth? How about spoon handles that can be squeezed to play notes? How about spoons that smell like roses? Listing hundreds of features is not uncommon. Figure 7.4 shows a shorter list, but perhaps it suggests what must be done. Successful users claim that just citing FIGURE 7.4 Dimensional Attributes of a Flashlight Using dimensional analysis, here are 80 dimensions. There were almost 200 in the analyst’s original list. A change in any one of them may make a new flashlight. Overall unit: Weight Rust resistance Balance Gripability Shock resistance Shear force Heat tolerance Insulation material Automatic flasher Manual flasher Distance visible Length Hangability Stain resistance Cold tolerance Flexibility Insulation color Translucence Focus of beam Closure type Lining material Buoyancy Flammability Malleability Compressibility Reflectiveness Surface area/color Closure security Material of case Color Number of body seams Water resistance Diameter Washability Weight of metal Explosiveness Smell of unit Number of tags Snagability Sealant material Lens: Material Opacity Color Strength Texture Springs: Number Material Length Strength Style Switches: Number Pressure Noise Type Location Bulb: Number Shape Size Gas type Thread strength Length of stem Filament shape Thread size Filament material Shatter point Thread depth Amperage Batteries: Number Size Terminal type Direction Rechargeability Reflector: Depth Diameter Shape Durability Surface Color Temperature limit 182 Part Two Concept Generation a unique dimension sparks ideation, and that the technique has to be used to be believed. Checklists From early forms of dimensional analysis evolved one of today’s most widely used idea-generating techniques—the checklist. The most widely publicized checklist was given by the originator of brainstorming: Can it be adapted? Can something be substituted? Can it be modified? Can it be magnified? Can it be reversed? Can it be minified? Can it be combined with anything? Can it be rearranged in some way? These eight questions are powerful; they do lead to useful ideation. Business and industrial goods analysts use such features as source of energy, materials, ease of operation, subassemblies, and substitutable components. (See Figure 7.5 for an abbreviated list of such industrial checklist questions.) Checklists produce a multitude of potential new product concepts, most of them worthless. Much time and effort can be spent culling the list. Another more recent technique for generating new concepts systematically manipulates the existing product’s attributes in certain specified ways. One attribute could, for example, be made dependent on another: A child’s bath mat changes color if the water is too hot. An attribute might be removed, leading to an essentially different product: Removing the internal floppy drives on a PC results in an ultra-thin model. And so on. Four strategies for creatively generating new concepts in this way are presented in Figure 7.6. FIGURE 7.5 Checklist of Idea Stimulators for Industrial Products Can we change the physical, thermal, electrical, chemical, and mechanical properties of this material? Are there new electrical, electronic, optical, hydraulic, mechanical, or magnetic ways of doing this? Find new analogs for parallel problems. Is this function really necessary? Can we construct a new model of this? Can we change the form of power to make it work better? Can standard components be substituted? What if the order of the process was changed? How might it be made more compact? What if it was heat-treated, hardened, alloyed, cured, frozen, plated? Who else could use this operation or its output? Has every step been computerized as much as possible? Chapter Seven Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques 183 FIGURE 7.6 Templates for Creativity Goldenberg and Mazursky present several “Creativity Templates” that can be used to manipulate the existing knowledge base encoded in product attributes to discover innovative new products. Procedure: Begin by identifying the determinate attributes, then manipulate these according to the four creativity templates. The templates are: 1. Attribute Dependency Template: Find a functional dependency between two independent variable attributes. The interaction may suggest a creative new product. Example: the color of the ink on a coffee cup is dependent on the contents, and a warning message can be revealed if the beverage is too hot. 2. Replacement Template: Remove one of the components of the product and replace it with one from another environment. The function the removed component performed is done by another component. Example: the antenna on a Walkman is replaced by the headphone cord. 3. Displacement Template: Remove an intrinsic component and its function, in such a way as to functionally change the product. This may create a new product for a new market. For example: removing floppy and CD drives on laptop PCs resulted in the ultra-thin PCs. 4. Component Control Template: Identify and create a new connection between a component internal to the product and one that is external to the product. Examples: Toothpastes with added whiteners, or suntan lotions with added skin moisturizers. Source: From Jacob Goldenberg and David Mazursky, Creativity in Product Innovation, Cambridge University Press, 2002. Relationships Analysis Several of the concept-generating methods we have been looking at compare things: Perceptual maps compare attributes, and group creativity is stimulated by reasoning from a known to an unknown, for example. But the comparisons are incidental to a larger issue in those methods. We will now look at two analytical attribute techniques that go right to the point—forcing things together for examination. These two techniques are the two-dimensional matrix and the morphological matrix. Both are examples of types of relationships analysis, so named because they require the respondent to find relationships among dimensions to generate new product concepts. About the Dimensions Used in Relationships Analysis Recall that Figure 6.1 said attributes are features (such as length), functions (such as coating hair with protein), and benefits (such as economy and health). But other aspects of products are not always included as attributes in definitions—for example, different places of use, occupations of users, or other items the product is used with. Relationships analysis techniques use them too. We seek any and all dimensions that help, and there is no fixed set of these. It is hoped that the examples shown in this chapter will suggest the view you should take in creating the matrixes. Two-Dimensional Matrix The simplest format for studying relationships is seen in Figure 7.7, which shows two attribute sets for insurance. Only partial lists of two dimensions (event insured against and person/animal insured) are used, but just these two provide 50 cells 184 Part Two Concept Generation FIGURE 7.7 Two-Dimensional Matrix Used for New Insurance Products Person/Animal Insured Event Insured Against NewTroubled Rich borns Geniuses Kids Uncles Injury from fire Getting lost Normal death Being insulted Being kidnapped 1 11 21 31 41 2 12 22 32 42 3 13 23 33 43 4 14 24 34 44 Dogs/ Cats 5 15 25 35 45 Tropical Birds 6 16 26 36 46 Saltwater Fish 7 17 27 37 47 New JobNewly- New holders weds Parents 8 18 28 38 48 9 19 29 39 49 10 20 30 40 50 Examples of new product concepts: An insurance policy that protects new parents if their child gets lost (20), or that protects newlyweds from the risks of being kidnapped while on their honeymoon (49), or that protects geniuses from the damage of being insulted (32). Relationships analysis methods mostly produce nonsense, but like the others, the two-dimensional matrix often produces a surprise that, upon careful thought, makes sense. to consider. Notice that only by forcing relationships could we expect to come up with a special policy that protects new parents if they happen to misplace their new child, or that protects newlyweds from the costs of overcelebrating their honeymoon. In the case of the insurance example, to analyze the results we would just start with 1, think about it, then to 2, and so on. In contrast to most of the methods studied to this point, relationships analysis goes directly to a new product idea (e.g., aerosol ice cream). The number of twodimensional matrixes that can be prepared is almost unlimited. Keep looking at different ones until satisfied with the list of new possibilities found or convinced that the technique “just isn’t for me.” A slightly different kind of relationships analysis employs as its dimensions the product’s utility lever (how the product affects the customer’s life) and buyer’s experience cycle (at what point does the product affect the customer). Altering one or both of those dimensions can result in successful new product ideas. Figure 7.8 shows how several companies have come up with product ideas that ultimately were very successful by “stretching” on one or both dimensions. Morphological or Multidimensional Matrix The next method, morphological matrix, simultaneously combines more than two dimensions. The matrix can include many dimensions, and the technique originated many years ago when a scientist was trying to further development on what became the jet engine.16 16 The scientist used 11 parameters (dimensions), each of which had between two and four alternatives; that set yielded 36,864 combinations (possible engines). Incidentally, that matrix also yielded two combinations that became the German V-1 and V-2 rockets in World War II. See Fritz Zwicky, Discovery, Invention, Research: Through the Morphological Approach (New York: Macmillan, 1969). Chapter Seven Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques 185 FIGURE 7.8 Another Form of Dimensional Analysis W. C. Kim and R. Mauborgne note that firms can come up with winning new product ideas by considering two key dimensions: • • Utility lever: How the product will affect the customer’s life (such as simplicity, fun/image, environmental friendliness, reduced risk, convenience, and productivity). Buyer’s experience cycle: The stage when/where the product will affect the customer (purchase, delivery, use, supplements, maintenance, disposal). They provide examples of firms that altered one or both of these dimensions, resulting in great new product ideas: • • • Typical fast-food restaurants offer cheaper coffee, focusing on offering convenience or productivity to the customer at the purchase stage. Starbucks also aims at the purchase stage, but adds the utility lever of fun/image with its trendy coffee bars and exclusive blends. Computer makers offer the utility of productivity at the use stage. Dell’s innovation was to offer the productivity utility at the delivery stage as well by shipping direct. The Philips Alto disposable fluorescent light bulb offered a unique combination of enviornmental friendliness utility at the disposal stage. Note: In addition to the novel combination of utility and experience cycle stage, the firm must also set a strategic price to improve the likelihood of success. Source: Reprinted by permission of Harvard Business Review. Exhibit from “Knowing a Winning Business Idea When You See One,” by W. C. Kim and R. Kauborgne, September–October 2000. Copyright © 2000 by the Harvard Business School Publishing Corporation; All rights reserved. An example, shown in Figure 7.9, concerns the development of a new coffee maker. Five dimensions were identified, and for illustrative purposes, three alternatives for each dimension are shown. (In a real example, many more alternatives are generally found.) The new product manager’s task is to link combinations of those items. One common technique is to have a computer print out all possible combinations, which are then scanned for interesting sets. Other analysts just use a simple mechanical method of reading the rows across. The top row says: How about a new coffee pot, with an element right in the pot for FIGURE 7.9 A Morphological Matrix for a New Coffee Maker Dimension Heating Adding Coffee Filtering Coffee 1. Heating element in pot 1. By spoon 1. Filter paper 2. Open flame under pot 3. Microwave unit 2. With built-in measuring cap 3. Automatic feed 2. Porous ceramic filter 3. Centrifuge method Keeping Coffee Warm 1. Thermal (insulating) technology 2. Warming unit in pot 3. External heat source Pouring Coffee 1. Valve under pot 2. Pump in lid of pot 3. Espresso-like jets Source: Adapted from Stefan Kohn and Rene Niethammer, “Why Patent Data Can Be a Good Source of Comparative ‘Technology Intelligence’ in New Product Development,” Visions, January 2004. 186 Part Two Concept Generation heating the water, to which one adds ground coffee by spoon into a paper filter, which then keeps the coffee warm via a thermal insulator? A valve under the pot is turned on to pour the coffee. Didn’t like that one? There are many other combinations! After going through the rows, the analyst systematically exchanges one item in each row with one item in another, and so on. All analytical attribute techniques produce noise from which good ideas must be picked; but what at first appears to be noise may simply be a great new idea no one would have thought of easily without the matrix. In any event, the structure shown in Figure 7.9 should be followed. Creation of the columns was discussed at the beginning of this section of the chapter. The number of items in each column is either (1) the entire set, as in the survey above, or (2) a selection representing the full array. For example, a study of play wagons might have a column headed number of wheels, and the rows would be two, three, four, five, and six; but the height column might just have rows of 6 inches, 8 inches, and 12 inches (low, medium, and high). Analogy We can often get a better idea of something by looking at it through something else—an analogy. Analogy is so powerful and popular that it is used heavily as part of the problem-solving step in problem-based methods (Chapter 5). Just think of how many analogies are involved in computing terminology: cut-and-paste, recycle bin, browsing, surfing, briefcase, folder, and so many other terms are familiar to us in noncomputer contexts, and the use of these terms in computer settings is intuitively obvious to computer users. A good example of analogy was the study of airplane feeding systems by a manufacturer of kitchen furniture and other devices. Preparing, serving, and consuming meals in a plane is clearly analogous to doing so in the home, and the firm created several good ideas for new processes (and furniture) in the home kitchen. Amusement park designers watched cattle being herded and came up with the idea of queues for those waiting to go on popular rides! An analogy for bicycles might be driving a car—both incorporate steering, moving, slowing, curving, and so on. But the auto carries more passengers, has four wheels for stability, variable power, built-in communications, on-board service diagnosis and remedial action, and the like. Each difference suggests another new type of bicycle; some of these types are already available. The bicycle could also be compared to the airplane, to skating, to the submarine, to swimming, and at the extreme (for illustration) to a mouse in a maze. The secret, of course, is finding a usable analogous situation, which is often difficult. The analogy should meet four criteria: 1. 2. 3. 4. The analogy should be vivid and have a definite life of its own. It should be full of concrete images. It should be a happening—a process of change or activity. It should be a well-known activity and easy to visualize and describe. Chapter Seven Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques 187 Airplane feeding systems and driving a car qualify easily. And, perhaps surprisingly, an analogy of the machine gun ammunition belt helped seed company developers think of a roll of biodegradable tape studded with carefully placed seeds to be laid along a furrow. Analogy is used in several of the specialized techniques in Appendix B. Summary In this and the preceding chapter, we have presented a review of several analytical attribute techniques. Qualitative techniques included the very simple yet challenging dimensional analysis and more advanced methods such as the morphological matrix. Quantitative approaches included gap analysis and trade-off analysis. These can be used in complementary fashion: As seen above, the qualitative methods can be used prior to the more numbers-oriented models (to specify or double-check the attributes included in the analysis) or after the fact (to help interpret results). The essence of attribute analysis, in every case, is to force us to look at products differently—to bring out new perspectives. We normally have fixed ways of perceiving products, based on our sometimes long-term use of them, so forcing us out of those ruts is difficult. Anyone reading this in preparation for a specific ideation activity is encouraged to scan the list of over 40 other techniques in Appendix B. We are now finished with concept generation and hopefully have several good concepts ready for serious review and evaluation before undertaking costly technical development. We meet evaluation in Part III, Chapters 8–12, entitled Concept/ Project Evaluation. We will also find that several of the analytical techniques we encountered in these chapters will be of assistance in assessing customer preferences, specifying product design characteristics, and even beyond in the new product development process. Applications 1. “I guess I really like checklists best—they’re easy for me to understand and use. I’ve never seen this one by Small that you mentioned—wow, four pages of ways. Is all that really necessary? Couldn’t just as good a job be done with, say, one page? And incidentally, I must confess I’m slightly confused by the terminology. Tell me, what is the difference again between the checklists I like and what you call dimensional analysis?” 2. “As you can probably tell by now, I am an engineer by training and have always enjoyed playing around with one form of attribute analysis. We call it attribute extension, where we forecast the future changes in any important attribute of a product. You know, like the amount of random access memory in a PC. I recently asked our cable TV division to take five dimensions of a cable TV service and extend each out as far as they can see it going and tell me what ideas they get from it. I mentioned number of channels and types of payment as examples. Could you do something like this for me now . . . that is, take five dimensions of 188 Part Two Concept Generation cable TV service and extend them? It would help me get ready for their presentation Thursday.” 3. “Several of our divisions work in the women’s clothing markets. As you know, they are all specialized these days, this segment or that segment. It’s getting hard to come up with a new segment, one that has some size and would be responsive. So, when you were talking about morphological matrix, which I liked, I thought about women’s attire. One way to innovate would be to come up with new settings, or occasions, situations where we could devise a whole outfit. Sort of like wedding, or racetrack, or picnic, though we know of them and have clothing for them, of course. Sort of a package of apparel and accessories. But there must be many we don’t think of now. Would that morphological matrix method work on that?” Case: Rubbermaid17 Newell Rubbermaid is a global company that manufactures and sells a wide range of brands worldwide. Its divisions include: Tools (Lenox, Hilmor), Writing (Sharpie, Waterman, Paper Mate), Baby and Parenting (Graco), Home Solutions (Rubbermaid, Calphalon), Specialty (Mimio, Bulldog Hardware), and Commercial Products sold under the Rubbermaid name. Rubbermaid had been a successful product innovating company for years before its purchase by Newell in 1999, and as many as 200 new products per year are launched under the Rubbermaid name. The success of the Rubbermaid division is based partly on creating and producing high-quality, functional plastic products for anywhere in the house: kitchen, garage, laundry room, and bathroom, as well as closet organizers, car organizers, trash bins, and similar products. In recent years items have ranged from lunch boxes with snack compartments to stacking cereal containers, storage trunks and benches, power scrubbers, and many more. Category brands include TakeAlongs®, Lunch Blox™, Closet Helper™, and others. The firm makes almost a half-million different items, boasts a 90 percent success rate on new products, and obtains at least 30 percent of its sales each year from products less than five years old. The firm’s new product strategy is to meet the needs of the consumer. The new product rate is high, and diversification is desired. It is market-driven, not technology-driven, although in recent years the firm has identified such technologies as recycling new plastic parts from old tires for which it is seeking market opportunities. This practice of seeking opportunities for specific technologies will increase as a fallout of the firm’s current use of simultaneous product development. For idea generation, Rubbermaid depends on finding customer problems that can be built into the strategic planning process. Problems are sought in several ways, the principal one of which is focus groups. It also uses comments and complaints from customers, an example of which came when then-CEO Stanley C. Gault heard a Manhattan doorman complaining as he swept dirt into a Rubbermaid 17 This case was prepared from rubbermaid.com and many public information sources. Chapter Seven Analytical Attribute Approaches: Trade-Off Analysis and Qualitative Techniques 189 dustpan. Inquiry determined that the doorman wanted a thinner lip on the pan, so less dirt would remain on the walk. He got it. Each complaint is documented by marketing people, and executives are encouraged to read the complaints. One complaint by customers in small households, who found the traditional rack-and-mat too bulky to store, led to a compact, one-piece dish drainer. The firm generally finds its problems by using problem analysis in focus groups and solves them internally. They occasionally use scenario analysis to spot a problem. But scenario analysis is much less useful than problem analysis because the lead times are so short; their new product cycles make them concentrate mainly on already existing problems. The organization is kept conducive to newly created ideas by promoting cross-functional association among workers. Problem-find-solve is encouraged at all levels. Some other new items have been: Bouncer drinkware was created for people who fear using glassware around their swimming pools. A lazy susan condiment tray and other patio furniture products came from studies of lifestyle changes. People working at home told of problems that led to a line of home office accessories, including an “auto-office,” a portable device that straps onto a car seat and holds pens and other office articles. Generally speaking, Rubbermaid does not make much use of attribute listing and other fortuitous scan methods of ideation, including the various mapping approaches. It does find that product life-cycle models can be useful, and it closely tracks competitive new product introductions. Rubbermaid is, however, always looking for new ways by which it can come up with good new product concepts. They know from experience, for example, that there will be new ways by which problem-find-solve techniques can be used. And perhaps the fortuitous scan methods can be of greater use than now perceived. What would you recommend to Rubbermaid management? Should they use any of the concept generation techniques discussed in this chapter, in addition to the methods they traditionally use? Which ones, and why? 190 Part One Overview and Opportunity Identification/Selection FIGURE III.1 Concept/Project Evaluation From Figure II.1 (Concept Generation) Entry screen Spell out concept statement Fit with strategy confirmed Check technical feasibility Check marketing feasibility Full screen Customer screen Scoring model process Prepare concept boards and prototypes Define criteria, hurdles Detail concept test plan Implement concept test plan Iteration, conclusion Approved Project Technical screen Final technical assessment on latest version of the concept Concept statement Written protocol Top management support reconfirmed PIC strategy may be revised Pro-forma financials Team selected Budget provided Tentative development plan To Figure IV.1 (Development) P A R T T H R E E Concept/Project Evaluation Part II completed our study of the various methods of generating new product concepts. The next task is to undertake evaluation of these concepts. Evaluation takes place at many different times and in different ways, by different people, for different reasons. Therefore, a system of evaluations is needed, an idea that will be explained in Chapter 8. Then, beginning in Chapter 9, we will look at the different phases in that system. (See Figure III.1.) Concept testing is the first major tool and will be discussed there. Chapter 10 covers the activity generally called a full screen, a step in which the concept is judged by how well it fits the company and its marketing strengths. Once the project has cleared the high hurdles set at the full screen, it is approved for development and ready to move into the next phase of the process. Chapters 11 and 12 focus on specific topics appearing in the last box in Figure III.1: a financial analysis, a check to make sure the project (still) fits with the product innovation charter, and the development of a protocol. At that point, development can begin, teams are assigned (if they do not exist yet), and we move to Part IV of the book. The evaluation tools discussed in Part III are those that precede development. Once prototypes or service configurations begin to appear, evaluation begins again, first in the form of product use testing and later in market testing, and more. These are covered in later chapters. All of these efforts at evaluation are themselves major topics, so our discussions must be selective. Unfortunately, industry uses many of the tools in different ways, so they tend to blend together at the edges. When, for example, does a prototype concept test become a product use test? 192 Part Three One Overview Concept/Project and Opportunity EvaluationIdentification/Selection Likewise, industry often combines two or even three of the tools. For example, in some industries it is very easy to prepare prototypes, so some of these firms like to do an early customer survey that is partly market analysis, partly concept test, and partly prototype test, particularly when the idea first emerged in prototype form. Finally, standardized and fully accepted terminology often doesn’t exist. Therefore, we have had to do some standardizing of terms, and some of our decisions won’t be acceptable to all people. C H A P T E R E I G H T The Concept Evaluation System Setting Before looking into the various specific techniques used to evaluate new product concepts, we need something to give us an overview. Throughout the new product development process, we are doing evaluations, and there are evaluation techniques appropriate to each of the phases in the basic new product process. Furthermore, none of these techniques is used all the time or in all cases. Chapter 8 offers this overview, presenting models such as the cumulative expenditures curve and the A-T-A-R model as ways that can help us decide which evaluation techniques to use. We will cover potholes and surrogates, among other ideas. In Chapters 9 and 10, we will look more closely at concept evaluation and full screen techniques that are specifically appropriate to Phase III (concept/project evaluation), while Chapters 11 and 12 round out the discussion with a look at sales forecasting, financial and strategic analysis, and product protocol specification. You will recall from Chapter 2 that new products fail because (1) there was no basic need for the item, as seen by intended users; (2) the new product did not meet its need, considering all disadvantages; and (3) the new product idea was not properly communicated (marketed) to the intended user. In sum, they didn’t need it, it didn’t work, they didn’t get the message. Keep these factors in mind as you see how an evaluation system is constructed. What’s Going On in the New Products Process? New products build up the way rivers do. Great rivers are systems with tributaries that have tributaries. Goods that appear complex are just collections of metal shapes, packaging material, fluids, prices, and so on. A good analogy is the production of automobiles, with a main assembly line supported by scores of subsidiary assembly lines scattered around the world, each of which makes a part that goes into another part that ultimately goes onto a car in that final assembly line. If you can imagine the quality control people in auto parts plants evaluating each part before releasing it to the next step, you have the idea of a new product evaluation system. The new product appears first as an idea, a concept in words 194 Part Three Concept/Project Evaluation or pictures, and we evaluate that first. As workers turn the concept into a formed piece of metal, or software, or a new factory site preparation service, that good or service is then evaluated. When a market planner puts together a marketing plan, its parts are evaluated separately (just as minor car parts are) and then evaluated again in total after it is added to the product. The fact that we evaluate the product and its marketing plan as separate and divisible pieces is what lets us telescope the development process into shorter periods of time. There was an era when we went through a new product’s development step by step, nothing ahead of its time. But today we may be working on a package before we actually have finished the product, we may be filming part of a commercial before the trademark has been approved and finalized, or we may be preparing the trailer to promote an upcoming movie long before the final edits are completed.1 This sometimes causes some backtracking, but the cost of that is less than the costs of a delayed introduction. It does require, however, that we have thought through carefully the item’s overall development needs and decided which of those needs are crucial, and which are not crucial. Any evaluation system must cover the crucial ones. The Evaluation System for the Basic New Products Process Although the overall purpose of evaluation is to guide us to profitable new products, each individual evaluation step task has a specific purpose, keyed primarily to what happens next. Recall Figure 2.2, which showed that different evaluation tasks were appropriate to specific phases in the new products process. Figure 8.1 presents the same information, but adds the most common evaluation techniques used throughout the process. Before going any farther, this is a good place to note that conducting the evaluation tasks really does improve new product performance. The CPAS study (first introduced in Chapter 1) included an analysis of the most commonly used evaluation techniques: In all cases, the “Best” firms were significantly more likely to use these techniques than the “Rest,” and they ended up with better sales and profit results from their new products.2 In the process of Figure 8.1, ideas become concepts; concepts get refined, evaluated, and approved; development projects are initiated; and products are launched. Throughout this process, different questions need to be asked, and different evaluation techniques provide the required answers. For example, the very first evaluation precedes the product concept—in fact, it takes place in Phase I, when an opportunity or threat is identified and assessed. Someone decided the 1 There is a (perhaps apocryphal) story about an Alberto Culver shampoo. Television commercials were finished and at the networks ready for showing before the chemist could find an appropriate formulation! 2 Gloria Barczak, Abbie Griffin, and Kenneth B. Kahn, “Perspective: Trends and Drivers of Success in NPD Practices: Results of the 2003 PDMA Best Practices Study,” Journal of Product Innovation Management, 26(1), January 2009, pp. 3–23; and Stephen K. Markham and Hyunjung Lee, “Product Development and Management Association’s 2012 Comparative Performance Assessment Study,” Journal of Product Innovation Management, 30(3), 2013, pp. 408–429. Chapter Eight The Concept Evaluation System 195 FIGURE 8.1 The Evaluation System, Including Common Techniques New Product Process Phase Evaluation Task Opportunity identification and selection Direction Evaluation Techniques Opportunity identification Market descriptions Where should we look? Concept generation Initial review Concept/project evaluation Is the idea worth screening? The product innovation charter Immediate judgmental responses Preliminary market analyses Concept testing Full screen Should we try to develop it? Development Progress reports Have we developed it? If not, should we continue to try? Launch Checklists Profile sheet Scoring models Market testing Should we market it? If so, how? Protocol checks Prototype tests Concept tests Product use tests Speculative sale Simulated test market Informal selling Controlled sale Test marketing Rollout Economic summary In retrospect Caution: Keep in mind that the activities are never in such a neat sequence of steps as implied by this diagram. firm had a strong technology, or an excellent market opportunity, or a serious competitive threat—whatever. As discussed in Chapter 3 on strategy, a judgment was made that if the firm tried to develop a new product in a given area, it would probably succeed. This early evaluation step (direction) is shown at the top of Figure 8.1. Where should we look, what should we try to exploit, what should we fight against? The tool is opportunity identification and evaluation, also discussed in Chapter 3. This tool keeps us out of developments where we stand a poor chance of winning; in other words, it makes sure we play the game on our home field. This direction is provided in the product innovation charter. 196 Part Three Concept/Project Evaluation Now continue down Figure 8.1 to see how the evaluation tasks change as we progress through the basic new product process. In Phase II (concept generation), ideas begin to appear, and the purpose of evaluation changes: Now the goal is to avoid the big loser or the sure loser. We want to cull them out and spend no added time and money on them. We’re sometimes wrong, of course, but usually we’re right, and this step is essential if we are to focus limited resources on the worthwhile concepts and not get overwhelmed with the sheer number of available ideas. We will also try to spot potential big winners. Most good concepts are just that: good, nothing more. But a few are great, and we want to recognize them as soon as possible. These get added effort, usually in the form of a very complete concept testing and development program. While there is no one set screening procedure, certainly at this early stage we would be assessing the ideas in terms of the following criteria: • • • • • • Uniqueness: Is the idea original? Can it be easily copied by competitors? Need fulfillment: Does it meet a customer need? Feasibility: Do we have the capability to develop and launch it? Impact: How will our firm or organization be affected? Scalability: Can we become more efficient in production as volume increases? Strategic fit: Is there a good match with corporate strategy and culture?3 Many firms use some variant of this idea screening technique. For example, Unilever requires that a brief be written for each new idea under consideration. The brief must describe: the consumer need, any technical specifications, the “idea solution” (benchmarks and standards), “must-haves” (minimum requirements or specifications), “killers” (an assessment of what might cause it to fail), a statement of what is already known, a budget, and a timeline. A British consultant group, Oakland, recommends the 3A’s Matrix® as a way to model market opportunities against their potential value. The criteria they recommend are: • Applicability: does the opportunity meet strategic requirements? • Availability: can it be commercialized quickly? • Affordability: what is the associated cost?4 This kind of activity leads us to Phase III (concept/project evaluation) and the decision on whether to send the concept into full-scale development. This decision, if the amounts to be spent make it an important decision, will benefit from a very thorough scoring model application. Should we try to develop it? The decision to enter Phase IV (development) introduces the part of the process where the parallel or simultaneous technical and marketing activities are done (as seen in Figure 2.1). All through this phase we are continually asking, Have we got what we want? Is this part ready? Is that system subset cleared for use? Does the 3 Jacquelin Cooper, “How Industry Leaders Find, Evaluate, and Choose the Most Promising Open Innovation Opportunities,” Visions, 36(1), 2012, pp. 20–23. 4 The Unilever and Oakland examples are from J. Cooper, Visions, op. cit. Chapter Eight The Concept Evaluation System 197 software not only work, but produce what the customer needs? A protocol check tells us whether we are ready to develop a product for serious field testing. Development is naturally iterative: One new discovery leads to another; directions are changed; specific attempts fail, and we have to back up. At Hollingsworth & Vose, an industrial specialty paper company, gaskets are tested five times in this stage—in-house lab test, customer lab test, customer engine test, car manufacturer engine test, and fleet test. Sooner or later the technical efforts yield a product that evaluators say meets the customers’ request. We then enter Phase V (launch), and attention turns to launching the item. The evaluation issue now is whether the firm has proven itself able to make and market the item on a commercial scale. This is usually resolved by some form of market testing. Later on, of course, the developers (and others in the firm too, unfortunately) will be asking, in retrospect, should we have done all this? The purpose is not to find a guilty party for a product that bombed but, rather, to study the evaluation process to prevent a repetition.5 Product Line Considerations in Concept Evaluation Keep in mind that any one product being evaluated is not alone. Most organizations have several products under development, sometimes scores or even hundreds of them. As we saw in Chapter 3, managers often think in terms of a portfolio of products and evaluate new product projects in terms of how well they would fit with corporate strategy. We will see how product projects are selected relative to strategic concerns (such as strategic portfolios) in more detail in Chapter 11. Especially at the early phases, or fuzzy front end, of the new product process, there are risks involved in making project selection decisions. Depending on the evaluation mechanism chosen, the firm may let through too many bad ideas or reject good ideas. There is no one right way to optimize project evaluation, but some experience can help set the best rules for a given firm or industry. For example, a firm that needs new product help fast may skip early checkpoints and narrow down to just one or two alternative formats during development. They will tend to put in one major check late in the process to make sure the marketing plan communicates and the distribution system is in place. In an industry like pharmaceuticals, a firm might bring two or more ideas through to development: With more potential products, there is a greater chance that one will be a winner, and the payoff for winning is large enough to offset the extra costs incurred in developing more products. Having hurdles that are too high may reduce failure rate, but contribute to major, costly delays in new product launch. If a firm makes products with very short cycle times (such as computer games), it has to control the number of products in the process queue at any given time so that products receive development funds in a timely manner.6 5 For a thorough discussion of new product evaluation techniques and their use at different phases in the new products process, see Muammer Ozer, “A Survey of New Product Evaluation Models,” Journal of Product Innovation Management, 16(1), January 1999, pp. 77–94. 6 For more on this topic, see Donald G. Reinertsen, “Taking the Fuzziness Out of the Fuzzy Front End,” Research-Technology Management, November–December 1999, pp. 25–31. 198 Part Three Concept/Project Evaluation The Cumulative Expenditures Curve As we have seen, the new product evaluation system flows with the development of the product. What evaluation occurs at any one point (how serious, how costly) depends greatly on what happens next. Figure 8.2 shows a key input to the design of any evaluation system: In the middle of that figure, a gradually upward-sloping curve represents the accumulation of costs or expenditures on a typical new product project from its beginning to its full launch. This generalized curve, taken from various studies over the years, is just an average. It need not reflect any one firm, but it is typical of many durable consumer goods, nontechnical business-to-business products, and many services. Shown with the average curve are two others. The early expenditures curve is representative of product development in technical fields, such as pharmaceuticals, optics, and computers. R&D is the big part of the cost package, and marketing costs are relatively small. The lower curve in the figure shows the opposite type of firm, say, a consumer packaged goods company. Here the technical expenditures may be small, but a huge TV advertising program is needed at introduction. These are generalizations, and individual exceptions do occur, such as when P&G spends years developing a fat substitute called Olestra or Upjohn markets a line of generic drugs. The point is, whoever develops a concept evaluation system needs to know what situation it is for. No evaluation decision is independent of considerations on what will be done next, how much will be spent, or what points FIGURE 8.2 Cumulative Expenditures— All-Industry Average Compared to Occasional Patterns 100 90 80 Early expenditures 70 60 Percent of expenditures 50 (cumulative) 40 Average 30 20 Late expenditures 10 0 10 20 30 40 50 60 70 Percent of time spent 80 90 100 Launch Chapter Eight The Concept Evaluation System 199 FIGURE 8.3 Matrix of Risk/Payoff at Each Evaluation Decision is to: A Stop the project now B Continue to next evaluation A. It would fail AA (no error) BA (go error) B. It would succeed AB (drop error) BB (no error) If the product were marketed Comment: Cells AA and BB are “correct” decisions. Cells BA and AB are errors, but they have different cost and probability dimensions. of no return are passing. An old Chinese proverb says, “Spend your energy sharpening the edge of the knife, not polishing the blade.” The Risk/Payoff Matrix Figure 8.3 applies these ideas in a risk/payoff matrix. At any single evaluation point in the new product process, the new products manager faces the four situations shown. Given that the product concept being evaluated has two broad ultimate outcomes (success or failure) and that there are two decision options at the time (move on or kill the project), there are four cells in the matrix. The AA cell and the BB cell are fine; we drop a concept that would ultimately fail, or we continue on a concept that would ultimately succeed. The managerial problem arises in the other two cells. AB is a “drop error”: A winner is discarded. But BA is a “go error”: A loser is continued to the next evaluation point. Which error does the manager most want to avoid? The answer depends on the dollars. First, throwing out a winner is very costly, because the ultimate profits from a winning product are bound to be much greater than all of the development costs combined, let alone those in just the next step. So error AB is much worse than BA. The exception, of course, is the opportunity cost. What other project is standing by waiting for funding? When good candidates wait in the wings, the losses of dropping a winner are much less because the money diverted will likely go to another winner. The point is, a manager must think of these matters when deciding what evaluation to do. If the net costs of the next step in any situation are low, then a decision will probably be made to go ahead, perhaps with very little information. For example, P&G supported both Febreze (an odor eliminator) and Dryel (which lets you wash dry-clean-only clothes at home) with substantial market testing, including lengthy test marketing, as they were seen as risky, newto-the-world (and also new-to-P&G) products. But P&G would support a simple detergent line extension with much less extensive testing (relying on some of the alternative methods we will see in Chapter 18), as that would be considered a much less risky launch. You will read later about how Starbucks extensively tested Via instant coffee, because they saw it as a potentially risky product launch and 200 Part Three Concept/Project Evaluation needed to mitigate these risks. Consider: who knows more about detergents than P&G? Who knows more about coffee than Starbucks? Yet even these companies recognized that high-risk situations called for more extensive market testing. In general, the new products team should consider four generic risk strategies: Avoidance: Eliminate the risky product project altogether, though an opportunity cost is incurred (what if they had pushed through with the project and it succeeded?). Mitigation: Reduce the risk to an acceptable, threshold level, perhaps through redesigning the product to include more backup systems or increasing product reliability. Transfer: Move the responsibility to another organization, in the form of a joint venture or subcontractor, for example. The other party would be better equipped to handle the risk. Acceptance: Develop a contingency plan now (active acceptance) or deal with the risks as they come up (passive acceptance).7 The Decay Curve The risk matrix decisions lead to the idea of a decay curve, as shown in Figure 8.4. That figure depicts the percentage of any firm’s new product concepts that survive through the development period, from the 100 percent starting out FIGURE 8.4 Mortality of New Product Ideas—the Decay Curve Curve A — Slow decay, carry ideas along Curve B — Average decay Curve C — Rapid decay, avoid development cost Number of ideas A B C Time Concept Generation Development Launch Source: Hypothetical representation based on empirical data in various sources, including New Products Management for the 1980s (Chicago: Booz Allen & Hamilton, 1982), p. 14. 7 Gregory D. Githens, “How to Assess and Manage Risk in NPD Programs: A Team-Based Risk Approach,” in P. Belliveau, A. Griffin, and S. Somermeyer (eds.), The PDMA Toolbook for New Product Development (New York: John Wiley, 2002), pp. 187–214. Chapter Eight The Concept Evaluation System 201 before concept testing to the 2 percent (estimated from various studies) going to market. The discarded 98 percent drop off at various times during the process, and when they drop off is primarily determined by the analysis of the risk matrix. Decay curve C is roughly the shape of one decay curve from a leading company in the paper industry that wanted to kill off all possible losers early and spend time developing only those proposals worthy of marketing. This was their strategy, and their evaluation system implemented it faithfully. Decay curve A represents one for a service firm that had very low development costs and wanted to drop a project only when there was solid evidence against it. The paper company spent time making careful financial analyses even before technical work began; the service firm started up a project and just let it keep going until the contrary evidence built up. Thus, the decay curve is partly a plan and partly a result. The two should be synchronized. Its value as a managerial concept lies in helping the manager see the need for thinking through the stream of development costs and the risk/payoff matrix (above) for each new product concept as it starts its journey through development. When it is working, you will hear statements such as, “On that chip, let’s make sure the customer will want it if we can make it; no sense in spending all that money only to find there’s no buyer for it.” And, in the building next door, “Don’t worry about Ed’s doubts at this time; we can reposition the fertilizer spreader at the last minute if we have to, even change several of the key attributes if we want. Let’s just get going, now!” Planning the Evaluation System The previous considerations help set the tone for management decisions on an appropriate evaluation system for any particular new product concept. There are four other relevant, but less demanding concepts that help us decide whether to concept test, how long to run a field use test, whether to roll out or go national immediately, and how thorough a financial analysis to demand. Everything Is Tentative It’s easy to imagine that building a new product is like building a house—first the foundation, then the frame, then the first floor, and so on. Unfortunately, product aspects are rarely locked in that way. Occasionally they are, as when a technical process dominates development, or when a semifinished product is acquired from someone else, or when legal or industry requirements exist. We usually assume everything is tentative, even up through marketing. Form can usually be changed, and so can costs, packaging, positioning, and service contracts; so can the marketing date and the reactions of government regulators. So can customer attitudes, as companies with long development times have discovered. This means two long-held beliefs in new product work are actually untrue. One is that everything should be keyed to a single Go/No Go decision. Granted, one 202 Part Three Concept/Project Evaluation decision can be critical—at times, for example, when a firm must invest millions of dollars in one large facility or when a firm acquires a license that commits it to major financial outlays. But many firms are finding ways to avoid such commitments by transferring risk: by having another supplier produce the product for a while before a facilities commitment, or by negotiating a tentative license, or by asking probable customers to join a consortium to ensure the volume needed to build the facility. The other fallacy is that financial analysis should be done as early as possible to avoid wasting money on poor projects. This philosophy leads firms to make complex financial analyses shortly after early concept testing, although the numbers are inadequate. The paper products firm whose decay rate was presented in Figure 8.4 (curve C) rejected hundreds of ideas before realizing that early financial analysis was killing off ideas that would have looked great after further development. The financial analysis is best built up piece by piece, just like the product itself. We will see later how this works. Still another tentative matter is the marketing date. Marketing actually begins very early in the development process (for example, when purchasing agents are asked in a concept test whether they think their firm would be interested in a new item). Rollouts (discussed in Chapter 18) are now so common it is hard to tell when all-out marketing begins. No one pulls a switch and marketing instantly begins. Rather, marketing activity has been rolling forward gradually, picking up steam, which clearly affects the evaluation system. What results in some cases is a sort of rolling evaluation. The project is being assessed continuously, figures are penciled in, premature closure is avoided, and participants avoid mindsets of good and bad. This is, in a way, dealing with risk via acceptance or mitigation. We know product development projects are risky, so we evaluate, move to the next phase if warranted, and continuously upgrade the quality of information available to us throughout the process to minimize the chances of failure (mitigation) and to expect contingencies and deal with them as they come up (acceptance). Potholes One critical skill of product developers is the ability to anticipate major difficulties, the potholes of product innovation. In automobile travel, potholes are always a problem, but they only become costly when we fail to see them coming in time to slow down or steer around them. The same thinking applies to new products: We should carefully scan for the really damaging problems (the deep holes) and keep them in mind when we decide what evaluating we will do. If the pothole is deep enough, the development team may have to seriously consider the risk avoidance option: Drop the project! For example, when Campbell Soup Company undertakes the development of a new canned soup, odds are in its favor. The company knows development, manufacturing, packaging, distribution, and promotion of such products extremely well. But experience has shown two points in the process when it may fail, and if it Chapter Eight The Concept Evaluation System 203 does, the product won’t sell. The first is manufacturing cost—not quality, as that’s one of the company’s key strengths. But there is always a question about whether the chosen ingredients can be put together to meet market-driven cost targets. The second is whether consumers think it tastes good. So the company’s evaluation system is set never to overlook these two points. A flour miller once said his biggest pothole was a quick entry by a price-cutter because that industry had virtually no patent protection or other barriers to competitive entry. He planned on it in every case and didn’t go ahead without an answer. A software developer said his biggest pothole was customer unwillingness to take the time to learn to use complex new products. He had several worthwhile products in the graveyard to prove it. Among the potholes faced by pharmaceutical manufacturers is the uncertainty regarding FDA approval: For that reason a firm may go to the expense of taking two similar products through the approval process in the hopes that at least one of them makes it. In fact, if a manager thinks through the matter of potholes carefully (scans the road ahead), there are more benefits than just to the evaluation system. The People Dimension Product developers also have to remember they are dealing with people, and people cause problems. For example, although R&D workers are quite enthusiastic early in the life of a new product, the idea may have little support outside of R&D; it is fragile and easy to kill. Late in the development cycle, more people have bought in on the concept and are supportive because they have played a role in getting it to where it is. Consequently, the now-strong proposal is tough to stop. This means that an evaluation system should contain early testing that is supportive. In fact, concept testing is sometimes called concept development to reinforce the idea of helping the item, not just killing it off. Later in the cycle, hurdles should usually be tough and demanding, not easily waved aside. One firm designated its market research director as a manager of screens. His task was to impose absolute screens, such as, “A new food product, in home placement testing, must achieve a 70 percent preference against the respective category leaders.” If less than 70 percent of the testers preferred the new item, it was stopped, period. This sounds severe and arbitrary, but it shows how difficult it sometimes is to kill off marginal products late in development. Another people problem relates to personal risk. All new product work has a strong element of risk—risk to jobs, promotions, bonuses, and so on. Consequently, some people shy away from new product assignments. We’re always under the gun from someone—an ambitious boss, a dedicated regulator, an aggressive competitor, a power-hungry distributor, or an early critic who was overruled within the company. A good evaluation system, built on a thorough understanding of the road the new item will follow as it winds its way through development, protects developers from these pressures. The system should be supportive of people and offer the reassurance (if warranted) that players need. 204 Part Three Concept/Project Evaluation Surrogates The timing of factual information does not often match our need for it. For example, we want to know customer reactions early on, even before we develop the product, if possible. But we can’t really know their reactions until we make some of the product and give it to them to try out. So, we look for surrogate questions to give us pieces of information that can substitute for what we want to learn but can’t. Here are four questions to which we badly need answers and four other questions that can be answered earlier (thus giving clues to the real answer): Real Question Surrogate (Substitute) Question Will they prefer it? Will cost be competitive? Will competition leap in? Will it sell? Did they keep the prototype product we gave them? Does it match our manufacturing skills? What did they do last time? Did it do well in field testing? Note that each response has little value except to help answer a critical question that cannot be answered directly. Surrogates often change at different times in the evaluation process. For example, let’s go back to one of the questions just above: Will cost be competitive? At different times during the project, the surrogate used might be: Time 1: Does it match our skills? Time 2: Are the skills obtainable? Time 3: What troubles are we having in making a prototype? Time 4: How does the prototype look? Time 5: Does the manufacturing process look efficient? Time 6: How did the early production costs turn out? Time 7: Do we now see any ways we can cut the cost? Time 8: What is the cost? Time 9: What is the competitive cost? Only when we know our final cost and the competition’s cost can we answer the original question. But the surrogates helped tell us whether we were headed for trouble. The last tool that we use for designing an evaluation system for each new project as it comes along is based on how we forecast sales and profit on a new item. The calculation is much like a pro forma income statement, an array of figures allowing us to see what the profits will look like based on where we are at any one time in the development. The basic formula, shown in Figure 8.5, is based on what is known in the marketing field as the A-T-A-R concept (awareness-trial-availability-repeat). Chapter Eight The Concept Evaluation System 205 FIGURE 8.5 The A-T-A-R Model Profits = Units sold × Profit per unit Units sold = Number of buying units × Percentage who become aware of the product × Percentage who opt to try the product if they can get it × Percentage of intended triers who can get the product (if it is available to them) × Measure of repeat: 1 + (the percentage of triers who like the product enough to repeat their purchase X the number of additional units bought by repeaters in a year) Profit per unit = Revenue per unit (unit list price less trade margins, promotional allowances, freight, etc.) − Costs per unit (usually costs of goods sold plus direct marketing costs) Therefore: Profits = Buying units × Percent aware × Percent trial × Percent availability × Measure of repeat × (Revenue per unit – Costs per unit) The A-T-A-R Model This is taken from what is called diffusion of innovation, explained this way: For a person or a firm to become a regular buyer/user of an innovation, there must first be awareness that it exists, then there must be a decision to try that innovation, then the person must find the item available to him or her, and finally there must be the type of happiness with it that leads to adoption, or repeat usage.8 We want to use the formula to calculate all the way to profit, so we expand it to include target market size (potential adopters), units purchased by each adopter, and the economics of the operation. But at the heart of the calculation is A-T-A-R. We use one form of this model here in predicting first-year profitability; in Chapter 11 we revisit A-T-A-R as a market-share forecasting tool, and we see it again later in Chapter 18 in the context of simulated test markets. Let’s take a simple example to explain how it works. Assume our company is developing the next generation of smartphone. We presume that the product is analogous to existing smartphones with video display (that is, the new product will be comparable in many ways to these cell phones: similar price, similar target 8 The basic A-T-A-R sequence has been broken down further into many microsteps. One example of this extension is John H. Antil, “New Product or Service Adoption: When Does It Happen?” Journal of Consumer Marketing, Spring 1988, pp. 5–16. Some people use this model in abbreviated form, stopping at unit sales. They calculate market share and make conclusions on that. 206 Part Three Concept/Project Evaluation market, similar benefit provided). A rough estimate, then, of the potential for the new product is the size of the market for the analogous existing product (more about the use of analogous products in forecasting in Chapter 11). To apply the A-T-A-R model, we need the following (hypothetical) data: • Number of owners of smartphones (who are our potential buying units for the new product): 10 million. • Percentage of owners of smartphones we think we can make aware of our new generation phone the first year on the market: 40 percent. • Percentage of aware owners who will decide to try the new phone during the first year and set out to find it: 20 percent. • Percentage of customary consumer electronics retailers whom we can convince to stock the new phone during the market introduction period: 70 percent. (To keep things simple, assume that potential buyers probably will not seek beyond one store if they cannot find it there.) • Percentage of the actual triers who will like the product enough to repeatpurchase within the first year: 20 percent. • Number of additional units bought by these repeat buyers, on average: 1 (that is, they buy a total of two phones, possibly one for home use and one for office use). • Dollar revenue at the factory, per device, after trade margins and promotion discounts: $100. • Unit cost, at the intended volume: $50. The profit contribution forecast, based on the A-T-A-R model as depicted in Figure 8.5, would be: Profit contribution 5 Potential 3 AW 3 T 3 AV 3 R 3 margin 5 10 million 3 0.40 3 0.20 3 0.70 3 1.20 3 ($100 2 $50) 5 $33,600,000, where AW 5 awareness and AV 5 availability, and R is calculated as shown in Figure 8.5 as 1 1 (percentage of repeaters 3 number of additional units) 5 1 1 (0.20 3 1) 5 1.20. What we did was prepare a mathematical formula and run it through one set of data. Since the development was about finished when the calculation was made, the forecast was fairly solid. But the formula could have been used at the very beginning as well. Only a few figures (e.g., number of potential adopters) are known at the start, but estimates can be plugged into the other spots and the model can be set up for use down the line. As with the other parts of this chapter, the A-T-A-R model gives us guidance on evaluation system design. You can immediately see the importance of awareness, trial, and so on. That means tests will have to be run in which customers are checked for their interest in trying, their reactions after trying (how likely would they be to try again?), and whatever else contributes to the formula. Chapter Eight The Concept Evaluation System 207 There is nothing magic in the formula; it simply states the critical factors and shows their relationship to one another and to the sales and profit forecasts. Two things are important about this model’s sales and profit forecasts for the new cell phone: 1. Each factor is subject to estimation, and in every development phase we are trying to sharpen our ability to make the estimates. For example, we may be trying to check the introductory promotion’s awareness-building capability or just how much price discounting we must do to motivate a first purchase of the device. We may be worried about how we’re going to get enough distribution to make the product available when the intended market seeks it. 2. An inadequate profit forecast can be improved only by changing one of the factors. For example, if the forecast of $33,600,000 profit contribution is insufficient, we look at each factor in the model and see which ones might be changed and at what cost. Perhaps we could increase the retail margin by 5 percent and get another 10 percent of retailers to stock it. On the other hand, perhaps an increase in advertising would produce more awareness. Qualitative changes (such as a new advertising theme) can be made in addition to the quantitative. The proposed changes are then run through the formula again, which yields another set of results, some more changes, and so on. Sometimes the issue raised is so fundamental that it is more efficient to cycle back to an earlier phase in the development. The fact that the model is set up in spreadsheet format makes for easy simulations and what-if tests. A-T-A-R is a term that came from consumer products marketing. Industry has traditionally used slightly different language, so a natural question is, “Does the model apply to all types of new products, including industrial ones, and services too?” The answer is absolutely, though each term may be defined slightly differently in different settings. See Figure 8.6 for the definitions of terms that vary. A consumer buying unit may be a person or a home. For office furniture, it will perhaps be a facility manager; for industrial products, it will generally be a purchasing or engineering person FIGURE 8.6 Definitions Used in the A-T-A-R Model Buying unit means purchase point; may be each person, household, or department who participates in the decision. Aware means someone in the buying unit hears about the existence of a new product with some characteristic that differentiates it; subject to variation between industries and even between developers. Available means the percentage chance that if a buyer wants to try the product, the effort to find it will be successful; often “percent of stores that stock it.” Direct sellers have 100 percent availability. Trial is variously defined; may be use of a sample in an industrial setting where such use has a cost associated with it; in most situations, means an actual purchase and at least some consumption. Repeat is also varied; on packaged goods, means to buy at least one (or two or three) more times; on durables, may mean be happy and/or make at least one recommendation to others. 208 Part Three Concept/Project Evaluation (part of a team); and for a consumer bank loan, it will once again be a person or a family. Product developers know what these definitions should be; the target users were selected partly because we know them well. Without a precise definition there can be no worthwhile measurement. In each case, something about the term tells you how to define it. For awareness, we want to know if the buying unit has been sufficiently informed to stimulate further investigation and consideration of trial. If it has only heard the product’s name, it probably won’t. For trial of our new product, we might imagine an in-store situation where the prospective customer tries out the smart phone and sees if the product is satisfactory. For other kinds of new products (such as a new electronic security device to be installed on a car), you may wonder how a potential buyer could try the product, that is, try it in risky situations, waiting for a thief to challenge it. The answer is that we get as close to the perfect answer as we can, and that sometimes calls for ingenuity. Otis Elevator Company, for example, is not selling cake mixes—they simply take prospective buyers to a site where the elevator under consideration is already installed. The trial is not perfect, but it is close enough for real customer learning. Sometimes firms use vicarious trial where a person or firm who did try something shares results with someone who can’t try it. But trial there should be, and Chrysler once wanted a new item tried so badly that they paid people $50 if they would take a demonstration ride (and later show proof of purchase of a new car within a month or so). It can be done. In a trial, we want two things to happen: 1. The buying unit went to some expense to get the trial supply—if there was no cost, then we can’t be sure there was evaluation of the product message and interest created. Anyone can taste some sausage in a supermarket, but that doesn’t mean the taste was a true trial. 2. The buying unit used the new item enough to have a basis for deciding whether it is any good. For availability, we want to know whether the buyer can easily get the new product if a decision is made to try it. This factor is more standard, and for consumer products it is usually the percent of those outlets where our target buyers shop where the firm has stocking of the new item. If the firm sells direct, there is always availability (unless the factory has extended back-orders). Another measure commonly used is all commodity volume, or ACV, which is the percentage of the market that has access to the product in local distribution channels. Business-tobusiness often uses distributors of some type, usually under some franchise or semifranchise agreement, again pretty much assuring availability. But many small firms cannot be sure of availability and spend much of their marketing money on trying to get it. Repeat is easy for consumer packaged goods (usually, a repeat purchase), but it really means the trial was successful—the buying unit was pleased. For onetime purchases (industrial or consumer), we have to decide what statistic will tell us that. Some people use the direct one: “Were you satisfied?” Sometimes an indirect one—such as “Have you had occasion to recommend the product to others?”—is better. In the case of the cellular picturephone-camera, buying a second unit would Chapter Eight The Concept Evaluation System 209 be a good measurement. In any case, a firm should arrive at some acceptable definition and stick with it, thus building up experience to measure against. Where Do We Get the Figures for the A-T-A-R Model? The evaluation techniques shown in Figure 8.1 (primarily concept testing, product use testing, and market testing) will provide the data needed for the A-T-A-R model. You are not yet acquainted with the various tests, but they will be tied into the A-T-A-R model as they come up. Though various evaluation events can help on several of the key factors, we are usually most interested in the one event that makes the biggest contribution—noted as Best in the figure. And we should know which these are prior to starting the evaluation. That way, we spend our limited funds first on the best steps and then on others if funds are available. Also, if we have to skip a step (for example, the concept test), we immediately know we are leaving open the question of whether users are likely to try the item when it becomes available. If we are going to do product use testing, then it should be set up in a way that lets us go through a concept test in the process of getting people to sign up for the use testing. It’s later than we wanted, but better now than not at all. Further Uses of the A-T-A-R Model In this chapter, we outlined the use of A-T-A-R relatively early in concept evaluation as a rough forecasting tool (i.e., what is the potential profit contribution of this product, is it satisfactory, and how could it be improved?). The A-T-A-R model is useful at this early point, as it provides an early sales and profit forecast based on estimates specific to the new product (i.e., the A, T, A, and R)—it calls for numbers that usually can be researched, and it uses them in a managerial way. We will use A-T-A-R at later phases in the new products process and therefore will return to it occasionally in this book. In Chapter 11, we will use it as the basis of a somewhat more detailed sales forecasting model. A-T-A-R is implicit throughout the discussion of market launch planning (Chapters 16 and 17): What else could be more important for the marketing effort to do than achieve awareness, trial, availability, and repeat use?9 Finally, in Chapter 19, we visit it again, this time as a tool to assess the launch, identify where the problem areas are, and steer it back on course. Summary This chapter looked at the factors that aid in designing an evaluation system for the basic new products process, designed to provide pieces of information that guide the project in its journey to the market. First came the cumulative expenditures curve, the risk/payoff matrix, and the decay curve. Then we looked at several descriptors of most situations, the primary one being that almost everything about a process situation is tentative. The product itself is still evolving, at least 9 One of our top sales forecasting experts addressed the new product situation, particularly the issues surrounding the many techniques. Robert J. Thomas, “Issues in New Product Forecasting,” Journal of Product Innovation Management, 10(3), September 1994, pp. 347–353. 210 Part Three Concept/Project Evaluation until it sells successfully; the actual date of marketing is increasingly unclear as firms adopt limited marketing approaches; evaluation actually begins with the innovation charter well before ideation; and a product is an assemblage of many parts, each requiring its own evaluation. Lastly, we introduced the A-T-A-R model, which tells us some of the critical steps, how our information about them can be used to forecast sales and profits, and how to design an evaluation system accordingly. What are the specific tools, what can each do, and what are their weaknesses? The ones we use in Phase III of the basic process, prior to entering the development phase, are covered in the next two chapters. Others come later. Applications 1. “During a recent management meeting, two of my division managers in the United Kingdom got into quite a tussle over the programs they use to evaluate new product ideas. One of them said he felt evaluation was very important; he wanted to do it quite completely, and he certainly didn’t want anyone working to further the development of an item unless the prospects for it looked highly promising. The other manager objected to this, saying she wanted products to move rapidly down the pike, saving the serious evaluation for the time when she had the data to make it meaningful. Both persons seemed to have a point, so I just let it ride. What do you think I should have said?” 2. “I don’t know what business school professors would say, but it often seems to me that we might be just as well off if we didn’t do any evaluation on new products. Just produce the ones we’re convinced will sell the best and really support those. Let’s face it—we never have reliable data anyway, and everyone is always changing minds or opinions. Never knew so many people could say I told you so.” 3. “Let me tell you another funny thing about evaluation. It seems as though the folks involved in it never use the facts or data that they should and instead use some sort of surrogate data. I don’t see why you have to beat around the bush. Why not just gather the real facts in the first place and not use those substitutes?” Case: Chipotle Mexican Grill10 In 1993, Steve Ells opened a burrito-and-taco restaurant in a Denver storefront, not far from the University of Denver campus and popular with students. He named it Chipotle Mexican Grill, after the dried pepper common in Mexican cooking. A trained chef and graduate of the Culinary Institute of America, Steve’s idea was 10 Information from this case was obtained from Anonymous, “Chipotle: Fast Food with ‘Integrity,’” BusinessWeek.com, February 16, 2007; Anonymous, “Chipotle’s Chef Has His Payday,” BusinessWeek.com, January 27, 2006; Marc Gunther, “Can Fast Food Be ‘Good’ Food?,” cnnmoney.com, September 13, 2006. Chapter Eight The Concept Evaluation System 211 for Chipotle to be a cash cow to help him finance a “real,” upscale restaurant. Chipotle, however, began branching out: first to several locations in and around Denver, then eventually nationwide. In 1998, McDonald’s bought a 91 percent stake in Chipotle; this was followed by a 2006 initial public offering in which McDonald’s retained 69 percent of the stock and 88 percent of voting rights. By the end of 2005, there were about 500 Chipotle outlets generating approximately $600 million in sales annually. Currently, about 15,000 people are employed by Chipotle. Steve Ells still serves as the chairman and CEO. What accounts for the success? For starters: a simple menu, skilled cooking techniques, fresh preparation, served quickly, and a “cool” setting. The menu is described as “fast casual” and is at first glance rather limited: only tacos and burritos. (As the only real change in over a decade, salads were very recently added.) Steve notes, however, that there is a wide variety of flavors to choose from, and by focusing on a couple of things, Chipotle has been able to ensure that they do them very well. He argues that too big a variety leads to too much pre-preparation or processed ingredients, and notes that since its earliest days, Chipotle items are always made-to-order. He admires and tries to emulate In-N-Out Burger, a 50-yearold chain that sells only fries, hamburgers, and milkshakes, but offers high quality for which people are willing to pay a premium. He also admired Steve Jobs of Apple, and feels that one can learn a lot from Jobs’s “passion for not accepting mediocre stuff.” There are some other factors at work here as well. The pork used by Chipotle comes from pigs raised naturally, without hormones, on family farms. In 2005, Chipotle switched its dairy purchasing policy: Since then, all sour cream comes from cows that are not given the hormone rBGH. The restaurants use fresh avocados, tomatoes, and peppers, prepared from scratch. And Chipotle believes in the “open kitchen” format: People can see for themselves that the food is fresh. Steve’s term for Chipotle’s vision is “food with integrity.” He notes that he loves seeing high school students going into a Chipotle, spending a couple of dollars more for a meal than they might elsewhere, and maybe getting a bottle of water instead of a soda. Chipotle’s has never advertised as a place for kids or teenagers to eat. In fact, it does very little advertising. Steve feels that advertising the “food with integrity” vision won’t work; to use his term, it would “be too preachy.” Rather, he lets the food quality, value, and convenience do the talking. The open kitchen also helps promote Chipotle’s freshness and quality. He estimates that no more than 5 percent of his customers know about “food with integrity.” The rest come in because they like the taste or the value, or just because “the place looks cool.” But Steve believes that “food with integrity” can mean much more. He points to the popularity of organic food stores such as Whole Foods. People respond positively to organic, sustainably grown vegetables, humanely raised meats, and fewer preservatives. At Chipotle, he has considered switching over to all-organic produce, but does not want to price a dining experience at Chipotle out of the average fast-food customer’s range. According to one estimate, going organic overnight would make the retail price of a burrito jump to $15. Nevertheless, switching to natural pork increased the price of carnitas from $4.50 to $5.50, but sales also went up. 212 Part Three Concept/Project Evaluation Totally organic is perhaps a long-term goal, and there are certainly interim steps. About 30 percent of its beans are organic, though other vegetables are generally not. About 60 to 70 percent of the chicken and about 40 percent of the beef is sourced naturally, as well as all of the pork. The sour cream is still not organic, though it is hormone-free. The other thing to keep in mind is that as Chipotle grows, it gets more power in the supply channel. As a tiny chain, it was unable to get natural chicken thighs from high-end supplier Bell and Evans, but at its current size, it can do so today. Though today’s Chipotle would seem to be among the leaders in providing healthy fast food to the public, Steve feels that he is still lagging behind. His goal is for all Chipotle restaurants to offer only organic, pesticide-free ingredients, lacking preservatives and artificial flavors and colors, and all natural, humanely raised meat. He would be even more delighted if every restaurant were to follow the same vision. Let’s call this the “all-organic concept” for short, recognizing that organic is only a part of the whole vision here. If you were advising Steve Ells, what could he have done to evaluate the allorganic concept? Is the concept viable at all? How would he be able to estimate the price elasticity (that is, how high does price have to get before he begins losing significant numbers of customers)? Given the fierce competition in this industry, is his concept pleasant but unrealistic? Or does the organic position provide Chipotle with sustainable competitive advantage? Specifically, consider the challenges posed by competitors such as Taco Bell that are adding to their product lines and trying to move into the “healthier” Mexican food product space. How protected is Chipotle from this kind of competitive action, now or in the future? Case: Concept Development Corporation11 Late in 2012, three bridge-playing friends in a southern college town decided to start their own firm. One, Bob Stark, worked for General Motors as a planning manager in a local assembly operation. The second, Betsy Morningside, was a speech and theater professor at the college. The third, Myron Hite, was a CPA who worked for one of the Big Eight accounting firms. All three were exceptionally creative and especially enjoyed their bridge sessions because they had a chance to brag about their new creations and to hear the creations of the others. It was all for fun until one evening it struck them that it was time to stop the fun and start making some money from their many ideas. So they quit their jobs, pooled their savings, rented a small, three-room office, hired a couple of people, coined the name Concept Development Corporation, and started serious work. A professor from the college was asked to “make a contribution to local entrepreneurship” by setting up a system to evaluate their ideas. They fully realized they were better at thinking up things than evaluating them. They also were aware of their deficiencies: little staff, little money, little experience in making things like 11 This is a real situation, slightly camouflaged. Chapter Eight The Concept Evaluation System 213 the ones they created, and little time before their meager savings disappeared completely. They began with two product areas. One was toys, broadly defined as things children played with, especially educational activities. The other area was writing services, something they had not intended to work on but which arose as temporary spin-offs from the abilities of one of the two people they hired. These services primarily involved designing and writing instruction sheets for area firms (training manuals, copy for package inserts, instruction signs—anywhere words were used to instruct people in doing things). The individual had some background in instructions and was experienced in writing and layout work. So they decided to develop new items along that line as well. Their strategy was to develop unique toys that required little up-front expenditures (for example, dies and packaging equipment). All three were too creative to settle for imitation. Most toys would have some game or competitive aspect, be educational, and involve paper, color, numbers, and the like. They figured “most of the stuff would be for children under 12.” And, of course, they needed products that would catch on fast and sell well. The writing services would be partly reactive in that they would do whatever clients asked them to do. But, being creative, they also planned to create innovative services—new ways of meeting industry and business needs. For example, they wanted to offer a special test/training service, whereby after developing a training manual or instruction sheet they would have some employees for whom the piece was developed come to a special room where they would read the material, apply it in some fashion, be tested on it, and so on. What they delivered to the client would be proven to work. They had many such ideas. The professor went back to the college and decided to let a new products class assist in the assignment. The students were asked to think about the new firm’s situation, the general evaluation system in Figure 8.1, and the various purposes and special circumstances discussed in Chapter 8 and then come up with one general guideline statement of evaluation policy for the toy ideas and another for the new services. They hadn’t yet studied specific techniques (such as concept testing), but they could clearly indicate which of the phases in Figure 8.1 were the most critical, where the toughest decisions would be, and so on. The professor was especially interested in the differences between goods and services. He wanted the students to state, as specifically as possible, what they felt were the major differences between the evaluation of tangible goods (like toys) and services, why these differences existed, and what the consequences were with respect to evaluation techniques and methods. C H A P T E R N I N E Concept Testing Setting This chapter is the first of two spelling out the various tools for evaluating new products (goods and services) prior to undertaking technical development. Chapter 9 will cover the product innovation charter and market analysis activities, which occur before the idea appears, and the initial reaction and the concept testing, which occur immediately after the idea appears. We will be investigating the different kinds of evaluation steps leading up to, but not including, the full screen as shown in Figure III.1. Chapter 10 will examine the full screen in detail. All the evaluation steps shown in this chapter and Chapter 10 should be viewed as investments—the additional information provided far outweighs their cost, and cutting corners in getting important early information from customers can prove costly in the long run. Recall that, at the end of Chapter 6, we had left unresolved the issue of whether customers would actually buy products corresponding to the gaps we had identified. We need to be able to relate customer needs and preferences to these gaps to ensure that we don’t develop the “wrong” product. In this chapter, we will show how we can use perceptual mapping and conjoint analysis to analyze market needs and preferences, to segment the market according to benefits sought, and to test how well our concept will be accepted by the market. The Importance of Up-Front Evaluations In recent years there has been much more activity at this phase of the process, prior to development. There still is not nearly enough, but the practice is spreading, as product managers are under continual pressure to boost quality and reduce time to market while not incurring cost overruns. There’s another reason: This is when finalizing the positioning statement, a cornerstone of our entire marketing strategy, becomes a focus of attention. The biggest cause of new product failure is that the intended buyer did not see a need for the item—no purpose, no value, not worth the price. It is in concept testing, a key part of this chapter, where we get our first confirmation that this will be a quality product. We save time by gathering information and making decisions that help assure the product will move through development fast, and with a minimum of looping back to correct some problem. Spending time here saves Chapter Nine Concept Testing 215 time overall, and there is good evidence of this.1 We lower cost in several ways, one coming when we avoid the rising cumulative expenditures curve you just met in Chapter 8—with the cost curve ever rising, the best time to get off a loser is at the bottom of the curve. Another cost cutter is the elimination of the many losers naturally picked up in an aggressive concept generation program. It’s difficult to cut at this point, but we have to, so we want to do it in the correct way. Finally, on cost, information gathered here helps us make cost forecasts—just how close are we going to be to competition on the proposed item and how draconian must our efficiencies be. Quality, time, and cost—there is no better reason for taking action at this point. But highly relevant is that this is also the phase where we set the basic marketing strategy on firm ground. We confirm the target market (the user whose needs we are trying to find and solve) and settle on a product positioning statement (just how the new item will be better than others already out there). The positioning statement guides all the rest of the marketing activities. In this chapter, we look at what happens at this point in the process, what firms need to be doing, and in particular how one sets about doing what seems to be the very best approach—concept testing. The Product Innovation Charter The earliest evaluation that a firm makes is of itself and its situation. That evaluation yields a priori conclusions about new product proposals. The firm reaches these conclusions while making basic strategic decisions, as discussed in Chapter 3 on the product innovation charter. These decisions decide what types of new products fit best. We saw earlier that Kellogg’s sought snack foods that capitalized on existing food technologies and familiar brand names and trademarks. The PIC itself will eliminate many new product ideas. In advance and without knowing the concepts, the firm decides to reject ideas that violate PIC guidelines. Following the PIC should result in excluding the following kinds of ideas: • Ideas that require technologies the firm does not have. • Ideas to be sold to customers about whom the firm has no close knowledge. • Ideas that offer the wrong degree of innovativeness (too much or too little!). • Ideas wrong on other dimensions: not low cost, too close to certain competitors, and so on. 1 Several studies show this, one being Albert L. Page and John S. Stovall, “Importance of the Early Stages in the New Product Process,” Bridging the Gap from Concept to Commercialization (Indianapolis, IN: Product Development & Management Association, 1994). Others are Robert G. Cooper and Elko J. Kleinschmidt, “Determinants of Timeliness in Product Development,” Journal of Product Innovation Management, 11(5), November 1994, pp. 381–395; and Mitzi M. Montoya-Weiss and Roger Calantone, “Determinants of New Product Performance: A Review and Meta-Analysis,” Journal of Product Innovation Management, 11(5), November 1994, pp. 397–417. 216 Part Three Concept/Project Evaluation The charter given to new products management thus eliminates more product ideas than all the other evaluations combined. By coming at the beginning of the new products system, it precludes the unfortunate practice of having unwanted proposals eat up valuable development funds before they are detected. Market Analysis The second evaluation that precedes appearance of the concept is an in-depth study of the market area that the product innovation charter has selected for focus. The study takes place immediately after the PIC is approved, and the depth of the study depends on how well the firm already knows the market selected. Ongoing ideation in support of present product lines takes place within a standing type of PIC, and no special study is necessary (assuming current product managers do their jobs correctly). Initial Reaction Concept generation follows the market analysis just discussed. Concepts begin flowing in, usually very fast, and opinions on them are formed instantly. But most firms have evolved some special technique to handle this deluge more systematically, which we will call initial reaction. Quick and inexpensive initial reactions must resist the “bazooka effect” (where suggestions are quickly blasted out), so several provisos apply: 1. The idea source does not usually participate in the initial reaction. A person who has an idea may want to explain it and argue for it, but this person should probably not have a vote in the decision to advance the idea or drop it. 2. Two or more persons are involved in any rejection decision, based on the “fragility of new ideas” concept discussed in Chapter 8. The rejection percentage is much higher here than at any other stage, but involving two or more persons dilutes the biases of a single person. 3. The initial reaction, though quick, is based on more than a pure intuitive sense. The evaluators are trained and experienced; records are kept and reviewed; and objective aids are sought. One of several techniques used in this initial reaction is the product innovation charter. Knowing such things as whether a firm wants to be first or last, high risk or low risk, internally or externally developing, and in shoes or handbags leads to quick and decisive action. Most firms also make use of heuristics (rules of thumb) for this rough screening. For example, managers look at the scale required (is it in our league?), the competitor we would have to face, the state of the art the idea would require, and the fit with our manufacturing and marketing operations. One suggested way for firms to do a rough early screen is to evaluate it on three factors: • Market worth: What is the attractiveness of the new product to the targeted customer population? Chapter Nine Concept Testing 217 • Firm worth: Is the new product project viewed positively by management? Does this new product project enhance the firm’s competencies? • Competitive insulation: Can the product’s advantage be maintained against competitive retaliation?2 Some managers prefer to use a small-scale informal survey at this initial reaction point, particularly when some aspect of the proposal extends beyond the evaluator’s experience. But such a survey should be held to the level of telephone checks with professional colleagues. Concept Testing and Development Years ago, when Alan Ladd Jr. reigned as top judge of new movie scripts at Twentieth Century Fox Film Corporation, he revealed that his product proposal evaluation system ended about as soon as it began. He would just read a script and decide whether to make the movie. He and his small staff knew their markets well, had a guiding product innovation charter, and combined their knowledge and the charter with personal judgment to reach decisions. They did not use concept testing, full screening, or product use testing. Ladd said, “It’s based on my intuition and experience. There’s no way to put it on a chart or graph or formulate it.”3 Perhaps. Some agree with Mr. Ladd, but most do not. Most major firms make frequent use of concept testing. It is a mandatory part of the process for makers of consumer packaged goods. And use is growing in industrial firms, which actually invented it. Business-to-business firms have always spent much time talking with users about their needs and problems, what suggestions they have, what they think about various ideas, and so on. They just never called it concept testing. But first, let’s deal with some concerns about this activity—there are times when it doesn’t help. When the prime benefit is a personal sense, such as the aroma of a perfume or the taste of a new food, concept testing usually fails. The concept cannot be communicated short of actually having some product there to demonstrate. A type of kids’ gum popular in the early 1990s (sour gums called Cry Baby and Warhead) tasted so bad that even product use testing showed children hated it. But, when the gum became available, kids became masochistic to the tune of almost $100 million a year. Second, concepts embodying new art and entertainment are tough to test. Whistler could not have concept-tested his idea for a painting of his mother. The inventor of the Ferris wheel could not have surveyed people to ask what they thought 2 Rita Gunther McGrath, “Advantage from Adversity: Learning from Disappointment in Internal Corporate Ventures,” Journal of Business Venturing, 10(2), March 1995, pp. 121–142. 3 Earl C. Gottschalk, “How Fox’s Movie Boss Decides That a Script Is a Powerful Winner,” The Wall Street Journal, May 17, 1979, p. 1. Many years later, he was still doing the fast reaction and had some major successes—e.g., Star Wars, Nine to Five, and Thelma & Louise. But he had also worked at several different studios and had marketed some misses—The Right Stuff, Quigley Down Under, and Not Without My Daughter. Ronald Grover, “Can Alan Ladd Jr. Make Leo the Lion Roar?” BusinessWeek, August 12, 1991, pp. 65–66. 218 Part Three Concept/Project Evaluation of it. The thrill simply had to be experienced personally. A study of audience testing of television pilots showed how inaccurate these tests could be. Long-running shows including classics like Seinfeld, Gilligan’s Island, and All in the Family did poorly in tests, while others that scored high in audience testing never caught on. As a result, some network executives find it best to go with “gut instincts.”4 Third, when the concept embodies some new technology that users cannot visualize, it is also a weak tool. Alberto-Culver realized this when it first tested the concept of hair mousse. Women accustomed to sprays could not imagine putting “stuff like that” on their hair. Only after the company developed the product and set up training classes in salons did women agree to try the mousse. As another example, physicians rejected the concept of a heart pump—they could not know the full attributes (and thus the risks) of such a product before work was completed. Fourth, there are times when firms mismanage concept testing and then blame the tool for misleading them. In one of the most famous market research flubs of all time, Coca-Cola asked their customers to taste-test New Coke in comparison to both regular Coke and Pepsi, and received favorable replies (we know how that worked out for Coca-Cola). But they then took that taste testing to mean customers would buy the product when it got a new name. This is actually the same problem as the heart pump—customers were asked to predict their behavior without knowing all the facts. They can’t, but will if asked, and will deceive developers who aren’t careful. Another mismanagement was when several fast-food chains asked customers if they wanted diet burgers. Not only was it an unknown taste situation (see above), but people are notoriously inclined to predict “worthy” behavior and then do something else. Fifth, consumers sometimes simply do not know what problems they have. We discussed this in the chapter on problem-based ideation. Steelcase, for example, found they could not use concept testing on special furniture for use by teams. The team members had no feeling for what they didn’t have, so Steelcase observed them in action and came up with a winner: furniture that lets them do some of their work collaboratively and some privately. The microwave oven was another similar example—we didn’t know what to do with it even after it hit the market, and we certainly could not have responded helpfully to researchers asking us what we thought about the concept.5 More recently, some customers wondered what in the world they would do with an iPad; presumably many of these people now use theirs daily and cannot imagine life without it. Oddly, in spite of evidence to the contrary, some new products people have doubts about concept testing on business and industrial products and on services. Regarding the former, if one sticks to situations where the customer has the ability to make judgments, those judgments are worth gathering; but major technological breakthroughs don’t qualify for that, and we just have to take the risk. On services, there is no question as to whether people can tell us what they see useful if they 4 Brooks Barnes, “Trusting Gut Instincts, WB Network Stops Testing TV Pilots,” The Wall Street Journal, May 3, 2004, pp. B1, B7. 5 Some of these examples are discussed in Justin Martin, “Ignore Your Customer,” Fortune, May 1, 1995, pp. 121–128. Chapter Nine Concept Testing 219 can see it (but watch for the intangibles above). They can. But because there is usually little technical development, there is less need to do concept testing. If it is simple to go from concept to full service description (a form of prototype), then the services firm can proceed to what is called prototype concept testing. Such testing is, of course, much more reliable with a physical prototype to talk around. Concept testing is useful in most cases, and right now the burden of argument lies with the person who wants to skip it. Unfortunately, we will for a long time hear about such firms as CalFare Corporation, which, without concept testing, developed shopping carts with a “special fifth wheel” that locked into place if the cart was taken from the premises (and run over a rough surface). The cart would then only go in circles. But most store managers said they were not interested. They feared negative publicity, and customers being scared away. A competitor said it often happened that the cart went awry and started circling around the dairy departments. The developers were caught off guard by the very negative reactions they got.6 What Is a New Product Concept? Webster’s says a concept is an idea or an abstract notion. Businesspeople use the term concept for the product promise, the customer proposition, and the real reason why people should buy. We have, of course, previously seen it in Chapter 4, describing it as a stated relationship between product features (form or technology) and consumer benefits (needs satisfied). That is, the product concept is a claim of proposed satisfactions. This promise is open to four interpretations: 1. 2. 3. 4. The producer’s perception of the features of the new product. The consumer’s perception of the features of the new product. The producer’s estimate of the benefits delivered by that set of features. The consumer’s estimate of the benefits delivered by that set of features. These are only forecasts or guesses at this time—not reality, even with a prototype in hand. They rest on expectations. Thus a complete new product concept is a statement about anticipated product features (form or technology) that will yield selected benefits or problem solutions relative to other products already available. An example is “A new electric razor whose screen is so thin it can cut closer than any other electric razor on the market.” Sometimes a part of the concept can be assumed; for example, saying “a copier that has twice the speed of current models” assumes the benefits of speed can go without saying. The Purposes of Concept Testing Recall that concept testing is part of the prescreening process, preparing a management team to do the full screening of the idea by providing input into the full 6 David Jefferson, “Building a Better Mousetrap Doesn’t Ensure Success,” The Wall Street Journal, November 18, 1991, p. B2. 220 Part Three Concept/Project Evaluation screen just before beginning serious technical work. We look at information to help the screeners use scoring models and write out product protocols in Chapter 10. Therefore, the first purpose of a concept test is to identify the very poor concept so it can be eliminated. If music lovers, for example, don’t care about a new kind of compact disc that will last forever (because they download all their music from online sources) and thus reject it out of hand, the concept is probably a poor one. If the concept passes the first hurdle, a second purpose is to estimate (even crudely) the sales or trial rate that the product would enjoy—a sense of market share or a general range of revenue dollars. Some people believe this buying prediction is worthless. Others claim a clear, positive correlation between intention and purchase. One longtime practicing market researcher claimed to have confidential data showing correlations of 0.60 and well above.7 The buying intention question appears in almost every concept test. The most common format for purchase intentions is the classic five-point question: How likely would you buy a product like this, if we made it? 1. 2. 3. 4. 5. Definitely would buy. Probably would buy. Might or might not buy. Probably would not buy. Definitely would not buy. The number or percentage of people who definitely would buy or probably would buy are usually combined and used as an indicator of group reaction. This is called the top-two-boxes score, as it is the total number of times one of the top two boxes on the questionnaire (definitely or probably) were checked. Incidentally, Nabisco says “try” (not “buy”), because buyers really are still quite tentative at this point. Whether this many people actually purchase the item is not important. Researchers have usually calibrated their figures, so they know, for example, that if the top two boxes total 60 percent, the real figure will be, say, 25 percent. They do this from past experience, discounting what people tend to say in interview situations. Direct marketers can do the best calibration because they will later be selling the tested item to market groups they surveyed; they can tell exactly how actual behavior matches stated intentions. The data banks of the BASES Group, the largest supplier of concept tests and now part of A. C. Nielsen, literally let a client company calibrate all of its concept test questions by product type. For a price, BASES translates a client’s raw intentions data into probable intentions.8 Incidentally, sometimes experience calibrates the probable intention higher than the respondents say now. On complex products, people often use caution at concept testing time but end up buying the product when they have a chance to see the final item and hear all about it. (Recall the heart pump.) 7 Personal communication with Anthony Bushman, now professor of marketing, San Francisco State University. 8 BASES is only one of many concept testing suppliers that provide this service. Chapter Nine Concept Testing 221 Obviously, the concept’s sales potential will be closely related to how well it satisfies customer needs or offers desired benefits to the customer. Later sections of the chapter show more advanced analytical procedures that identify customer segments based on benefits sought. Knowing the benefit segments that exist in the marketplace, the firm can identify concepts that would be particularly desirable to specific segments or niches. The third purpose of concept testing is to help develop the idea, not just test it. Concepts rarely emerge from a test the way they went in. Moreover, a concept statement is not enough to guide R&D. Scientists need to know what attributes (especially benefits) will permit the new product to fulfill the concept statement. Because the attributes frequently oppose or conflict with each other, many tradeoffs must be made. When better to make them than when talking with people for whom the product is being developed? Near the end of this chapter, we will see how conjoint (trade-off) analysis, a technique we discussed in Chapter 7, is frequently used for this task. Considerations in Concept Testing Research Prepare the Concept Statement A concept statement states a difference and how that difference benefits the customer or end user: “This new refrigerator is built with modular parts; consequently, the consumer can arrange the parts to best fit a given kitchen location and then rearrange them to fit another location.” If you think this sounds somewhat like a positioning statement, you are correct. And if the interviews are with a logical target group of potential buyers, the principal parts of a marketing strategy are in place—target market and product positioning. This is consistent with the basic new products process, where we say that the product and its marketing plan are developed simultaneously. Format Practitioners urge that any concept statement should make the new item’s difference absolutely clear, claim determinant attributes (those that make a difference in buying decisions), offer an element of familiarity by relating in some way to things familiar to the customer, and be completely credible and realistic. And short, as short as possible, although there have been concept statements of 3–5 pages that worked very well in complex technical situations.9 This information is usually presented to potential buyers in one of several formats: a narrative (verbal) format, a drawing or diagram, a model or prototype, or in virtual reality. Early in concept testing, it apparently does not make too much difference which of these formats is used, as all yield about the same answers from 9 Regarding clarity, Anheuser-Busch said consumers had difficulty understanding Bud Dry, even when it was marketed. Perhaps the reason lies in what an executive said it was: “A cold-filtered draft beer—not pasteurized—with no aftertaste, basically a full-alcohol, light beer, a cleaner beer.” (So, is other beer not clean?) 222 Part Three Concept/Project Evaluation FIGURE 9.1 Concept Test Format—Plain Verbal Description of the Product and Its Major Benefits A major soft-drink manufacturer would like to get your reaction to an idea for a new diet soft drink. Please read the description below before answering the questions. New Diet Soft Drink Here is a tasty, sparkling beverage that quenches thirst, refreshes, and makes the mouth tingle with a delightful flavor blend of orange, mint, and lime. It helps adults (and kids too) control weight by reducing the craving for sweets and between-meal snacks. And, best of all, it contains absolutely no calories. Comes in 12-ounce cans or bottles and costs 60¢ each. 1. How different, if at all, do you think this diet soft drink would be from other available products now on the market that might be compared with it? Very different Somewhat different Slightly different Not at all different 2. Assuming you tried the product described above and like it, about how often do you think you would buy it? Check one More than once a week About once a week About twice a month About once a month Less often Would never buy it Source: NFO Research, Inc., Toledo, Ohio, now part of TNS, a worldwide market information company. See www.tns-global.com. the respondents.10 All the concept testing techniques we discuss here are commonly used for business-to-business product development, though in those cases it is especially important to provide sketches, models, and/or other renditions of the concept such that meaningful, objective reactions can be obtained.11 Figure 9.1 shows an example of the narrative format. Some people prefer a very brief presentation, giving only the minimum of attributes and letting the respondent offer additional ones. Others prefer a full description, approaching what a diagram or prototype would provide. Drawings, diagrams, and sketches comprise a second way to present concepts to respondents. Figure 9.2 demonstrates the use of a drawing. Drawings and the like usually must be supplemented by a narrative statement of the concept. 10 Gavin Lees and Malcolm Wright, “The Effect of Concept Formulation on Concept Test Scores,” Journal of Product Innovation Management, 21(6), November 2004, pp. 389–400. 11 Ronald L. Paul, “Evaluating Ideas and Concepts for New Business-to-Business Products,” in M. Rosenau, A. Griffin, G. Castellion, and N. Anscheutz (eds.), The PDMA Handbook of New Product Development (New York: John Wiley, 1996), pp. 207–216. Chapter Nine Concept Testing 223 FIGURE 9.2 Concept Test Format— Verbal Description plus Sketch Aerosol Hand Cleanser A large-size can of hand cleanser concentrate that completely eliminates those lingering unpleasant odors that come from handling fish, onions, garlic, furniture polish, etc. Not a covering odor! Just press the button and spray directly on the hands, rub for a few seconds, and rinse off under the faucet. 24-ounce aerosol can will last for months and can be easily stored. Costs $2.25. 1. How interested would you be in buying the product described above if it were available at your supermarket? Check one I would definitely buy I would probably buy I might or might not buy I would probably not buy I would definitely not buy Responses in sample (%) 5% 36% 33% 16% 10% 100% Total Note: These hypothetical response percentages are for illustrative purposes only. Source: NFO Research Inc., Toledo, Ohio, now part of TNS, a worldwide market information company. See www.tns-global.com. Figure 9.2 also shows what the results might look like. As shown, 5 percent of respondents said they would definitely buy the product and 36 percent said they would probably buy it, so the top-two-boxes score is 5 1 36 5 41 percent. Note that while Figures 9.1 and 9.2 present classic concept tests administered by mail, these concept statements can obviously be converted to online testing with minimal difficulty. Prototypes, or models, are a third, more expensive form of concept statement because many decisions have to be made about the new product to get it into a prototype. Whoever builds an early prototype makes lots of decisions about the item that probably should be kept open at this early date. Prototypes are useful only in special situations, as, for example, with simple-to-prepare food products or, at the other extreme, with concepts so complex that the buyer cannot react without more knowledge than a simple narrative would give. A firm in Canada was trying to get reactions to the concept of a traveling medical examining unit that would be driven to various corporation offices where examinations would be given. The answer was to build a small model of the unit, showing layout, equipment, and so on.12 12 Robert G. Cooper, Winning at New Products, 3rd ed. (Cambridge, MA: Perseus Publishing, 2001), p. 162. 224 Part Three Concept/Project Evaluation The fourth type of concept format, virtual reality, captures the advantages of the prototype without most of the disadvantages. Steelcase, the office supply firm, has a software system that allows them to virtually build three-dimensional images of office concepts. The interviewee can actually walk around rooms, seeing things from any angle.13 The real question is “What does it take to communicate to the buyer what we have in mind?” From that point on, it is a question of the cost of better displays versus the need for that information in making forecasts of buying intentions. For office furniture, most buyers want lots of details, but for turnipflavored yogurt, one sentence would probably work.14 Commercialized versus Noncommercialized Concept Statements One decision that needs to be made here is whether to present a commercialized or noncommercialized concept statement. Compare these two concept statements: Light Peanut Butter, a low-calorie version of natural peanut butter that can provide a tasty addition to most diets. A marvelous new way to wake up your diet has been discovered by General Mills scientists—a low-calorie version of ever-popular peanut butter. As tasty as ever and produced by a natural process, our new Light Peanut Butter will fit most weight-control diets in use today. The first is a noncommercialized concept statement (or a stripped description), which presents just the facts (this can even be in bullet points); the second is a commercialized concept statement (or an embellished description) that sounds more like how the product would actually be promoted or advertised to customers. It is not clear that one form is better than the other.15 Due to their similarity to advertising copy, some say that commercialized statements will produce more realistic customer evaluations (that is, greater acceptance); nevertheless, good or poor advertising copywriting might bias the results. Some favor the noncommercial form; they argue, why evaluate the advertising when all we want at this time is reaction to the concept?16 Neither form is better than the other, and many managers simply go for a compromise: a gentle sell that puts advantages in language stakeholders are used to. Some practitioners say that it is most important to keep the idea simple, to be clear and realistic, and not to oversell the concept. Also, if testing several concepts, be 13 Information from William Miller, director of research and business development, at a Product Development & Management Association conference in Southfield, Michigan, January 1995. 14 Actually, Green Giant Vegetable Yogurt in four “flavors” (cucumber, beet, tomato, and garden salad) scored well on concept tests (87 percent top two boxes). But the firm couldn’t deliver what the concept seemed to promise to consumers (should it be crunchy?). This one failed in the marketplace. 15 G. Lees and M. Wright, “The Effect of Concept Formulation on Concept Test Scores,” Journal of Product Innovation Management, 21(6), pp. 389–400. 16 Jeffrey B. Schmidt, “Concept Selection Matrix,” in Jagdish N. Sheth and Naresh K. Malhotra, Wiley International Encyclopedia of Marketing, Volume 5, Product Innovation and Management (West Sussex, UK: John Wiley, 2011), p. 27. Chapter Nine Concept Testing 225 consistent: don’t mix commercial with noncommercial concept statements, and don’t mix radically new concepts with minor improvements.17 Offering of Competitive Information Customers of all types know much less about their current products and other options than we would like. A new concept may well offer a benefit that the customer doesn’t realize is new. One solution is to provide a full data sheet about each competitive product. Many new product managers, however, don’t like to overload the concept statement; it diffuses the message and confuses the customer. Price Another issue turns on whether to put a price in the concept statement. The examples in Figures 9.1 and 9.2 both mention price. Some people object, saying reaction to the concept is wanted, not to its price. Yet price is part of the product (actually, a product attribute in the customer’s eyes), and buyers can’t be expected to tell purchase intentions without knowing price. An exception occurs for those complex concepts (for example, the medical examinations van, above) requiring many decisions before the cost is known. Define the Respondent Group We would like to interview any and all persons who will play a role in deciding whether the product will be bought and how it might be improved. When the New Zealand Wool Testing Authority came up with a new wool testing service, it had to test the concept with three levels in its channel—brokers who sell the raw wool, scourers who scour the wool and prepare it for shipment, and exporters who sell the wool to manufacturers.18 A cement company, which created a new concept in cement for use in construction, had to seek advice from brick makers, siding makers, architects, builders, designers, and regulators, among others, in addition to the people who would be buying the buildings. Some industrial products may involve 5–10 different people at each buying point, and durable consumer goods usually involve more than one person. Yet that peanut butter mentioned above could probably be tested with just one person in a family setting—the household member who does the buying—or could it? The solution is to think in terms of stakeholders—any person or organization who has a stake in the proposed product. Our new product wastebaskets are filled with products that made sense to the end users but could not get to them—for example, when professional sanitary engineers refuse to endorse a new system of water treatment. 17 Brian Ottum, “Market Analytics,” in K. B. Kahn, S. E. Kay, R. J. Slotegraaf, and S. Uban (Eds.), The PDMA Handbook of New Product Development (Hoboken, NJ: John Wiley, 2013), Ch. 15, p. 260; see also Ned F. Anscheutz, “Evaluating Ideas and Concepts for New Consumer Products,” in M. Rosenau, A. Griffin, G. Castellion, and N. Anscheutz (eds.), The PDMA Handbook of New Product Development (New York: John Wiley, 1996), pp. 195–206. 18 Arch G. Woodside, R. Hedley Sanderson, and Roderick J. Brodie, “Testing Acceptance of a New Industrial Service,” Industrial Marketing Management 17(1), February 1988, pp. 65–71. 226 Part Three Concept/Project Evaluation Reaching this full set of influencers sounds simple, but it is complex and expensive. Some people try to seek a small number of lead users (see Chapter 4), or influencers, or large users. This approach saves some money and gets more expert advice but often fails to reflect key differences (and misunderstandings) in the marketplace. It would seem to be a technique for situations where there is a right understanding or perception or preference. Of course, we should always watch out for critics, people who have a reason for opposing the concept. A developer came up with a device that read electrocardiograms and needed the reactions of cardiologists, but the obvious conflict of interest made the interviewing tricky. Some new products people, aware that they will first have to interest the innovators and early adopters in a market, concentrate their concept testing solely on them. If this group is interested, it’s a good bet others will be also. Select the Response Situation There are two issues in the response situation: (1) the mode of reaching the respondent, and (2) if personal, whether to approach individually or in a group. Most concept testing takes place through personal contact—direct interviewing. Survey samples typically run about from 100 to 400 people, though industrial samples are usually much smaller. Personal contact allows the interviewer to answer questions and to probe areas where the respondent is expressing a new idea or is not clear. Examples earlier in this chapter show that mail contact is frequently used instead of personal contact, and firms have also used the phone and the Internet as other, less expensive means to conduct these tests. Some research suppliers offer a service of interviewing in which the client can submit product concepts on a shared-cost basis. In the Omnibus program at Moskowitz Jacobs, a fully equipped central testing facility conducts periodic waves of interviewing, which yield 100 interviews at a cost per concept of around $3,000. Other research firms use pseudo stores in vacant locations at shopping malls. The second issue concerns individual versus group. Both are widely used. Groups (in this case, focus groups) are excellent when we want respondents to hear and react to the comments of others and to talk about how the product would be used. Newer methodologies allow for almost instantaneous evaluation of a great number of product concepts. One such technique, the real-time response survey, combines the best features of focus groups and surveys and has proved useful in screening new consumer product concepts. Briefly, about 100 participants observe price, positioning, and attribute information about the concept, perhaps via a simulated ad. A moderator guides the respondents through a computer exercise, in which they use a keypad to input their purchase intentions, responses to proposed prices, and similar data using 11-point scales. The responses are sent to a central computer where they can be read directly, in real time, by the moderator and client. Based on these early results, the moderator can develop original open-ended questions and ask them while the respondents are still present. Responses to the open-ended questions might suggest whole new concepts or attribute combinations, which can then be further evaluated by the respondents. Response rates are Chapter Nine Concept Testing 227 virtually guaranteed; hundreds of questions can easily be asked in a three-hour session using the keypads; and dozens of concepts can be evaluated in a single session (thus reducing the number of sessions needed).19 Another similar technique now used in concept evaluation is to employ group support systems (GSS) software (see Chapter 5) in a focus group setting and to have the participants react to different versions of products. For the aerosol hand cleanser of Figure 9.2 as an example, different spray applicators, package sizes, effectiveness levels, and price points could be tried. The group’s responses can be averaged and immediately displayed at the front of the room, and good concepts can be selected and even improved upon. Prepare the Interviewing Sequence Simple interviewing situations state the new product concept and ask about believability, buying intentions, and any other information wanted. The whole interview may take only two or three minutes per product concept if the item is a new packaged good and all we really want is a buying intention answer. Usually we want more than that. In such cases, we first explore the respondent’s current practice in the area concerned, asking how people currently try to solve their problems, what competing products they use, and what they think about those products. How willing would they be to change? What specific benefits do they want? What are they spending? Is the product being used as part of a system? This background information helps us understand and interpret comments about the new concept, which are asked for next. The immediate and critical question is, “Does the respondent understand the concept?” Given understanding, we then seek other reactions: Uniqueness of the concept. Does it solve a problem? Believability of the concept. How much they like the concept. Importance of the problem. How likely would they buy? Their interest in the concept. Their reaction to the price. Is it realistic, practical, useful? Problems they see in use. We are especially interested in what changes they would make in the concept, exactly what it would be used for and why, what products or processes would be replaced, and who else would be involved in using the item. You can see that services present a problem here. A service offers an image, or a feeling, or a hard-to-measure convenience. This makes it difficult for the respondent to give useful information along the lines just listed. 19 Lynne R. Kahle, Douglas B. Hall, and Michael J. Kosinski, “The Real-Time Response Survey in New Product Research: It’s About Time,” Journal of Consumer Marketing, 14(3), 1997, pp. 234–248. 228 Part Three Concept/Project Evaluation In all this interviewing, remember we are not taking a poll but, rather, exploring what people are doing and thinking. Only a few questions will be in standard form for tabulation. Each new concept addresses a very specific problem (or at least it should), and we need to know what people think about that problem in the context of the new concept. It doesn’t pay to get too formal in the questioning, unless conducting many concept tests where there is a database for comparison. Variations There are variations to all these procedures. The above procedure assumed oneon-one contact with potential buyers. The real-time response surveys employing GSS, discussed earlier, can provide information on buying intentions efficiently as groups of customers respond to the product concepts presented to them. In another example, Avon markets 50 new products every 60 days, with a three-month development cycle. Every two weeks they meet with some of their test bank of 150 sales reps. Many ideas are shown to them by computer-driven projectors for their quick reaction. Their appeal rating correlates very well with sales, in some cases more accurately than field consumer concept testing predicted. A garment must be cut to fit a body, and bodies vary greatly. Note also that there are research firms that offer concept evaluation as part of their package of product development services. One example is the Inno Suite Concept Screener, a product of TNS (www.tns-global.com). Analyzing Research Results A great number of firms rely on a simple top-two-boxes score (or top-two-boxes plus 30 percent, based on industry experience) in doing concept testing. Occasionally, more information is needed. We cannot assume all customers will have the same needs or look for the same benefits when making a purchase. In fact, through benefit segmentation, a firm may identify unsatisfied market segments and concentrate its efforts on developing concepts ideally suited to the needs of these segments. We now turn our attention to ways in which we can identify benefit segments in our desired market and develop products that will be most preferred by key benefit segments.20 Identifying Benefit Segments Let’s return to the swimsuit example of Chapter 6. Recall that, when we were collecting respondents’ perceptions of the existing swimsuit brands, we also asked them to rate how important each attribute was in determining their preference among brands. These importance ratings can be used to model existing brand preferences and predict likely preferences for new concepts. 20 Note that, in our typology, benefits are one type of attribute (the others being features and functions). The terms “benefit segmentation” and “benefit segments” are commonly used for the procedure described in this section and should not imply that only benefit-type attributes can be considered. Chapter Nine Concept Testing 229 FIGURE 9.3 Importance Map Showing Benefit Segments Comfort Segment 3 Segment 2 Fashion Segment 1 Suppose there were only two attributes to consider: comfort and fashion. It might be very simple to identify benefit segments on an importance map as in Figure 9.3. Each customer is indicated by a dot in this figure, according to the importance she attaches to each of the two attributes. In this simple case, three obvious benefit segments emerge of approximately equal size: customers that think comfort alone is important, those that think fashion alone is important, and those that think both are important. Rarely are the benefit segments so easily visualized, however. In this case, like most others, there were many more than two attributes that are important to customers in forming their preferences. We need to turn to one of the many computer programs that can do cluster analysis, which puts observations (in this case, individuals) together into relatively homogeneous groups on an importance map. Like factor analysis, cluster analysis is also a data reduction method. In Chapter 6 we learned that factor analysis reduces the data cube by grouping many attributes together into a small number of underlying factors; cluster analysis groups together many individuals into a small number of benefit segments. Different criteria and rules of thumb can be used to select the best number of clusters (benefit segments) that exist in the market since there is no one correct answer. Generally, practical judgment or experience play an important role. For example, in this case we may feel it is unlikely that more than five or six benefit segments exist. When cluster analysis was run on the swimsuit importance data, a satisfactory solution with three benefit segments was obtained. Conceptually, it is not too different from Figure 9.3, even though we considered many more attributes: The three clusters we obtained more or less correspond to those depicted in that figure. 230 Part Three Concept/Project Evaluation Joint Space Maps We can now overlay the benefit segments onto our perceptual map (previously built in Chapter 6). The result is called a joint space map, and it allows us to assess the preferences of each benefit segment for different product concepts. Joint space maps can be developed using ideal brand ratings or preference regression. Ideal Brands The most direct way is to get customers to rate their ideal brand on each attribute. Using the factor score coefficient matrix (which we obtained in Chapter 6 from the perceptions of existing brands), we convert the ideal brand ratings to factor scores and plot the ideal brand positions directly on the perceptual map. Clusters of individuals may be detected visually from this map—each cluster represents a segment with its own ideal brand positioned at the center of the cluster. Figure 9.4 shows what a joint space might look like if three segments existed as shown in Figure 9.3. The preferences of each segment can be obtained from Figure 9.4. We expect the brand that is located closest to a segment’s ideal brand will be preferred by that segment. Market share estimation models often assume that the market shares obtained by the various brands are inversely proportional to the square of the distance of that brand from the ideal point: This technique makes brands very close to the ideal highly preferable. In Figure 9.4, Segment 1 is likely to prefer Sunflare, while Segment 2 seems to be satisfied with either Aqualine or Islands. The brand nearest to Segment 3’s ideal point is Molokai, but none of the brands is really that close. Thus, a new brand high in both fashion and comfort has a chance to draw substantial market share from competitors. FIGURE 9.4 Joint Space Map Showing Ideal Points Comfort 5 Aqualine 3 2 Islands Molokai 1 2 3 Splash 2 1 Numbers along the axes represent factor scores. 4 5 Sunflare Fashion 1 Chapter Nine Concept Testing 231 FIGURE 9.5 Benefit Segment Profiles Segment Experience Seekers Benefits Need to haul people and belongings Good performance Practical Safe Preferred Vehicle Preferred Way to Seek Car Information Male/Female Median Age Children Median Income Pragmatic Performance Seekers Affordable Performance ** * * Safety Conscious ** * ** * SUV Hybrid Visit Read dealerships Consumer Reports 50/50 35/65 40 49 80% 60% $70K $60K ** Sedan Luxury Performance Visit dealerships Performance Web and dealerships Web 75/25 42 30% $85K 65/35 33 20% $35K 35/65 40 50% $60K Note: **This cluster’s factor score is very high for this benefit. *This cluster’s factor score is relatively high for this benefit. Source: Adapted from Brian Ottum, “Segmenting Your Market So You Can Successfully Position Your New Product,” in A. Griffin and S. M. Somermeyer, The PDMA Toolbook 3 for New Product Development, John Wiley & Sons, Inc., 2007, Ch. 7. Reprinted with permission of John Wiley & Sons, Inc. Figure 9.5 provides a fully worked-out benefit segmentation of the car-driving market. At the left are four benefits identified through factor analysis: need to haul people and belongings, good performance, practical, and safe. The figure shows that five segments were identified: 1. Experience Seekers: While performance and safety are important, they really care about hauling lots of stuff. 2. Pragmatic: They care mostly about practical transportation. 3. Performance Seekers: They seek high-performance cars only. 4. Affordable Performance: They care about performance but also about practicality. 5. Safety Conscious: Only the safety benefit is important. What the figure then shows is how all of this information is put to use by management. Additional rows in the figure suggest the kinds of cars each of these segments might prefer, show how each segment tends to get information about car purchases, and provide key segment demographics. Information like this is very useful to managers in developing ideal new products for targeted segments, and also for making positioning decisions (to be taken up later in Chapter 16).21 21 The car example is adapted from Brian Ottum, “Segmenting Your Market So You Can Successfully Position Your New Product,” in A. Griffin and S. M. Somermeyer, The PDMA Toolbook 3 for New Product Development (New York: John Wiley, 2007), Ch. 7. 232 Part Three Concept/Project Evaluation Preference Regression Preference regression is another method that can be used to identify the optimum combination of attributes desired by the market. This method relies on a different kind of numerical input—often, rankings of brands are obtained (paired comparisons can also be used). In preference regression, we do a regression analysis to relate the factor scores of each brand to the rankings of brands. The relative sizes of the regression coefficients we obtain give us an indication of the relative importance of each factor. Preference regression can also be done on attribute ratings instead of factor scores. Instead of the importance ratings we discussed above, assume that customers were asked to provide rank orderings of the five existing brands, where 1 5 most favored, and 5 5 least favored. First, we reverse scale the rank orderings such that higher numbers represent more favored brands. Then we can solve the regression equation: (reversed) preference rank 5 b0 1 (b1 3 attractiveness score) 1 (b2 3 comfort score) 1 e If we ignore benefit segments and put all the respondents together, we find the values of b1 and b2 are 0.28 and 0.21. Thus, the relative importance of fashion to customers is 0.28/(0.28 1 0.21), or 57 percent, and the relative importance of comfort is 43 percent. Hence, while fashion comes out as the more important factor, we cannot ignore the fact that customers also place a lot of importance on comfort when we are assessing product concepts. We can also plot the regression line on the perceptual map as shown in Figure 9.6.22 This line is referred to as the FIGURE 9.6 Comfort 5 Joint Space Map Showing Ideal Vectors Aqualine 4 Molokai Islands 2 3 Splash 4 X 5 Fashion 1 fit ne Be ent 1 ll gm Overa Se Benefit t2 Segmen Sunflare 2 1 Numbers along the axes represent factor scores. 22 Though we actually do estimate b0 in the regression equation, we ignore it when drawing Figure 9.6, as we are most interested in the relative importance of the revealed weights b1 and b2. The b0 term simply defines the scale. Thus, in Figure 9.6, the regression line is shown passing through the origin. The e is an error term. Chapter Nine Concept Testing 233 ideal vector, as it visually represents the optimum proportion of attributes desired by this market. A product concept lying near the regression line at Point X on the map is in a desirable position for this market. We can also cluster analyze the rank orderings to get benefit segments. In this case, two benefit segments were found to exist in this market, represented by the two lighter lines in Figure 9.6. One of these appears to consider only fashion (the relative importance of this factor is 94 percent), while the other considers a blend of fashion and comfort (the relative importances are 30 and 70 percent, respectively). Product concepts aimed at one or the other of these segments may do better than the concept represented by Point X, which in fact may not directly appeal to either segment. Conjoint Analysis in Concept Testing We were first introduced to conjoint analysis in the context of concept generation. In actuality, conjoint analytic techniques are extremely useful in concept testing as well and are frequently used at this point. In the conjoint analysis of Chapter 7, you had assumed the role of product manager for a line of salsas. We selected three key attributes of salsa and two or three levels for each attribute, and used conjoint analysis to identify high-potential gaps: combinations of attributes that (a) customers like and (b) are not on the market yet. Without going through the quantitative analysis again, it should be clear how conjoint analysis can be used at concept testing. The model identified the levels of the key attributes that are preferred by customers and rank ordered the possible combinations from most to least preferred. Each of these combinations could be thought of as a concept, and the top-ranking concept or concepts are the ones that hold the highest potential and should be considered for further development. Of course, the model also identified the real losers! As was mentioned earlier, many attributes and levels can feasibly be tested in conjoint analysis using a reduced set of cards. The most preferred concept(s) will still come out ranked on top, even if they were not included in the original set of cards. Overall, conjoint analysis is extremely useful in concept testing because of its ability to uncover relationships between attributes (features, functions, benefits) and customer preferences, as illustrated in the salsa example. We had used a set of product description cards as the stimuli in our original example, since we were at the very earliest phases of the new products process. Note, however, that conjoint analysis can easily use concept statements in other forms as stimuli. At concept testing, we may have concept statements in any of the forms discussed earlier in this chapter (verbal narratives, drawings, sketches, models or prototypes, even virtual-reality representations). The analysis would proceed in the same way regardless of the stimuli used. To illustrate: Conjoint analysis was used to evaluate how well drivers in New York and New Jersey would respond to the EZPass electronic toll collection system.23 With the EZPass system, drivers attach an electronic “tag” to their 23 Terry G. Vavra, Paul E. Green, and Abba M. Krieger, “Evaluating EZPass: Using Conjoint Analysis to Assess Consumer Response to a New Tollway Technology,” Marketing Research, Summer 1999, pp. 5–16. 234 Part Three Concept/Project Evaluation windshield. The tag is read using high-speed radio waves as the car passes through the toll lane and the toll amount is deducted from the driver’s prepaid account. The tag can be read while the car is moving; thus EZPass eliminates the need for the driver to stop completely to pay the toll and ultimately should result in fewer traffic tie-ups at toll plazas. The transportation authorities had already decided to adopt EZPass, but needed assistance on how it should be designed in order to meet driver needs. Seven key attributes were identified by the transportation authorities: • • • • • • • The number of EZPass accounts a user would need to open. How to apply and pay for an account. The number of EZPass lanes available at each toll plaza. Transferability of the EZPass tag to another vehicle. Acquisition cost and/or service charge (if any) for the tag itself. The toll price with EZPass. Other uses for EZPass, such as airport parking or gasoline purchase. Because EZPass was a new product concept for most drivers at the time of the study (1992), the concept description took the form of an 11-minute video demonstration of the system “in action” and its effectiveness in relieving toll plaza congestion. Respondents were mailed a copy of the video together with a questionnaire and eight scenario cards (each showing a different combination of the above attributes). The conjoint analysis revealed that the most important attribute by far was the number of lanes available and how they would be controlled, while price of the toll, application procedure, and acquisition cost were also relatively important. The analysis also revealed which levels of each attribute were preferred. For example, the four options provided for acquisition cost were rated in order as follows: 1. 2. 3. 4. $10 deposit, plus $15 yearly service charge (most preferred). $2 per month service charge. $10 charge plus $1.50 per month service charge. $40 credit card charge if tag not returned, plus $20 annual fee. The results from the conjoint analysis were used to design the implementation plan for the EZPass system. The system was adopted rapidly by drivers in New York and New Jersey. By 1999, rush hour use of EZPass had reached 60 percent, about 2 million drivers in the two states that were enrolled in the program, and about 3.1 million tags had been distributed. Benefit segments can also be identified in conjoint analysis. Recall that conjoint identifies each customer’s value system, that is, the relative importances of the attributes to each customer and the preferred levels of each attribute. We took a shortcut in Chapter 7 by assuming that all customers had about the same value system, so we identified the medium-hot green salsa as the best combination. As we have seen earlier in this chapter, however, there may be underlying benefit segments. We noted in Chapter 7 that aggregating all customers may disguise the fact that half of the market might like extra-hot salsa, and half the market Chapter Nine Concept Testing 235 might like mild. We can apply cluster analysis techniques to the importances and preferences generated by conjoint analysis to identify benefit segments of customers who have similar value systems. For example, in the industrial service example of Chapter 7, price came out as the most important variable (with a relative importance of about 27 percent) when all the respondents were aggregated. Follow-up cluster analysis revealed as many as five benefit segments, which varied widely with respect to the importance ascribed to price. In one segment, more concerned with performance quality, price’s relative importance was under 9 percent, while in a second, price-driven segment, the comparable figure was about 35 percent!24 The benefits sought by potential subscribers to TrafficPulse System, by Mobility Technologies Inc., were assessed using a variation of conjoint analysis. This system allows subscribers to get information on traffic conditions, travel times, and preferred routes. The analysis found five benefit segments, differing in their levels of interest in a personalized system, a voice/wireless system, and an Internet-only system.25 Market Research to Support Concept Testing There are a few well-known research firms that support the concept testing phase. One of the best known is BASES, part of the A. C. Nielsen company and briefly mentioned earlier. BASES helps firms evaluate and optimize new product concepts and also predict sales, operates worldwide, and studies over 10,000 new product ideas per year. BASES provides three levels of concept testing studies.26 Pre-BASES is a concept test that provides rough sales predictions. BASES I is a more advanced concept test that incorporates media selection decisions, levels of consumer and trade promotion, and extent of distribution to estimate awareness and availability, and can attain a forecasting accuracy range of 25 percent. BASES II combines the concept test with a customer taste test and post-taste customer responses, and achieves an accuracy range of 20 percent. Nestlé Refrigerated Foods relied on BASES for support during concept development of their Contadina refrigerated pasta and pizza products. To concept-test a line of refrigerated pastas and sauces, a BASES I methodology was used. Preliminary study assessed the appeal of the refrigerated pasta concept and current levels of customer satisfaction. Then the concept was tested among 300 adult female respondents. Each stated a purchase intention and also assessed what they liked, 24 Y. Wind, J. Grashof, and J. Goldhar, “Market-Based Guidelines for Design of Industrial Products,” Journal of Marketing, July 1978, pp. 27–37. 25 Abba Krieger, Paul Green, Leonard Lodish, Jim D’Arcangelo, Chris Rothey, and Paul Thirty, “Consumer Evaluations of ‘Really New’ Services: The TrafficPulse System,” Journal of Services Marketing, 17(1), 2003, pp. 6–36. 26 This section is adapted from the Nestlé Refrigerated Foods: Contadina Pasta and Pizza (A) case by V. Kasturi Rangan and Marie Bell, Case no. 9-595-035 (Cambridge, MA: Harvard Business School Press, 1995). 236 Part Three Concept/Project Evaluation FIGURE 9.7 Summary of Concept Test Results— Refrigerated Pasta Total (%) Likes General variety Filled variety Natural/not artificial Quick/fast/saves time Easy to prepare/already prepared Good/reasonable price Fresh/made fresh and dated Favorable (%) Unfavorable (%) 28 16 28 20 17 8 26 28 16 30 22 20 9 27 28 16 23 16 11 4 21 Dislikes Too expensive Not like green/spinach color Not like spinach taste 8 6 3 3 5 2 23 11 5 Concept Uniqueness Extremely new and different Very new and different Somewhat new and different 15 38 35 17 41 32 8 32 41 Explanation: These are the results of the concept test for the refrigerated pasta product. The figure reports the percent of respondents agreeing with each statement, broken down by (1) overall percent, (2) percent of respondents who are favorable toward the product concept, and (3) percent of respondents who are unfavorable toward the product concept. For example, 20% of those who liked the concept thought it was easy to prepare, while only 11% of those who disliked the concept thought it was easy to prepare (the overall average was 17%). The statements are organized into likes, dislikes, and statements of uniqueness. Source: V. Kasturi Rangan and Marie Bell, “Nestlé Refrigerated Foods: Contadina Pasta and Pizza (A),” Case no. 9-595-035, Cambridge, MA: Harvard Business School Press, 1995. disliked, and found unique about the product. A summary of the results is found in Figure 9.7. The refrigerated pasta earned a 75 percent top-two-boxes score (24 percent stating “definitely would buy” and 51 percent stating “probably would buy”). An advantage of using BASES is that it can compare these figures to similar products in its database as a rough benchmark. In this case, the median top-two-boxes score for similar products was 61 percent, so these results are encouraging so far for the pasta. The respondents were then split into two groups: favorable (the 75 percent with positive purchase intentions) and unfavorable (everyone else). Both groups liked the same things about the new product: It’s natural, it offers variety, it’s fresh, it saves time, and it’s easy to prepare. Of those who were unfavorable to the product, they most often mentioned price. Although not shown in the table, BASES also tested three different product positioning statements for the concept: Homemade (positioned to match homemade taste and quality), Pasta Dinner (a hearty-enough dinner to satisfy meat-andpotato cravings), and Superior (a new line that is better than any pasta or sauce you have ever tried). The Superior positioning was slightly preferred over the other Chapter Nine Concept Testing 237 two on the attributes shown in Figure 9.7 (more likes and fewer dislikes), and was therefore selected for further consideration. Next, the raw top-two-boxes results were converted to an A-T-A-R sales forecast. BASES calls on its database to convert statements of “definitely will buy” and “probably will buy” to actual purchase behavior. While this information is proprietary and varies across industries, we will apply a simple rule of thumb: We expect that 80 percent of the “definitely” and only 30 percent of the “probably” will actually buy. Our adjusted trial is therefore: (0.8 3 definitely) 1 (0.3 3 probably) 5 (0.8 3 24%) 1 (0.3 3 51%) 5 34.5% The next two components of the A-T-A-R model, awareness and availability, are assessed using managerial input. To keep the example simple, Nestlé is planning to spend $13 million on advertising, which is enough to achieve 48 percent awareness, and they are funding an intensive distribution strategy in which the product will be available to 70 percent of the population. We now have the first three parts of the A-T-A-R model: (awareness 3 trial 3 availability) 5 0.48 3 34.5% 3 0.70 5 11.6% The total number of target households was 77.4 million, so the number of trial households is 11.6% 3 77.4 million 5 9 million. Finally, repeat rate is assessed at 39 percent based on similar products, with the average customer repeat purchasing 2.5 times, and with 1.4 units bought per purchase occasion. (This methodology is a little different from how repeat was calculated in earlier examples, but it makes sense for small, frequently purchased packaged goods such as this one.) Repeat therefore is calculated to be: 39% 3 2.5 3 1.4 5 136.5% Putting it all together, BASES predicts total sales to be: 9 million 3 136.5% 5 12.3 million Nestlé also had the greatest amount of uncertainty in the 39 percent repeat rate, so BASES redid the calculations with a worst-case and best-case repeat rate. Even with a mediocre 27 percent repeat rate, the sales forecast still reaches 8.5 million units. To support the launch of the follow-up product, Contadina refrigerated pizza and topping kits, a BASES II methodology was employed. This situation was a little different, because customers will be already familiar with two alternative product forms: frozen pizza and takeout pizza from the local shop. It was important, then, to determine if customers thought this new product was anything different or special relative to more familiar choices. First, top-two-boxes scores were obtained from a sample of about 600 respondents. The pizza-and-topping combo scored 76 percent, while a pizza-only concept scored only 58 percent, so the choice was made to move the pizza-and-topping concept forward. As noted above, the distinctive feature of BASES II is that customers actually try the product (in this case, in an in-home use test). At this point, the results gave some cause for concern. Respondents were asked how “new and different” the pizza-and-topping concept was compared to takeout or frozen pizza. Before use, the top-two-boxes 238 Part Three Concept/Project Evaluation score (“extremely new and different” plus “very new and different”) reached 59 percent; this declined to 49 percent after use, suggesting some disappointment among the customers. Afteruse attribute testing identified a list of improvements as suggested by the respondents: Improve the overall taste and lower the price were the most important. With this information obtained from BASES II, Nestlé is in a position to decide whether to approve the product for launch, run more extensive testing (we will discuss market testing methods later in this book), or drop the concept altogether.27 Conclusions The advantages of concept testing and development prior to full screening are many. It can be done quickly and easily, gives the screeners invaluable information for the sorting out of less valuable concepts, proves market research technology exists, is reasonably confidential, helps us learn a lot about buyer thinking, and enables segments and positionings to be developed in tandem with the concept. Unfortunately, some developers (especially industrial designers) still refuse to do concept testing. Herman Miller, for example, was unable to market successfully a Hygiene System that incorporated a toilet, sink, and tub. It had not been concept tested, and after it failed, the designer claimed that industry people still did not understand it. Nevertheless, concept testing is a bit treacherous—mistakes are easy and can be costly. It is not a tool for amateurs. There have been classic flops, most of which passed concept tests—dry soups, white whiskey, clear soda, and so on. The original chewable antacid tablet floundered because the concept test missed the idea that people then wanted water with antacids. One firm studied executions of a single new product idea by three copywriters and found that the most important determinant of high scores in the concept test was the skill of the copywriter. People find reacting to entirely new concepts difficult without a learning period; the stimulus of a concept statement is very brief; many situation variables will change by the time the product is marketed; and certain attributes cannot be measured in a concept test—for example, rug texture, shower nozzle impact, and what color will be “in” next season. Perhaps most troublesome, the technique has just enough slippage in it that persistent product champions often argue successfully against its findings. Summary This was the first chapter covering the tools used to evaluate new product proposals. Because evaluation actually begins prior to ideation (that is, deciding where to seek ideas), we first looked at the product innovation charter. By focusing the creative activity in certain directions, the charter automatically excludes all other directions and thus, in effect, evaluates them negatively. 27 Interestingly, Nestlé decided to launch the pizza-and-toppings product without further market testing, but it did poorly. The name Contadina was sold to Del Monte Foods a few years later, and Nestlé is still in the refrigerated pasta and sauce business (under the Buitoni name), but not in the pizza business. Chapter Nine Concept Testing 239 Once the strategic direction is clear, most firms undertake a market analysis of the opportunity described by it. The customer should be a major input to any product innovation program, and immediately after strategic decisions have been made is an excellent time to seek this input. Then, as the ideas begin to roll in, an initial response is made—highly judgmental, quick, and designed primarily to clean out the worthless ideas. Once an idea passes that test, more serious evaluation begins. The tool at this point is concept testing, or concept development, which now has a lengthy history of successful use. The chapter gave the overall procedure for concept testing, including its purposes, options in concept format, respondent selection, and the interviewing procedure. An immediate benefit of concept testing is that it gives management the information needed to make the judgments required by the scoring models used in the following step: the full screen of the concept, which is the subject of Chapter 10. Applications 1. “You know, most of our new products people do a great deal of marketing research—concept testing, attitude surveys, and the like. But let me read something that one automobile designer thought about marketing research.” (She then read from a yellowed clipping on her desk.) Market research is probably the greatest single deterrent to excellence in modern business. It’s a crutch for managers with no vision and no conviction. On the surface, it sounds sensible enough: Find out exactly what the buyers want before you come to a design. But in practice, it’s impossible. The public doesn’t know what it wants without being shown the choices, and even then, preference is apt to veer off in the direction of Kmart. Market research gives you Malibus with Mercedes grilles, refrigerators in avocado hues, and Big Macs with everything. You do not, however, produce greatness with this technique.28 “Perhaps you would comment on that statement.” 2. “A cosmetics competitor is trying to speed up its new product work on lipsticks by a system that uses (1) brainstorming to create ideas (392 in a recent session); (2) evaluation of those ideas by the same group of people, down to only the best 50 ideas; and then (3) focus group sessions for concept testing those ideas down to the few that should be developed rapidly. Do you see anything wrong with this system?” 3. “I would be curious to test your personal judgment on some new ideas from one of our recent idea sessions. They were all accepted in later concept tests with consumers, and that concerns me. Are we safe to go ahead? a. A gasoline-powered pogo stick. b. A combination valet stand and electric pants presser. c. Transistorized golf balls and an electric finder. d. An arm wrestling device so you can arm wrestle with yourself. e. An electrically heated bath mat. f. Chocolate candy in an edible chocolate box.” 28 “The Best Car in the World,” Car and Driver, November 1979, p. 92. 240 Part Three Concept/Project Evaluation Case: Domino’s29 By 2009, Domino’s had a track record of nearly 50 years of fast, dependable pizza delivery and claimed to be the “World Leader in Pizza Delivery.” From a single pizza store in Ypsilanti, Michigan in 1960, founder Tom Monaghan grew the business to over 8,000 stores by 2006, about 10 percent being company owned and the remainder franchised. This total included about 3,000 international stores in over 70 countries. Domino’s is one of the leading companies worldwide in online transactions as well since instituting online ordering in 2007. It had also launched a line of oven-baked sandwiches in 2008, immediately launching it into the number one position in sandwich delivery. But not all was well within Domino’s. Their promise of easy ordering and 30-minute delivery might have been their ticket to success in the early years, but by 2009 it was clear that customer were demanding something more: better taste. Customer testing at the time, including focus groups, alarmed Domino’s management, who heard customers complaining that the “pizza was cardboard” or that it tasted “wet and flavorless.” Lost-buyer analysis suggested that the main reason Domino’s lost customers was because of the menu, and the pizza in particular. In 2008, Domino’s tried an ad campaign stressing the many years of reliable delivery service: the slogan “You got 30” reminded viewers of the 30-minute delivery promise. When the campaign failed, it was clear that action was required. Domino’s senior executive J. Patrick Doyle viewed the negative customer comments and the failed ad campaign as a new product challenge. Domino’s initiated the “pizza turnaround,” led by Doyle and two other senior managers, Russell Weiner and Brandon Solano. The goal of the pizza turnaround was not just to improve the flavor of Domino’s pizza. The CEO at the time, Dave Brandon, said that “incrementally better” would not be good enough. According to Doyle, the goal was to completely reformulate the flagship product from the ground up so that it could actually beat competitors in a taste test. The new product team started by rethinking an unstated assumption: Domino’s was viewing good quality and quick delivery as trade-offs. As Doyle said, “There is no reason that we can’t deliver terrific food and do it in the same amount of time.” To achieve this, however, required Domino’s to rethink their platform for pizza innovation. Traditionally, pizza product development had been very incremental: new toppings, for example. But at this point, the success of the oven-baked sandwiches in the previous year was fresh in the minds of the new product team. Not only were the sandwiches a radical, new-to-the-firm innovation for Domino’s, they were also developed in record time and were successful. It was the success of the new sandwich line that emboldened the new product team to undertake the radical innovation needed with pizza. It was decided to take on the challenge of completely rethinking the pizza product. Based on early consumer research, Solano says the company began asking, 29 This case is adapted from Greg Githens, “Domino’s Pizza Reinvents Itself: The Story-Behindthe-Story of the New Product Launch,” Visions, 34(4), 2010, pp. 10–13, and from www.dominos.com and www.pizzaturnaround.com. Chapter Nine Concept Testing 241 “Could we make a better cheese? Dough? Sauce? What made a sauce better? What characteristics of cheese made it better?” This led to further consumer tests of different combinations of attributes. Solano explains the process: We had three components (three crusts, three sauces, two cheeses) yielding 18 pizzas. We tested many of these combinations and modeled the ones we did not test. We had a winner coming out of this . . . it was the product with all our new components. We did it! Further testing confirmed that the favored product was viewed as not just better and different, but much better. Cost was also a consideration, according to Weiner: “The pizza had to be significantly better, but it could not cost more or take longer to make [at the store]. There were financial and operational parameters.” The new products team knew that they would need to make the case to top management. Internally, they knew there would be opposition from some managers who felt the product was just fine. Solano said that they would need to be prepared to face senior management, but that unlike many other companies, they could informally “walk right in and talk about what we just learned.” When R&D personnel were shown the customer complaints, they were alarmed and recognized that this time, product development could not be “business as usual.” Weiner also noted that they would need to convince franchisees. He said, “We did research pretending we were the franchisees . . . for example, the sauce is spicier. [Franchisees might ask], could that make people sick after eating a lot? . . . So we did a road show with franchisees. They did a blind taste test; most of them liked the new better.” A major part of the launch campaign was a four-and-a-half minute promotional “pizza turnaround” video, launched on the website www.pizzaturnaround.com and also available on YouTube. The video showed extremely negative customer comments and clips from focus groups, then went into detail on how the company completely redesigned the crust, sauce, and toppings. The video emphasized that some companies might hide customer criticism, but Domino’s acted on it to improve the product. Doyle noted that it might have been risky to talk about negatives in their rollout promotional campaign, but as Weiner said, “we . . . did a lot of testing to get the right balance of negative to positive [comments]. If we were too negative we got a poor result and if we were too positive we got a poor result.” The pizza turnaround was a stunning success. Taste was improved to the point where Domino’s claims that three out of five people prefer Domino’s to competitors. Soon after launch, new customers increased by 30 percent and repeat purchase was up by 65 percent, indicative of excellent customer loyalty. These figures translated into an increase of 14.3 percent in quarterly same-store sales, even during a U.S. economic slump. What do you think caused the crisis within Domino’s that led to the pizza turnaround decision? Comment on how the new product team at Domino’s used attribute analysis to test various improved-pizza concepts. In your opinion, how important was this analysis to the team in selling the pizza turnaround internally? What else did the new product team do to boost their chances of success? What can be learned in general from this case about the use of analytical attribute approaches for concept testing? How could other concept testing approaches presented in this chapter have been used to guide the new product team? 242 Part Three Concept/Project Evaluation Case: Comparing Smartphones (B)30 Refer back to the Comparing Smartphones (A) case at the end of Chapter 6. In addition to the competitive information made available there, assume that you have commissioned additional customer research. You have gathered customer preference data, and this data set was used to identify “ideal brands” and assess the number and size of customer benefit segments in the marketplace. (For this study, consider only screen size and talk time as the most important attributes). Three segments were identified. Segment 1 (about 20 percent of the market) prefers the largest screen size and doesn’t care much about talk time; Segment 2 (about 30 percent of the market) likes long talk times but doesn’t care much about screen size; and Segment 3 (about 50 percent of the market) values a combination of the two attributes. Draw a positioning map for these two attributes, using the data from case (A). Which are the most serious competitors in each segment? What are the competitive implications? 30 See the Comparing Smartphones (A) case in Chapter 6 for further information. The customer preference segments are for illustrative purposes only. C H A P T E R T E N The Full Screen Setting Some years back, printer manufacturer Lexmark International had instituted a new products process and reported some success with it: the company experienced increased ability to launch successful products on time and within budget, and had also aligned its business and technical processes. Even so, management felt there was still room for improvement. In general, the most frequent problem was that too many projects were still getting through without adequate resources, and management was unable to prioritize the product projects effectively.1 This problem is all too common: Managers often say that concept selection is one of their biggest challenges. All the product projects that have made it this far in the process have cleared all the hurdles and look promising. But there are not enough financial or human resources to go around, so what should be done? Too often, managers lack a good selection procedure and do one of two things: guess (and probably select the wrong project), or approve everything (and consequently underfund every project). We know that project selection can be quite a challenge, even for top product developing firms. In a recent benchmarking study of product managers, over half of managers reported that they had too many projects in their portfolios given available resources, and almost half noted that they were managing portfolios of more than 50 products.2 These results suggest that managers have difficulty with concept selection, resulting in too many projects getting into the pipeline. We tackle concept selection in this chapter by introducing you to the full screen. In this chapter, we cannot present what any particular firm should do. That’s up to the new products manager. But we can present the range of alternatives and a middle ground that actually fits most firms. It can easily be modified. See Figure 10.1 for how screening relates to concept testing and the protocol step that follows from it. 1 Ed Crowley, “Building a Gated Product Development Process at Lexmark International,” Visions, 29(4), October 2005, pp. 22–23. 2 Carrie T. Nauyalis and Maureen Carlson, “Portfolio Pain Points: New Study Reveals That Companies Are Suffering from a Lack of Streamlined Product Portfolio Management Processes,” Visions, 34(1), 2010, pp. 13–18. 244 Part Three Concept/Project Evaluation FIGURE 10.1 Development Flow of New Product Concepts through Screening and Protocol Recycle Full screen Concept testing Project 1 Project 2 Write protocols Rank Project 3 Project 4 Reject Purposes of the Full Screen Recall where we are in the product innovation process. After the original idea emerged, we put it into concept format and then gave it a brief initial exposure for reaction by key players. Concept testing then enabled us to add the thoughts of potential users to the set of market and other data collected since the time of the product innovation charter. Along the way, we have been compiling the inputs of key functional people in the firm—technical, marketing, financial, operational, and so on. This work (which is situation dependent and may take a couple days up to several months) culminates in a step called the full screen. The term “full” here means that we now have as much information as we’re going to get before undertaking technical work on the product. The full screen often involves the use of a scoring model, which is an arrangement of checklist factors with weights (importance) on them, though we will see some variations in this chapter.3 Why do the full screen? Actually, the full screen accomplishes three objectives. First, it helps the firm decide whether it should go forward with the concept or quit. Keep in mind that if a concept passes the full screen, the next phase in the new products process is development. The concept will become a new product development project and will require a serious increase in commitment of financial and human resources. The full screen helps us decide whether these resources (R&D personnel, systems design for services, engineering, etc.) should be devoted to the project and, if so, how vigorously. This decision rests on whether we can do the job and whether we want to do it. “Can do” means feasibility—is technology up to the task, do we have it, can we afford it? “Want to” means will we get out of the project the profits, market share, or whatever it is we are doing product innovation 3 For a discussion and comparison of many of the most common full screen techniques, see K. L. Poh, B. W. Ang, and F. Bai, “A Comparative Analysis of R&D Project Evaluation Methods,” R&D Management, 31(1), January 2001, pp. 63–75. Chapter Ten The Full Screen 245 for? Sometimes these are called feasibility of technical accomplishment and feasibility of commercial accomplishment, and assessing these two types of feasibility (often through a scoring model) is central to most full screens. Second, the full screen helps manage the process by sorting the concepts and identifying the best ones. The best of the concepts can be rank ordered or prioritized such that we have some options on standby when an ongoing project stalls or is canceled, while unacceptable, but possibly worthwhile, concepts get cycled back into concept development where more work may make them acceptable. Further, a record is kept of rejected concepts to prevent reinventing the wheel when a similar concept comes up again later. This latter may seem a trivial point, but to managers who screen hundreds or even thousands of new product concepts a year, it is not trivial. A good corporate memory helps settle arguments later. An old adage says that a winning new product finds scores of “parents” who proposed it, whereas a losing new product is always an orphan. In firms that like to reward creativity, it helps to know who suggested what, and when. Third, the full screen encourages cross-functional communication. Scoring sessions are peppered with outbursts like: “Why in the world did you score that ratchet idea so low on such-and-such a factor?” The screening process is a learning process, particularly in making managers more sensitive to how other functions think. And it flushes out all basic disagreements about a project (including the everpresent politics) and sets them up for discussion. These disagreements put the spotlight on “potholes” or hurdles that the concept will face during development and show where new people may be needed. Many firms have difficulty with the full screen. They either select the wrong projects, or select too many projects. Inefficient screening means that financial resources and new product people are spread out over too many projects. New project approval should be made with human and financial resources and constraints in mind.4 Some firms bypass the full screen. Smaller firms not doing much new product work may prefer what really is an opinion poll where one or more people make a judgment on some informal checklist.5 In some cases participants may have a printed list of evaluation points as memory joggers, taken from the more formal lists that follow. Some packaged goods firms whose development process is rather nontechnical (me-too products and simple variations on what is already on the market) also may skip the full screen. Technical feasibility and the firm’s ability to market the product are already known, and the only issue is whether consumers will like the product if it were marketed. To compensate for the lack of a full screen, these firms may do a more complete concept test (Chapter 9) and what they call premarket testing sales forecasting models, which we will meet in Chapter 11. When there are major issues of technical feasibility (and more often than not, there are), even the packaged goods firms won’t depend just on concept testing and will rely on a full screen employing a scoring model as seen below. 4 Robert G. Cooper, “Your NPD Portfolio May Be Harmful to Your Business Health,” Visions, April 2005, pp. 22–26. 5 Even some very capable firms feel they can’t answer the issues in the more complete scoring models shown later. One unit of AT&T uses: Do customers care? Do we care? Can we do it? and Can we stay ahead if we do? 246 Part Three Concept/Project Evaluation The Scoring Model Scoring models are simple but powerful things. Let’s look at them through the eyes of a student who has a decision to make. Introduction to Scoring Models Assume a student is trying to decide what social activity to undertake this weekend. The student has several options, and more options may appear between now and then. The student could list criteria on several decisions that are personally important, specifically: 1. 2. 3. 4. It must be fun. It must involve more than just two people. It must be affordable. It must be something I am capable of doing. These four criteria (commonly called factors, but don’t confuse these with the factors we discussed in factor analysis) are shown in Figure 10.2. Of course, 20 or 30 factors might be involved in this student’s weekend social decisions, but let’s stick with the four. These factors are not absolutes; they can all be scaled— some fun, lots of fun, and so on. Figure 10.2 shows a four-point scale for each factor. Next, each scale point needs a number so we can rank the options. With that done, the student can proceed to evaluate each option (as indicated in Figure 10.2) and total the score for each. The final answer is to go boating—even though it isn’t quite as much fun—primarily because it can involve lots of people, it is cheap, and the student is a capable rower. FIGURE 10.2 Scoring Model for Student Activity Decision Values Factors 4 Points 3 Points 2 Points 1 Point Degree of fun Number of people Affordability Student’s capability Much Over 5 Easily Very Some 4 to 5 Probably Good Little 2 to 3 Maybe Some None Under 2 No Little Student’s scorings: Fun People Affordability Capability Totals Answer: Go boating. Skiing 4 4 2 1 11 Boating 3 4 4 4 15 Hiking 4 2 4 3 13 Chapter Ten The Full Screen 247 But suppose the student protests at this point and says, “There’s more to it than that. If I go hiking, I’ll get more exercise; but if I go skiing, a certain person is apt to be there.” Or the student may argue that affordability is more important than the other factors because without enough money, there is no need to score the other points. Or the student may say, “Having fun is really more important than skill, so let’s double the points for fun.” (That would be a kind of simple weighting system, in which the fun points are doubled before adding. Check: does that make any difference in the final recommendation? Why or why not?) And then there may be many objections, such as “skiing really is not all that much fun, boating is more expensive than you think.” A scoring process is what we actually use in making decisions like this, whether we realize it or not. The student’s objections contain the basic problems of new product scoring models, and we will see how the criticisms can be handled to fashion a system that works pretty well. The Screening Procedure It takes a while to develop a system; but once it is running, the fine-tuning does not require much effort.6 What Is Being Evaluated In the case of the above student, we chose to base the model on four arbitrarily selected factors. Selecting factors in real life is not that easy, and how we pick them is no accident. First, if we could, we would use only one factor. There is one factor that covers both technical and commercial accomplishment, a financial term called net present value of the discounted stream of earnings from the product concept, considering all direct and indirect costs and benefits. That mouthful is simply the finance way of saying “the bottom line on an income statement for the product, where we have included all costs (technical, marketing, and others) and then discounted back the profits into what their value is today.” That factor is shown on Level One in the abbreviated graphic of Figure 10.3. If it happens we can make a reasonably good estimate of that net present value, no other factors would be needed. But we almost never can; at this early point all financial estimates are quite shaky. In that case, we use surrogates (or substitutes) for it. Level Two in Figure 10.3 shows the obvious two: the likelihood of technical accomplishment (whether we can create something that will do what customers want) and the likelihood of commercial accomplishment (whether we can sell it profitably). There is again nothing left to do. Those two convictions would predict financial success of Level One, and we are finished. 6 Though quite easy when done in the mode of the scoring model example given later in this chapter, we should note that an immense body of theory lies behind all scoring decisions. For example, our scoring model is technically a linear compensatory model. That model, plus the conjunctive, disjunctive, and lexicographic models, is discussed (and compared in a new product screening exercise) in Kenneth G. Baker and Gerald S. Albaum, “Modeling New Product Screening Decisions,” Journal of Product Innovation Management, 3(1), March 1986, pp. 32–39. 248 Part Three Concept/Project Evaluation FIGURE 10.3 Source of Scoring Model Factors Level If know, OK Profit: Net present value One Don't know Two Chances of technical accomplishment Chances of commercial accomplishment If know, OK Sales If know, OK Don't know Three Margins Expenses Don't know Four Newness of wholesalers Familiarity with the market Product advantages Unfortunately, experience shows we usually can’t make these two estimates either. So we reach for more surrogates, this time at Level Three. To save space, Figure 10.3 shows only the three that produce commercial accomplishment; if we know our sales, our margins on those sales, and our marketing and administrative expenses, we have the commercial half of the answer. However, again we fall short; we don’t have a very good understanding of those figures either at this early point. Note, however, that the packaged goods firms developing marginally different new products, discussed above, can make these estimates and do so in their forecasting models. Most firms have to seek surrogates for the Level Three factors too. This leads us to Level Four, which is where the action is. Level Four factors have answers, or at least answers we can estimate better than the factors at higher levels. Figure 10.3 lists only three of the many factors at this level. The reasoning goes like this: If you tell me the new product will enter a market with which we already have great familiarity, chances are we will be able to communicate with buyers in that market. This raises the chances for good sales (up to Level Three), and greater sales make for more likely commercial fulfillment (up to Level Two), which, in turn, leads to profit (Level One, our objective). So the trick is to spot those Level Four factors that contribute to the technical and commercial operations in this firm on this particular product concept. Level Four factors comprise the scoring model shown in Figure 10.4. Some firms include profit, sales, and so on as factors even though their surrogates should be there already. Chapter Ten The Full Screen FIGURE 10.4 Scoring Model for Full Screen of New Product Concepts Technical accomplishment Category Factor Technical task difficulty Research skills required Development skills required Technical equipment/ processes Rate of technological change Design superiority assurance Security of design (patent) Technical service required Manufacturing equipment/processes Vendor cooperation available Likelihood of competitive cost Likelihood of quality product Likelihood of speed to market Team people available Dollar investments required Legal issues 1 2 Scale 3 4 5 Score Weight Weighted score Very difficult Have none required Have none required Easy Perfect fit Perfect fit 4 5 2 4 3 5 16 15 10 Have none required High/erratic None None Have none required Have them Stable Very high Have patent Have it all . . . . . . . . . Have none required None in sight Well above competition Below current levels Two years or more None right now Over 20 million Major ones Have them now Current relationship Over 20% less Leadership Under six months All key ones Under 1 million None in sight . Commercial accomplishment Total 210 Market volatility Probable market share Probable product life Similarity to product life Sales force requirements Promotion requirements Target customer Distributors Retailers/dealers Importance of task to user Degree of unmet need Likelihood of filling need Competition to be faced Field service requirements Environmental effects Global applications Market diffusions Customer integration Probable profit High/erratic Fourth at best Less than a year No relationship Have no experience Have no experience Perfect stranger No relationship Trivial No relationship None/satisfied Very low Tough/aggressive No current capability Only negative ones No use outside national No other uses Very unlikely Break even at best Very stable Number one Over 10 years Very close Very familiar Very familiar Close/current Current/strong Critical Current /strong Totally unmet Very high Weak Ready now Only positive ones 2 5 3 5 6 25 . . . . . . . . . Fits global Many other areas Customer seeks it ROI > 40% Total 240 Concept: Date of screen: Action: Grand Total 450 249 250 Part Three Concept/Project Evaluation FIGURE 10.5 Industrial Research Institute Scoring Model Technical success factors: • Proprietary Position: developing a strong, defendable patent in the technology to be researched. • Competencies/Skills: Available technical resources have the competencies to undertake the research project. • Technical Complexity: The impact of technical complexity on product success. • Access to and Effective Use of External Technology: The availability of external technology and the firm’s ability to use it successfully. • Manufacturing Capability: Relates to whether the firm has internal or external capabilities to manufacture the product or incorporate the process into its operations. Commercial success factors: • Customer/Market Need: Is there a ready market for the product or the process, resulting from the project? • Market/Brand Recognition: The likelihood that the product will be accepted in the marketplace, due to company strengths and/or image. • Channels to Market: The ease with which the product will be introduced and distributed. • Customer Strength: The probability that the product will succeed or fail based upon the strength of the customer in the business area of interest. • Raw Materials/Components Supply: The effect of the availability of key components and materials. • Safety, Health, and Environmental Risks: The probability that any of these effects will hinder project success. Source: John Davis, Alan Fusfield, Eric Scriven, and Gary Tritle, “Determining a Project’s Probability of Success,” Research-Technology Management, May–June 2001, pp. 51–57. Reprinted with permission. In general, a firm should start with the list of factors in Figure 10.4, scratch out any that clearly are not applicable, insert any obviously omitted, and then use it for a few times to see how the scores set with the people involved. Over time the list should be reduced as much as possible and always kept fluid. Nothing about this system should be set in stone; after all, it is just an aid to decision.7 The Logitech case at the end of this chapter will show how each situation is somewhat different. A new scoring model was recently developed by the Industrial Research Institute on how best to determine the success of an individual technical project. The model, developed with the help of this institute’s member company managers, contains two parts: a set of technical success factors and a set of commercial success factors. Each project is rated on each of these factors on a 1 to 5 scale. Importance weights for each success factor are also established. Weighted sums of the technical success and commercial success factors are calculated; projects with the highest total scores are most likely to succeed. The Industrial Research Institute’s model factors are as shown in Figure 10.5. 7 For further information, especially from a more corporate management view, see Thomas D. Kuczmarski, Managing New Products (Englewood Cliffs, NJ: Prentice-Hall, 2002). From the consumer products view, see Larry A. Constantineau, “The 20 Toughest Questions for New Product Proposals,” Journal of Product and Brand Management, 2(1), 1993, pp. 51–54. Chapter Ten The Full Screen 251 The Scoring Given a scoring form such as that shown in Figure 10.4 or 10.5, the team members who will be doing the scoring first undergo a familiarization period during which they get acquainted with each proposal (market, concept, concept test results). Then each scorer starts with the first factor (in this case, the difficulty of the technical task) and rates each one by selecting the most appropriate point on the semantic differential scales given in the third column. These scorings are multiplied by the assigned importance weights, and the factor totals are extended. The scorings continue for the other factors, and the ratings are then totaled to get the overall rating for that concept by each individual. Various methods are used to combine the individual team member’s ratings, an average (mean) being the most common. Some firms use the Olympic method of dropping the highest and lowest ratings before averaging. Some firms have an open discussion after the averages are shown, so individuals can make a case for any view that is at odds with the group. Many firms have found that groupware (e.g., Lotus Notes) aids the process greatly.8 Unusual Factors On some factors, a bad score constitutes a veto. For example, in the case of the student seeking to decide what entertainment to pursue this weekend, a money shortage may block anything costing more than $30. This problem should be faced in the beginning so no time is wasted drumming up options costing more than $30. It is the same in industry, and a key role for the product innovation charter is to point out those exclusions. These are sometimes called culling factors.9 Another problem occurs when the factor being scored has all-or-nothing, yesor-no answers; for example, “Will this concept require the establishment of a separate sales force?” This type of factor is handled by using the end points on the semantic differential scale, with no gradations. If possible, such factors should be scaled as, for example, “How much additional cost is involved in setting up sales coverage for this concept?” Columns might be None; Under $100,000; $100,000 to $300,000; and so on. The Scorers or Judges Selecting the members of a scoring team is like selecting the members of a new products team. The four major functions (marketing, technical, operations, and 8 For another matrix scoring model, see Bob Gill, Beebe Nelson, and Steve Spring, “Seven Steps to Strategic New Product Development,” The PDMA Handbook of New Product Development (New York: John Wiley, 1996), pp. 19–34. 9 See Rodger L. DeRose, “New Products—Sifting through the Haystack,” The Journal of Consumer Marketing, Summer 1986, pp. 81–84. This article shows some direct connections between product strategy at Johnson Wax and the firm’s new product screening; for example, its screening factors include “only safe products,” “use existing capabilities,” and “reflect the company’s position and style.” 252 Part Three Concept/Project Evaluation finance) are involved, as are new products managers and staff specialists from information technology, distribution, procurement, public relations, human resources, and so on, depending on the firm’s procedure for developing new products. Top business unit managers (presidents, general managers) should stay out of the act, except, of course, in small firms. Such people inhibit the frank discussions needed when assessing the firm’s capabilities (for example, in marketing or manufacturing). Some CEOs are intuitively so good at this task they can’t be excluded.10 Screening experience is certainly valuable. So is experience in the firm and in the person’s specialty. Technical people generally feel more optimistic about probable technical success, and marketers are more pessimistic. Problems with individuals are more specific. Research indicates that (1) some people are always optimistic, (2) some are sometimes optimistic and sometimes pessimistic, (3) some are “neutrals” who score to the middle of scales, (4) some are far more reliable and accurate than others, (5) some are easily swayed by the group, and (6) some are capable but erratic. Scoring teams need a manager to deal with such problems. Some firms actually weight each evaluator’s scores by past accuracy (defined as conformity with the team’s scores). Dow Brands uses a computerized groupware approach primarily because they like the scorings to be anonymous. Weighting The most serious criticism of scoring models is their use of weights because the weightings are necessarily judgmental (an exception from new research will be discussed in a moment). Let’s go back to the student seeking a weekend activity. To a money-cautious student, affordability deserves more weight than the other factors. But how much more? Should it be weighted at 2 and the other factors at 1? Because of weighting’s importance, some firms measure its effect using sensitivity testing. Scoring models are actually just mathematical models or equations, so an analyst can alter the scorings or the weightings to see what difference the alterations make in the final score. Spreadsheet programs handle this easily, and so does most groupware. Profile Sheet Figure 10.6 presents an alternative preferred by some firms for its graphic capability. The profile sheet graphically arranges the 5-point scorings on the different factors. If a team of judges is used, the profile employs average scores. The approach does indeed draw attention to such patterns as the high scores given near the bottom of the profile (in Figure 10.6) compared to those near the top. 10 One leading packaged goods firm’s CEO was such an expert at selecting among product manager job applicants that other evaluations were considered unnecessary. Chapter Ten The Full Screen 253 FIGURE 10.6 The Profile of a New Product Proposal Factors Score 1 2 3 4 5 Market size Market relatedness Market growth Market regularity Distribution capability Competitive status Regulatory freedom International potential Marketing capability Manufacturing capability Financial capability Engineering capability Vendor/supply options Technical uncertainty Strategic fit A Screening Model Based on Project NewProd Project NewProd, a comprehensive study of new product success and failure, was undertaken by Robert Cooper in the late 1970s. Some 100 Canadian industrial firms cooperated in the original NewProd study, in which product managers identified a recent product success and a recent failure. Respondents provided information on dozens of descriptive variables that might have been related to the product’s success or failure. From this study, the original New-Prod screening model, similar to the scoring models seen above, was derived and used to predict likelihood of product success and failure, and also to identify weak spots that ought to be rectified before approving the new product project.11 Since that time, the original NewProd model has been expanded and enlarged with the inclusion of data from many more firms and input from other new product 11 Robert G. Cooper, “Selecting Winning New Product Projects: Using the NewProd System,” Journal of Product Innovation Management, 2(1), March 1985, pp. 34–44. 254 Part Three Concept/Project Evaluation managers. Most recently, Cooper and his coauthors have advocated a two-level screening model, which combines checklists with scoring models. The two levels of criteria are must-meet and should-meet criteria. Must-meet criteria include good strategic alignment between project and strategy, and acceptable risk-return ratio; should-meet criteria include strategic importance, product advantage to the customer, and market attractiveness. The full set of criteria are listed in Figure 10.7. As FIGURE 10.7 Must-Meet and Should-Meet Criteria Based on the NewProd Studies Must-Meet Criteria: rated Yes/No 1. 2. 3. 4. 5. 6. 7. Strategic Alignment—does the product fit the business strategy? Existence of Market Need—does it surpass the minimum required size? Likelihood of Technical Feasibility—is it technically reasonable? Product Advantage—does it provide the customer with unique benefits or good value? Environmental Health and Safety Policies—does it meet the standards? Return versus Risk—is the ratio acceptable? Show Stoppers—any “killer” variables? Should-Meet Criteria: rated on scales (like scoring models) 1. Strategic a. To what extent does the project align with business strategy? b. What is the strategic importance of the project to the business? 2. Product Advantage a. To what extent does the product offer unique benefits? b. To what extent does the product meet customer needs better than the competition? c. To what extent does the product provide excellent value for the money? 3. Market Attractiveness a. What is the market size? b. What is the market growth rate? c. What is the competitive situation? (the more intense and price based, the lower the score) 4. Synergies a. To what extent does the product leverage marketing, distribution, or selling strengths? b. To what extent does the product leverage technical know-how or expertise? c. To what extent does the product leverage manufacturing or operations expertise? 5. Technical Feasibility a. How big is the technical gap relative to other products? (the smaller the gap, the higher the score) b. How complex is the product technically? (the less complex, the higher the score) c. What is the technical uncertainty of the outcome? (the higher the certainty, the higher the score) 6. Risk versus Return a. What is the expected profitability (NPV)? b. What is the percent return (IRR or ROI percent)? c. What is the payback period—how fast is the initial investment recovered? d. What is the certainty of the profit or sales estimates (pure guess or highly predictable)? e. To what extent is the product low cost and fast to do? Source: From Robert G. Cooper, Winning at New Products, 3rd ed., Perseus Books, 2001. Copyright © 2001 Robert G. Cooper. Reprinted by permission of Basic Books, a member of the Perseus Books Group. Chapter Ten The Full Screen 255 might be expected, must-meet criteria are designed to weed out the bad projects and function as high hurdles for the new product project. In fact, the authors suggest using a simple yes-no checklist for these, and a single “No” response might be enough to screen out the project. The should-meet criteria are those that characterize good business propositions. No one project would rate high on every one of these, so the authors suggest using a scoring model to combine all the criteria and rank the best new product projects by total score. What is notable in Figure 10.7 is that the must-meet and should-meet criteria both include a combination of both financial and strategic considerations. This is characteristic of the better product developing firms, which look beyond simple financial considerations when choosing new product projects to support.12 We will return to this important topic in our discussion of financial analysis in Chapter 11. The Analytic Hierarchy Process Another technique for product project screening and evaluation is the Analytic Hierarchy Process (AHP).13 AHP, developed in the 1970s by Thomas Saaty, is a general technique that systematically gathers expert judgment and uses it to make optimal decisions. It has been used in dozens of business and nonbusiness settings over the years and can be applied in full screening as a way to prioritize and select new product projects. When used as a full screen technique, AHP gathers managerial judgment and expertise to identify the key criteria in the screening decision, obtain scores for each project under consideration relative to these criteria, and rank the projects in order of desirability. Commercially available software such as Expert Choice makes AHP very easy to use.14 The product manager begins by building a hierarchical decision tree. The tree will show the manager’s ultimate goal (in this case, choosing the best new product project) at the top. The next level below will include all the primary criteria the manager considers important in reaching the goal. There may be several levels of criteria (secondary, tertiary, etc.) under the primary criteria in the tree. Lastly, the choices (new product projects under consideration) are placed at the bottom of the tree. Next, the manager provides comparison data for each element in the tree with respect to the next higher level. That is, the criteria are compared in terms of their importance in reaching the goal, and the choices are compared in terms of their ratings on each criterion. The AHP software takes over from this point. It converts the comparison data into a set of relative weights, which are then aggregated to obtain composite priorities of each element at each level. Ultimately, the available choices (new product projects) are rank ordered in terms of their preferability to the manager. 12 Robert G. Cooper, Winning at New Products, 3rd ed. (Cambridge, MA: Perseus Publishing, 2001). For a full treatment of AHP, see Thomas L. Saaty, The Analytic Hierarchy Process (New York: McGraw-Hill, 1980). 14 Expert Choice is presented in Arvind Rangaswamy and Gary L. Lilien, “Software Tools for New Product Development,” Journal of Marketing Research 34, February 1997, pp. 177–184. Expert Choice has a simple online AHP tutorial on its Web site, www.expertchoice.com, and also allows the user to download a small trial version of AHP from the Web site. 13 256 Part Three Concept/Project Evaluation FIGURE 10.8 An Application of the Analytic Hierarchy Process (AHP) Goal: Select Best New Car Project Market Fit Tech. Fit Dollar Risk Uncertainty Product Line Design Payoffs Unmitigated Channel Materials Losses Mitigated Logistics Supply Timing Mfg. Tech. Price Mfg. Timing Sales Force Differential Advantage Choose one of the following: Sedan, Subcompact, Two-Seater, SUV A sample, real-life application of AHP in a new automobile project screening setting is provided in Figure 10.8.15 In this case, the product manager for one of the Big Three U.S. automakers screens projects with respect to four primary criteria: fit with core marketing competencies, fit with core technical competencies, total dollar risk profile of the project, and managerial uncertainty about the project’s outcomes. (Again, much like in the NewProd-based model, both financial and strategic criteria are considered, though the specific criteria are somewhat different and more specific to the auto industry.) As shown in the figure, each of these primary criteria can be assessed in terms of several secondary criteria. For example, market fit considers the new product’s expected fit with the existing product line, distribution channel, distribution logistics, market timing strategy, price, and sales force. Finally, there are four new automobile projects under consideration (a sedan, a subcompact, a sporty two-seater, and an SUV); these are placed at the bottom of the decision tree. After the decision tree is built, paired comparisons are obtained. Usually, this is done by asking the manager first to rate the relative importance of the primary criteria in pairwise fashion on a scale of 1 through 9 (for example, “how much more/ 15 Roger J. Calantone, C. Anthony Di Benedetto, and Jeffrey B. Schmidt, “Knowledge Acquisition in New Product Project Screening with the Analytic Hierarchy Process,” Journal of Product Innovation Management, 16(1), January 1999, pp. 65–76. Chapter Ten The Full Screen 257 less important is fit with marketing competencies as compared to fit with technical competencies?”). Expert Choice allows several other ways for the paired comparisons to be entered by the respondent. Next, the relative importance of the secondary criteria are obtained (for example, “how much more/less important is fit with product line as compared to fit with distribution channel?”). Finally, comparisons of the new product projects with respect to each secondary criterion are made. Using these data, the AHP software calculates overall global weights for each new product project. These weights can be interpreted as the relative contribution of each alternative to the overall goal. The AHP output, shown at the bottom of Figure 10.9, clearly shows the sedan to be the preferred project, having the highest overall global weight (0.381). The subcompact is second best at 0.275, while the other two choices are also-rans. FIGURE 10.9 AHP Results and Overall Project Selection Level 1 Weights Dollar Risk Level 2 Weights Highest Ranked Project 0.307 Payoffs Losses Market Fit 0.153 0.153 Sedan Sedan 0.094 0.064 0.063 0.036 0.014 0.014 Sedan Subcompact Sedan Subcompact Sedan Subcompact 0.088 0.047 0.032 0.027 0.023 0.010 Sedan Subcompact Subcompact Subcompact Subcompact Sedan 0.104 0.078 Sedan Sedan 0.285 Timing Price Logistics Channel Product line Sales force Technical Fit 0.227 Differential advantage Manufacturing timing Design Materials Manufacturing technology Supply Uncertainty 0.182 Unmitigated Mitigated Ranking of Alternatives: Project Sedan Subcompact Two-Seater SUV Overall Weight 0.381 0.275 0.175 0.170 xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxx 258 Part Three Concept/Project Evaluation While all the AHP results cannot be shown here, Figure 10.9 summarizes some of the key findings and provides some insights on how the sedan came to be the top choice. The Level 1 weights indicate the relative importance of the primary criteria. This manager views dollar risk to be the most important criterion, followed by market fit, technical fit, and uncertainty. Similarly, the Level 2 weights indicate how important each of the secondary criteria are to this manager. For example, under market fit, timing and price are rated more important than sales force or product line fit. The last column shows the project that was ranked highest on each secondary criterion. The sedan was ranked highest on most of the secondary criteria and almost all of the really important ones (as judged by the Level 2 weights). The subcompact tended to do a little better on several of the technical fit criteria, but technical fit is less important to this manager than dollar risk or market fit. So it is not surprising that the sedan comes out ranked first, with the subcompact in second place. Special Aspects A few other aspects round out our discussion of scoring models. One concerns the product champion (discussed fully in Chapter 14). Champions are sometimes needed to push past normal resistance to change and to see that the concept gets a fair hearing at all turns. They try to give the scorers all favorable information and may argue that standard forms don’t fit their special situations. Some developers are trying to use computer technology with expert systems (often called knowledge-based systems). Such systems are essentially scoring models, with the factors developed on the basis of expert experience.16 Last, experience shows that management sometimes misuses scoring models. One consumer products manufacturer threw out a scoring model system because it 1. Was rejecting products that would help round out the line. 2. Was rejecting products that would help forestall competitive entry into the market. 3. Was rejecting too many products, according to the sales department. The first two problems arose from either faulty factor selection or faulty factor weighting and were easily solved. The third arose because the cutoff score was set too high. Scoring models require competent management. Summary If an idea progresses through early concept testing and development to the point where it is a full-blown concept ready for technical workup, it must then be screened. Screening is commonly done with scoring models, whereby the firm’s ability to bring off the required development and marketing is estimated. If the concept scores well by whatever criteria the firm uses, it is sent into technical development. 16 For a look at expert system performance, see Sundaresan Ram and Sudha Ram, “Validation of Expert Systems for Innovation Management: Issues, Methodology, and Empirical Assessment,” Journal of Product Innovation Management, 13(1), January 1996, pp. 53–68. Chapter Ten The Full Screen 259 Just prior to that, however, some firms try to spell out a protocol—an agreed set of benefits and other requirements that the technical development and marketing phases must deliver. And once the team feels the product parts of the protocol have been achieved, the concept is in prototype form. It can be taken to the field for further concept testing. The concept test is much more productive when the concept is in prototype form, though it may be more expensive because substantial technical expenditures have already been made. These matters of protocol and prototype testing will comprise Chapter 12. Applications 1. “Our small electrical engines division recently threw out a screening system that was based on a fairly complete scoring model, as they called it. Seems the model kept rejecting too many of their product ideas, some of which looked like sure winners to them—and to me, incidentally. Under their new system, a topmanagement committee reviews these ideas personally, without all that paperwork, and it looks like things will be better. Do you have any reaction to that?” 2. “We experimented with a numerical scoring model some years back—it just didn’t work. Brought in all the senior people, sales managers, product managers, you name it. We selected several dimensions of technical and commercial viability, not too different from the scoring model you presented. Rated everything on a scale of 1 to 5. Guess what? All the projects that were the pet projects of senior management came out as 5s. The ones they all could care less about came out as 1s. And all the ones we really lacked good information on came out as 3s. A lot of help that was! What went wrong?” 3. “If it happens that one of our divisions absolutely must use a scoring model, as you call it, I strongly prefer the one that we use at my company: Just get answers to four questions: Do customers care, Do we care, Can we do it, and Can we stay ahead if we do? What else could be more relevant? That list covers technical feasibility and commercial feasibility both, doesn’t it?” Case: Logitech (A)17 Logitech produces a wide range of computer-related products such as mice, keyboards, speakers, and webcams, as well as various accessories for tablets and smartphones. It has also gotten into the home security business, through the purchase of WiLife, a venture started by two entrepreneurs in 2002. WiLife products were renamed Logitech Alert products at the time of purchase. The home security 17 Information for this case was obtained from Jeanne Lee, “Simple Surveillance,” cnnmoney .com, Feb. 1, 2006; Walter S. Mossberg and Katherine Boehret, “Setting Up Your Own Security Camera at Home,” The Mossberg Solution, ptech.wsj.com, March 29, 2006; Edward C. Baig, “LukWerks Lets You Put Kids, Pets on Candid Camera,” USA Today, www.usatoday. com, July 12, 2006; Paul Taylor, “Network Cameras on Watch for Intruders—and Family Pets,” Financial Times, www.ft.com, January 18, 2007; and the Logitech Web site, www.logitech.com. 260 Part Three Concept/Project Evaluation product line has been greatly updated and expanded under the Logitech name; now the line includes state-of-the art indoor and outdoor systems that offer nightvision as well as high-definition-resolution images. This case goes back to the founding of the WiLife company by the two entrepreneurs, and how they progressed from opportunity identification and a good idea to their first marketable product. Evan Tree had about two decades of experience working as a dealer-installer in the video surveillance industry when he founded WiLife, Inc. in 2002. His partner in this venture, Andrew Hartsfield, had entrepreneurial experience, most recently in the beverage business. Evan had operated a local security dealership, Double Tree Security, for about 10 years and had just sold the business to a nationwide concern. Evan’s experience in video surveillance suggested that there was a gigantic hole in the marketplace. Video surveillance systems currently available on the market might cost as much as $4,000, even for a basic model. These were typically sold to business operators by big security companies such as Honeywell or ADT through dealer-operators like Double Tree. Not only was the price steep, but running wires and mounting cameras was complex work, and a central location would require a dedicated computer for monitoring. Big customers, or those such as jewelry stores or pharmacies with special surveillance requirements, would make the big investment. But the majority of small business owners found the price tag for video surveillance much too high. Evan and Andrew recognized that a small, inexpensive video surveillance system would fill a real market need for the small businessperson. The two partners began thinking of the market opportunity and the technical and commercial factors that would be most critical to success. On the technical side, a place to start would be the weaknesses of the current systems. Systems available at the time used analog cameras. It would make sense to explore digital cameras for this application, especially as the costs would probably not be too high. Digital video clips could potentially be sent to a PC, or even to a mobile phone with video capability. It also makes sense to explore Ethernet networking to connect the cameras to the PC. This is a standard and readily available technology that would eliminate the need for new wiring, as it runs on the electrical power lines already in the building. One would then need to think about video image storage for review by business owners or the police (in the event of a break-in caught on tape, for example). Currently, images caught by analog cameras were videotaped, and the tapes were frequently reused after a certain number of weeks; this might be a reasonable starting point for storage requirements. Picture quality would have to be good, and it would be a nice feature to add digital time stamps for convenient playback and searching. Though Evan and Andrew thought first only of in-store surveillance, the concept might be developed for external cameras as well, though this might pose additional technical issues (making the cameras waterproof, for example, or using infrared technology so the camera can see in the dark). There are obvious commercial factors to consider here as well, first and foremost being price. It is almost a certainty that more small businesspeople would buy video surveillance systems if the price were not so high. Still, there are nonprice Chapter Ten The Full Screen 261 attributes to consider. As noted above, it should be technologically possible to send digital images to the user’s PC or phone; customers should like this because it eliminates the need for a dedicated PC. Plus, the system could be “smart,” alerting the user to any unusual activity by sending an alarm message to the user’s PC or phone. (If it is really smart, it should be able to distinguish a burglar breaking a window from a cat sitting on the windowsill.) Additionally, having the crew out to install video cameras is time-consuming and disruptive; ease of camera installation should be a consideration. Similarly, it should be painless to load the required software onto the PC. You don’t have the information to compose an entirely new scoring model for use on the new product concept discussed in this case, but you can put together the five most important factors under each part of the model in the chapter (technical factors and commercial factors). This case gives you a few suggested technical and commercial factors to consider; try to add a few of your own to these. Give all of them weights. Then apply your model to the new product concept Evan and Andrew have recently been looking at (the new inexpensive digital video surveillance system). Beyond the task of developing a scoring model for them, give some thought to the problems of implementing the scoring model system for this product. Would any of the results of the scoring session potentially cause big problems? Why? C H A P T E R E L E V E N Sales Forecasting and Financial Analysis Setting Now that we have finished the full screen, we know the product concept meets our technical capabilities (present or acquirable) and that it meets our manufacturing, financing, and marketing capabilities as well. Also, we know it offers no major legal problems, and so on. So we are ready to charge ahead. Or are we? Most managers don’t think so—they are very interested in the financial side of this proposition. In fact, they have been interested in money from the very start of a project—think back to the product innovation charter where we talked about the size of potential markets and objectives on market shares and profits. And they will still be interested in money when they look back and total up whether the whole project was worthwhile. In addition, knowledgeable managers have learned that looking at the financial projections is not enough. To make the best possible choices from all projects being considered, one needs to keep in mind how well each project fits with the organization’s strategic goals and competencies. Indeed, one of the biggest problems facing firms at this phase is that they commit to too many projects, spreading human and financial resources out too thin. That is, firms need to improve their project selection procedure—for many, that means considering strategic fit to a greater extent than previously.1 Now is a good time to take a closer look at the managerial side of analysis. How should we select and manage a new product project such that it achieves reasonable financial goals and is in keeping with the PIC? In this chapter we focus our attention on the financial analysis and in particular on the sales forecast, which is usually one of marketing’s most critical contributions to the financial analysis. We then reconsider the product innovation charter to determine whether the project(s) under consideration are consistent with the firm’s strategy for innovation. These activities make up part of the last box in Figure III.1: They are part of the project approval process. In the next chapter, we will develop a written protocol for the project—at that point, we are ready to move forward to the development phase. 1 Robert G. Cooper, “From Experience: The Invisible Success Factors in Product Innovation,” Journal of Product Innovation Management, 16(2), March 1999, pp. 115–133. Chapter Eleven Sales Forecasting and Financial Analysis 263 Sales Forecasting for New Products We begin the financial analysis with the sales forecast. As noted above, this is typically the responsibility of the marketing person on the new product team. Once sales have been projected over the next several planning periods, we can assess costs, make profit projections, and calculate key financial benchmarks such as net present value or internal rate of return. Other participants on the team (such as manufacturing engineers, R&D people, financial and accounting specialists, etc.) have a greater input in providing the costs and other data that will make up the financial analysis. One of the hardest challenges in financial analysis is developing a reasonable sales forecast, especially for a very new product based on rapidly advancing technology. In 2000, forecasters were predicting that by 2007 there would be 36 million satellite radio subscribers; a year later this forecast was reduced to about 16 million. The actual number achieved by the end of 2006 was about 11 million, and revenues to Sirius and XM Satellite have been much lower than expected.2 See Figure 11.1 for some other forecasts—good and bad—about today’s society and FIGURE 11.1 What the Future Looked Like in 1967 In 1967, noted authorities in science, computers, and politics made a series of long-term forecasts about the coming 30 years. Many of these turned out to be highly accurate: • We would have artificial plastic and electronic replacements for human organs by 1982, and human organ transplants by 1987. • Credit cards would virtually eliminate money by 1986. • Lasers would be in common use by 1986. • Many of us would be working at home by the 1980s, using remote computer terminals to link us to our offices. • By 1970 man would have walked on the moon. • By 1986 there would be explosive growth in expenditures on recreation and entertainment. While about two-thirds of the forecasts were remarkably accurate, about a third were just plain wrong. Samples: • • • • Manned planetary landings by 1980, and a permanent moon base by 1987. Private cars banned from city centers by 1986. 3D television globally available soon. Primitive life created in the laboratory by 1986. What can we learn from the correct, and from the incorrect, forecasts? Firstly, forecasts do not have to be absolutely perfect to be used for planning. Recall that old-time ship captains used maps that contained inaccuracies, but still got where they wanted to go. Secondly, incorrect forecasts seemed to fit into two categories: underlying factors driving the projections changed or the forecaster was overly optimistic in the speed of development. Space funding was substantially cut back after the 1969 moon landing, throwing off forecasts about future space exploration. 3D television may indeed be big a couple of decades from now—of course we were saying that about video phones back in the 1960s. Sources: From Edward Cornish, “The Futurist Forecast 30 Years Later,” The Futurist, January–February 1997, pp. 45–58. Originally published in The Futurist. Used with permission from the World Future Society, 7910 Woodmont Avenue, Suite 450, Bethesda, Maryland 20814 USA. Telephone: 01-656-8274; www.wfs.org. 2 See Sarah McBride, “Until Recently Full of Promise, Satellite Radio Runs into Static,” Wall Street Journal, August 15, 2006, pp. A1, A9. 264 Part Three Concept/Project Evaluation products made by a panel of futurists a few decades ago. What the figure suggests, however, is that expert forecasters often do quite a good job predicting how advancing technologies will result in new products, even 30 years or more into the future, provided they keep a level head. We must keep in mind several considerations when developing the sales forecast. First, a product’s potential may be extremely high, but sales may not materialize due to insufficient marketing effort. Advertising may not adequately create awareness, or inadequate distribution may make the product unavailable to much of the market. The A-T-A-R model we discussed in Chapter 8 will help us adjust sales forecasts based on awareness and availability. Second, sales will grow through time if we successfully get customers to try the product and convert many of these customers into repeat purchasers, if they pass along favorable word of mouth to their friends, if greater demand encourages more dealers to stock the product, and so on. After this growth period, sales will eventually stabilize. Thus, we will be interested in developing projections of long-run sales or market shares. Third, we should recognize that our product’s sales will depend on our competitors’ strategies and programs as well as our own. Forecasting Sales Using Traditional Methods Many standard techniques, such as those shown in Figure 11.2, can be taken to forecast a new product’s sales at this early phase in the new products process.3 In addition to the considerations of time and cost, one should also consider product FIGURE 11.2 Commonly Used Forecasting Techniques Technique Time Horizon* Cost Comments Simple regression Multiple regression Short Short-medium Low Moderate Econometric analysis Simple time series Advanced time series (e.g., smoothing) Short-medium Short Short-medium Jury of executive opinion Scenario writing Delphi probe Medium Medium-long Long Moderate to high Very low Low to high, depending on method Low Moderately high Moderately high Easy to learn More difficult to learn and interpret Complex Easy to learn Can be difficult to learn but results are easy to interpret Interpret with caution Can be complex Difficult to learn and interpret *Generally, a short time horizon means under three months; medium time horizon means up to two years; and a long time horizon means over two years. For more details on these and other forecasting techniques, please consult any good forecasting textbook. Source: Adapted from Spyros Makridakis and Steven C. Wheelwright, “Forecasting: Framework and Overview,” in Forecasting, S. Makridakis and S. C. Wheelwright (editors), Studies in the Management of Sciences, Vol. 12, Amsterdam: North-Holland, 1979. Reprinted with permission. 3 For an excellent resource, see Kenneth B. Kahn, New Product Forecasting: An Applied Approach (Armonk, NY: M. E. Sharpe, 2006). Also see Kenneth B. Kahn, “Using Assumptions-Based Models to Forecast New Product Introduction,” in A. Griffin and S. M. Somermeyer, The PDMA Toolbook 3 for New Product Development (New York: John Wiley, 2007). Chapter Eleven Sales Forecasting and Financial Analysis 265 FIGURE 11.3 New Product Forecasting Strategies Current Product Technology New Product Technology Current Market Type of innovation: cost reductions and process improvements Type of forecasting: sales analysis Type of innovation: line extension Type of forecasting: product line analysis, life cycle analysis New Market Type of innovation: new market or new product uses Type of forecasting: customer analysis, market analysis Type of innovation: new-to-the-world or new-to-the-firm Type of forecasting: scenario or “what-if” analysis Source: Adapted from K. B. Kahn, “Forecasting New Products,” in K. B. Kahn, S. E. Kay, R. J. Slotegraaf, and S. Uban (Eds.), The PDMA Handbook of New Product Development (Hoboken, NJ: John Wiley), 2013, Ch. 16, p. 276. and market newness when selecting the most appropriate forecasting model (see Figure 11.3). The most straightforward kind of forecast to conduct is a sales analysis, used for current technologies being sold into current markets (for example, Kellogg’s assessing how many boxes of corn flakes it will sell next year if it can reduce costs per package by 5 percent). Time series and regression forecasts are useful here. For selling a new technology into a current market (the new generation of HP printer replacing the old model), a product line or life cycle analysis is recommended. Despite some technology uncertainty, reasonable forecasts can be obtained by analogy: the new generation’s life cycle curve would be forecasted as similar to that of the current generation. For example, current printer sales might have been slow in the first couple of months, but peaked at months 7 and 8. This pattern might repeat for the new printer. The situation is reversed if a current technology is being sold into a new market (for example, the current HP printer being sold into foreign markets). In this case, customer and market analysis would be required to minimize the uncertainty surrounding the behavior of the new market. Finally, for new-to-the-world or new-to-the-firm products, the best forecasting methods would be scenario or “what-if” analyses. No data exist on past sales or even on whether the new technology would be accepted, so more subjective techniques are required here. The forecasting task is obviously a lot more difficult in this latter case, and some might think it to be almost impossible. Yet, Professor Kahn notes that accuracy rates of 40 percent for new-to-the-world products are achieved, on average, over a forecast time horizon of 36 months. By comparison, accuracy rates for simpler forecasting tasks, such as for cost reductions, product improvements, or line extensions, are in the 63–72 percent range, with a shorter time horizon of about 21 months.4 4 K. B. Kahn, “Forecasting New Products,” in K. B. Kahn, S. E. Kay, R. J. Slotegraaf, and S. Uban (Eds.), The PDMA Handbook of New Product Development (Hoboken, NJ: John Wiley, 2013), Ch. 16, pp. 276–278. 266 Part Three Concept/Project Evaluation Forecasting Sales Using Purchase Intentions Think back to concept testing (Chapter 9). Among other things, we gathered purchase intentions from respondents. When presented with a concept, they were asked (typically using a 5-point scale) to state their likelihood of purchasing that product if it were made available. As mentioned at that time, it is common to look at the top-two-boxes totals (the number of customers who stated they would either definitely or probably buy the product). This measure can be refined and calibrated through experience. As an example, recall that in our example of a concept test for an aerosol hand cleanser (Figure 9.2), we found that 5 percent of the respondents would definitely buy it, and 36 percent would probably buy it. Based on averages from data collected on similar products launched in the past, about 80 percent of those people who say they would “definitely” buy actually buy the product, and 33 percent of those who say they would “probably” buy actually buy. From this information, our first estimate of the percentage of potential purchasers would be (0.05)(0.80) 1 (0.36)(0.33) 5 16%. This estimate assumes 100 percent awareness and availability, and so it would have to be adjusted downward. If we expect that 60 percent of the market will be aware of the product and have it available to them at a nearby retail outlet, our predicted percentage of actual purchasers would be (0.16)(0.6) 5 9.6%. As a refinement to this method, we could also vary the concept and get separate purchase intentions for each variation. For example, we might have asked respondents to state their purchase intentions for an aerosol hand cleanser that disinfects as well as cleans, using the same 5-point scale. As another example, consider satellite radio again.5 In 2000, there were roughly 213 million vehicles in the United States. Let’s assume 95 percent availability (due to heavy distribution at satellite outlets) and 40 percent awareness (attributed to heavy promotion by Sirius and XM Satellite). Market potential adjusted for awareness and availability is (213 million) 3 (0.40)(0.95) 5 81 million. Market research suggests that half of this market could afford satellite radio; the forecast now becomes 81 million 3 0.5 5 40.5 million. Of these, what percentage actually intends to subscribe to satellite radio? One way to estimate this is to estimate the percentage of customers who are among the first to try a new technology. If this percentage is estimated at 16 percent, then the forecast becomes 40.5 million 3 16%, or a little over 6.4 million. Let’s take this as first-year (i.e., 2001) subscriptions and project yearly effective growth rate at 10 percent. (Effective growth rate means that we are considering new subscriptions as well as defectors.) By the end of 2006, we would project a little over 10 million subscribers—below the actual number attained but much closer than the industry estimate of 36 million! Indeed, the two rivals (Sirius and XM Satellite) agreed to merge in early 2007. 5 The satellite radio example is adapted from Kenneth B. Kahn, “Using Assumptions-Based Models,” op. cit. Chapter Eleven Sales Forecasting and Financial Analysis 267 Forecasting Sales Using the A-T-A-R Model In Chapter 8 we worked through a simple example of the A-T-A-R model in action and briefly mentioned where some of the data could be obtained. This simple model can be used to construct a sales or profit forecast, and market researchers long ago pushed the early, simple models into far more powerful forecasting devices. These advanced research models are used largely on consumer packaged goods, where firms have lots of new product experience on which to develop model parameters and to calibrate the raw percentages they get from consumers. The A-T-A-R model is the basis of many of the simulated test markets we will encounter in Chapter 18. This is one of the pseudo sale market testing methods used later in the new product process, typically when the physical product is available for the consumer to take home and try. Post-trial data are then collected from the consumer and used as input to the A-T-A-R model. At this early stage in the new products process, before product design and prototype manufacture, the A-T-A-R model can still be applied using data from other sources, and even assumptions. Trial and repeat rates that need to be achieved to reach sales or profit projections can be estimated early and adjusted as the product goes through later stages and more information becomes available. Here we are using a form of A-T-A-R that is commonly used in forecasting market shares. First-time product trial might be estimated using the purchase intention method described above. For frequently purchased consumer packaged goods, it is critical to get a good estimate of repeat purchase as well as trial, since long-run market share can be expressed as MS 5 T 3 R 3 AW 3 AV where T 5 ultimate long-run trial rate (the percentage of all buyers who ultimately try the product at least once) R 5 ultimate long-run repeat purchase rate (share of purchases of the product among those who tried the product) AW 5 percent awareness AV 5 percent availability Repeat purchase rate, R, can be obtained by analogy to similar products for which such data are available. It can also be calculated using a switching model.6 We can define Rs as the proportion of customers who will switch to the new product when it becomes available, and Rr as the proportion of customers who repeat 6 This switching model is an application of a Markov model (a form of model used to determine equilibrium states) in which the long-run repeat purchase rate is the equilibrium state. Details on the switching model are given in Glen Urban, “PERCEPTOR: A Model for Product Positioning,” Management Science, 21(8), 1975, pp. 858–871. 268 Part Three Concept/Project Evaluation FIGURE 11.4 100% 90 80 70 60 50 40 30 20 10 0 A-T-A-R Model— Results in Bar Chart Format 0.9 0.603 0.0965 Aware Available Trial 0.0614 Repeat purchase the product. The switching model estimates long-run repeat purchase rate, R, as Rs/(1 1 Rs 2 Rr ). If Rs and Rr are estimated as 0.7 and 0.6, respectively, repeat purchase is estimated as 0.7/(1 1 0.7 2 0.6) 5 0.636. If awareness and availability are 90 percent and 67 percent, respectively, and 16 percent of the market that is aware of the product and has it available to purchase tries it at least once, long-run market share is calculated as MS 5 0.16 3 0.636 3 0.90 3 0.67 5 6.14% Furthermore, if the total number of purchases in this product category is known, this market share can be converted into long-run sales. If total number of purchases is 1,000,000 units, the firm’s long-run sales are estimated to be 1,000,000 3 6.14% 5 61,400 units. The process of calculating market share is illustrated in the bar chart in Figure 11.4. The y-axis represents the total market (100 percent). The figure shows that 90 percent of the market is aware of the product; 67 percent of the “aware” market (67% 3 90% 5 60.3%) also has the product available to them; 16 percent of the “aware” market that has the product available tries it at least once; and 63.6 percent of the latter become repeat purchasers. As stated above, the accuracy of forecasts obtained using these methods depends on the validity of the measures. When building the forecasting models, one must also consider data availability, and also data precision. In the above example, we had assumed availability of 67 percent, but this might not be very precise; actual availability may be as low as 40 percent or as high as 80 percent. In such a case, it makes sense to do a what-if analysis. Substituting these values into the market share calculation, we see that the market share forecast falls into a range of 3.66 percent to 7.33 percent, representing the worst- and best-case scenarios.7 We shall return to A-T-A-R models of this type when we have a product prototype we are ready to test with customers, a little later in the new products process. 7 See Kenneth B. Kahn, “Using Assumptions-Based Models,” op. cit. Chapter Eleven Sales Forecasting and Financial Analysis 269 Techniques for Forecasting Product Diffusion Diffusion of innovation refers to the process by which an innovation is spread within a market, over time and over categories of adopters. The adopter categories, which we will look at more closely in Chapter 16, are often called innovators, early adopters, early and late majority, and laggards. In theory, individuals in the earlier adopter categories influence the purchase behaviors of later ones through word of mouth and other influence processes. The rate of diffusion of a product can be difficult to assess, especially at this early stage in the new product process, since it is unknown how influential the earlier adopter categories will ultimately be. We have already seen, in the satellite radio example, how important it is to get an estimate of the number of innovators and early adopters (i.e., those users who will be among the first to try the product). To get a handle on the growth potential of an innovative product, we can use an analogous existing product as a guideline. If we are assessing the market potential of a new kind of automobile tire (that could, say, run safely for 100 miles after being punctured), we could reasonably use common radial tires as an analogy. They are sold to the same populations (car manufacturers and service centers) and provide basically the same benefit. Thus, as a rough estimate, long-run market potential for our new tire is probably similar to the sales level achieved by radial tires. Managerial judgment regarding our new product might suggest that actual market potential be somewhat higher or lower than this initial estimate. Quantitative innovation diffusion models can also be used in predicting future product category sales based on historical product sales levels. A diffusion model commonly used for durable goods is the Bass model,8 which estimates the sales of the product class at some future time t, s(t), as: s(t) 5 pm 1 [q 2 p] Y(t) 2 (q/m) [Y(t)]2 where p is initial trial probability, q is a diffusion rate parameter, m is the total number of potential buyers, Y(t) is the total number of purchases by time t The Bass diffusion model is based on the diffusion curve of new products through a population. The initial diffusion rate (growth in total number of purchases) is based on adoption by innovators. Following these early purchases, the growth rate accelerates as word-of-mouth helps to promote the product and more of the market adopts the product. Eventually, however, we reach the point where there are not that many potential purchasers left that have not yet tried the product, and growth rate slows. 8 The model was originally published by Frank Bass, “A New Product Growth Model of Consumer Durables,” Management Science, 15(1), January 1969, pp. 215–227, and has since been extended in dozens of research articles. This stream of literature is reviewed in Vijay Mahajan, Eitan Muller, and Frank M. Bass, “New Product Diffusion Models in Marketing: A Review and Directions for Research,” Journal of Marketing, 54(1), January 1990, pp. 1–26. 270 Part Three Concept/Project Evaluation Managerial judgment, or standard procedures for market potential estimation, can be used to estimate m, the number of potential buyers. If the product category has been around for a while and several periods of data exist, one could use past sales to estimate the size of p and q. To set these values for a recent innovation, one might look at similar (analogous) products for which these values are known or rely on judgment or previous experience with this kind of model. Previous studies suggest that p is usually in the range of about 0.04, and q is typically close to 0.3, though these values will vary depending on the situation.9 A desirable feature of this growth model is that, once p and q are estimated, the time required to reach the sales peak (t*) can be predicted, as can the peak level of sales at that time (s*). These are given as: t* 5 (1/(p 1 q)) ln(q/p) s* 5 (m)(p 1 q)2/4q Let’s say you are working for a company that is assessing the viability of a new product category: a combination cappuccino maker–miniature convection oven. You believe the long-run potential for this product is in the area of 25,000,000 households. For similar small household appliances your company has sold in the past, innovation and imitation rates have tended to be in the area of 2 percent and 12 percent. Figure 11.5 presents a sales forecast derived for this new product category, based on applying the Bass model to these estimates. This preliminary forecast suggests that peak sales will occur about four years from now, and that total product category sales during that year will be a little over 4 million units. If these sales projections are combined with price, cost, and market share projections, FIGURE 11.5 4,500 Bass Model Forecast of Product Diffusion 4,000 3,813 3,500 4,056 3,704 3,127 2,917 3,000 Sales 2,500 ($000s) 2,000 2,029 2,300 1,500 1,287 1,000 769 442 500 0 9 0 1 2 3 4 5 Year 6 7 8 9 10 Parameter estimation issues are discussed in Vijay Mahajan and Subhash Sharma, “Simple Algebraic Estimation Procedure for Innovation Diffusion Models of New Product Acceptance,” Technological Forecasting and Social Change, 30, December 1986, pp. 331–346; and Fareena Sultan, John U. Farley, and Donald R. Lehmann, “A Meta-Analysis of Applications of Diffusion Models,” Journal of Marketing Research, 27(1), February 1990, pp. 70–78. Chapter Eleven Sales Forecasting and Financial Analysis 271 the product’s potential projected contribution to profit can be assessed. The Bay City Electronics case at the end of this chapter shows a set of sales projections for a new product (derived using Bass or some similar model) and takes you through these steps, finishing with an NPV analysis. Bass showed that, despite its simplicity, his model did a good job at predicting the time and magnitude of the sales peak for many durable consumer goods, including clothes dryers, television, coffee makers, irons, and many others. Later researchers have used the Bass model to forecast the diffusion of many high-tech product categories such as satellite TV or music CDs.10 Interestingly, it has also been used to predict the growth of Internet communities such as Facebook. The similarities to durable goods diffusion are striking: One is either a member of Facebook or is not; and some people will be the innovators and join right away. The more influential these innovators are in encouraging others to join, the faster the new community will grow.11 Other extensions of the Bass model have shown how it can be applied to nondurable goods where repeat sales need to be considered.12 Many other extensions have included more variables and better estimation techniques, resulting in even more accurate sales forecasts.13 Observations on Forecasting Models Model makers are rapidly accumulating experience and sharpening their models, which are now readily available to consumer packaged goods innovators, are quite inexpensive compared to test markets and rollouts, and allow diagnostic output as well as sensitivity testing. Unfortunately, they also require massive amounts of data to work well, are built heavily on assumptions, and are so complex that many managers are wary of them. Having been developed initially in the 1950s and 1960s, they often incorporate assumptions no longer valid—for example, reliance on mass advertising and easy-to-get distribution. But they are now a mature industry, a large and profitable one. 10 C. van den Bulte, “Technical Report: Want to Know How Diffusion Speed Varies Across Countries and Products? Try Using a Bass Model,” Visions, 26(4), October 2002, pp. 12–15. 11 D. R. Firth, C. Lawrence, and S. F. Clouse, “Predicting Internet-Based Online Community Size and Time to Peak Membership Using the Bass Model of New Product Growth,” Interdisciplinary Journal of Information, Knowledge, and Management, 1, 2006, pp. 1–12. See discussion of this topic in C. Anthony Di Benedetto, “Diffusion of Innovation,” in V. K. Narayanan and Gina C. O’Connor (eds.), Encyclopedia of Technology & Innovation Management (Chichester, UK: John Wiley, 2010), Chapter 16. 12 See Vijay Mahajan, Eitan Muller, and Frank M. Bass, “New Product Diffusion Models in Marketing: A Review and Directions for Research,” Journal of Marketing, 54(1), January 1990, pp. 1–26. 13 A brief discussion of this literature is in Deepa Chandrasekaran and Gerard J. Tellis, “Diffusion of Innovation,” in Jagdish N. Sheth and Naresh K. Malhotra, Wiley International Encyclopedia of Marketing, Volume 5, Product Innovation and Management (West Sussex, UK: John Wiley, 2011), pp. 49–51. 272 Part Three Concept/Project Evaluation It is interesting that the most successful firm by far uses the simplest methodology and requires the least data. In BASES II, Burke (a division of Nielsen) combines a concept test and a product use test, calibrates the trial and repeat percentages from their massive files of past studies, and uses a set of experiencehoned heuristics (rules of thumb) to translate those percentages into market shares. But product innovators outside of consumer packaged goods still most often use the simple version of the A-T-A-R model in Chapter 8, if they use any forecasting model at all. Research continues toward improving all of the sales forecasting models.14 Problems with Sales Forecasting Doing the sales forecasts poses no problem as such. We have an immense arsenal of forecasting methodologies, as seen above in Figure 11.2. We know, based on the A-T-A-R model we encountered in Chapter 8, what makes for sales. This model does an excellent job and serves as the basis for some very advanced mathematical systems used by sophisticated new product marketers. And every firm has people who can make an income-statement-based net present value (NPV) or internal rate of return (IRR) calculation (using discounted cash flow methods),15 as illustrated in the Bay City case at the end of the chapter. The real problems are getting the required information to do the financial analysis and not ignoring strategic issues when considering new product projects. One can use A-T-A-R to assist in building the sales forecast for the financial analysis. A-T-A-R, however, requires a solid estimate of how many people/firms will become aware of our new item, how many of those will opt to try the item in one way or another, and so on. Each of these figures, however, is very difficult to estimate. For example: • The folks at Google or Twitter did not know their Web sites would become that popular. • Apple did not know so many of us, even dyed-in-the-wool Windows users, would buy iPods or iPhones. • Amazon.com did not know we would buy millions of books over the Internet. • Samsung did not know that the Galaxy smartphones would become all the rage. 14 For a discussion of the use of forecasting techniques used in new product development, see Kenneth B. Kahn, “An Exploratory Investigation of New Product Forecasting Practices,” Journal of Product Innovation Management, 19(2), 2002, pp. 133–143; and Kenneth B. Kahn, New Product Forecasting, op. cit. 15 While we use NPV analysis in this chapter, some analysts suggest internal rate of return (IRR) instead for financial evaluation of projects since the latter tends to select the largest projects, not necessarily the highest-return projects! See Carey C. Curtis and Lynn W. Ellis, “Satisfying Customers While Speeding R&D and Staying Profitable,” Research-Technology Management, September–October 1998, pp. 23–27. Chapter Eleven Sales Forecasting and Financial Analysis 273 Also, the financial model requires product cost, prices, the current value of money, probable taxes on the future income, the amount of further capital investments that will be required between now and when we close the books on the product, and much more. These will never be certain, even after living out the product’s life cycle. Sales will be known, but we might have had a better marketing strategy. Costs are always just estimates. We will never know the true extent to which a new item cannibalized sales from another product. If we had not marketed the new item, a competitor probably would have. The fact is, we rely on estimates. Management’s task is to make the estimates as solid as we can and then manage around the areas of uncertainty in such a way that we don’t get hurt too badly. On minor product improvements we do this quite well—a new Troy-Bilt lawn mower with a pepped-up engine is not a wild guessing game. On near line extensions, we also do well, but with more misses. Really new products, using technologies never so applied before, are pure guessing games. For over 30 years business schools used a Polaroid pricing case in which Edwin Land was trying to decide whether people would pay $15, or maybe $25, or (dream on) $50 for his first instant camera! They paid the top price in huge numbers, making Land a very rich “financial and technical genius.” Summary of the Problems What makes forecasting so difficult? For one thing, target users don’t always know what the new product will actually be, what it will do for them, what it will cost, and what its drawbacks will be, nor will they have had a chance to use it. And if they do know, they may want to keep some information from us or offer outright falsehoods. Complicating this problem is that market research on these potential users is often poorly done—there is no lack of horror stories about focus groups. At the same time, competitors don’t sit still. In fact, they are trying very hard to ruin our data, just as we do to theirs. Resellers, regulators, and market advisers are in a constant flux. Information about marketing support—what kind of service will be available in the firm, for example—may be lacking. No sales manager can make promises a year ahead about sales time and support. Internal attitudes can be biased, and politics are always present. Many new products managers will not be ready to show just how good the new item is for some time, so they try to delay official forecasting. In their excitement to get to market, new products managers sometimes get themselves into trouble by rushing their products out, without stopping to fieldtest the new item. Steelcase management, responding to some disappointments, now demands that new office furniture systems be thoroughly tested in end-user offices. Finally, most common forecasting methods are extrapolations and work well on established products. New products don’t have a history. Even forecasting methods that seem free of history (use of leading indicators and causal models) use relationships established in the past. 274 Part Three Concept/Project Evaluation Actions by Managers to Handle These Problems Given that we badly need financial analyses and that good analyses are difficult to make, what is a manager to do? Improve the New Product Process Currently in Use Most of the horror stories given earlier from the trade press are embarrassing to their managers. In most of them a key step was skipped. In an effort to hurry, or to capitalize on the conviction of someone working in or around the project, a bad assumption was made. For example, New Coke was heavily taste-tested, but was not market-tested—that is, no one was actually asked to buy the product with that new name, and certainly no one was told in the test that if New Coke were launched, Classic Coke would be immediately dropped. So the emotional backlash over the loss of Classic Coke was totally missed. When it first launched the minivan in the early 1980s, Chrysler was wiser—it knew that consumers were negative toward the minivan because they couldn’t see the value in it until they actually drove it for a while. So Chrysler made sure customers drove it and learned that, despite its size, it handled like a car (since it was, and still is, built on a car platform). Top new product professionals today know good new products process, but many others don’t. They lack information and don’t realize it. All the standard forms will not make up for omissions of key data pieces. Use the Life Cycle Concept of Financial Analysis Firms sometimes err by focusing their financial analysis at one particular point— perhaps a stage in a phased system. That point is often right where we are in this book, at full screen. Another popular time is later, near where some major financial commitment must be made—for example, building a plant or releasing an expensive marketing introductory program. Managers talk about a point of no return. It is indeed a phase, a hurdle, and new product managers may spend weeks getting ready for the meeting. But both instances are exaggerated. Technical work can begin without committing the firm to a huge technical expenditure. Building a plant can often be avoided by contracting out early production or by building a large pilot system for trial marketing in a restricted rollout. It is far better for managers to see their project as a living thing—a bottom line that is created gradually, over the life of the project, never being completely accurate, even well after the item is launched (see Figure 11.6). A product innovation charter is accepted only because the management believes the combined technologies and market opportunities fit well with each other and with the firm. A PIC describes a home field where we can’t ever be sure of the final score, but where we should be able to win. A concept test result doesn’t assure financial success either, but it can say we are one step further along—the intended user agrees there is a need for something like our concept and wants some to try out. An early field use test with a prototype also won’t assure success, but it can say intended users like what they see. An advertising agency or a sales manager cannot guarantee success Chapter Eleven Sales Forecasting and Financial Analysis 275 FIGURE 11.6 Financial Analysis as a Living Thing: The Life Cycle of Assessment 5 100% 4 100 3 Completeness of financial analysis 2 1 100% Opportunity Concept Concept/ identification generation project and evaluation selection Development Technical Marketing 1 Concept test 3 Product use test 2 Full screen 4 Market test Launch Rest of market life 5 End-user decision either, but they can assess whether the new item will be brought to the attention of potential end-users and that it will be tried. If it delivers, it will sell, and if manufacturing is able to do what it felt it could, there will be profit in the item. And so on. The best we can do at any point is to ask whether progress to date is consistent with a successful life cycle. It is the same with financial analysis—where are we today, is what we know at this time consistent with profit goals, is there reason to change our past projections? Some financial analysts now prefer to set up with full financial sheets at the beginning and then compare progress against those spreadsheets. Many boxes are blank in the beginning, but will be filled in when we know them. But the profit figures at the bottom of the page are not current forecasts, just current goals. As long as current progress is consistent with those goals, we proceed. A successful cola taste-test is not a reliable indicator of consumers’ ultimate trial. If we get to trial, the taste-test says the chances are we will get repeat business. The life cycle concept of financial analysis enables us to avoid setting up systems where make-or-break decisions rest on one sales forecast or one cost forecast. Reduce Dependence on Poor Forecasts If it is difficult to make sales and profit forecasts, are there ways of avoiding having to make them? Yes, several, and many firms use them, though with precaution. Forecast What You Know This is actually an attitude toward forecasting. Why try to forecast what people in the marketplace will do, if there is no reasonable way we can do so? A blank in a spreadsheet can be filled in with a range of estimates to see where the failure point is. If that is very unlikely, then go ahead. Approve Situations, Not Numbers This is a variation on what was mentioned earlier. Analyze to find what the success factors are, and then look to see if the situation offers them. If so, go ahead, 276 Part Three Concept/Project Evaluation knowing that success should come even though we don’t know just how much. An extreme example of this occurred once when a marketing vice president was asked to predict what he would do if he could get a license to use the Coca-Cola trademark on a line of new products. His answer was, right now I don’t know, but with that trademark it’s only a matter of how much, not whether. One way of betting on a situation has a parallel in horse racing; some betters bet on the jockey, not the horse (about whom they may be able to learn very little). Many firms “bet” on a top-notch scientist, sales force, trademark, or reputation. Another situation variable is leadership. Some firms encourage the champion system. They expect champions to force their way past a restrictive financial system. This makes for a strange but very workable practice of evaluating teams and their leaders, rather than the ideas they come up with. These firms don’t seek great new products; they seek concepts that they can manage into great new products. You may have heard about the movie producer who builds a staff of outstanding creative people and depends on them to work miracles with ordinary scripts. A competitor invests in top scripts instead. But both were avoiding the necessity of relying on complex forecasts and financial analyses. People who love to fish do this all the time; they spend lots of money to find and reach top trout streams. One manager recently said, “If there is a good trout stream with lots of trout in it and a good angler with good equipment, we don’t need an accountant to tell us how many fish we will catch. Whatever happens will be good.” This strategy is not as folksy as it may sound. A firm must know what the success factors are in any situation. One of those two movie producers may be wrong. Notice how the manager included the trout stream, the angler, and the good equipment. Recall from Chapter 8 the two potholes to success identified by Campbell Soup: the taste of the soup and the manufacturing cost. Their name and skills could overcome any other limitations. Precise forecasting wasn’t necessary under this strategy, but being sure on taste and cost was. Commit to a Strategy of Low-Cost Development and Marketing There are times when a company can do the type of product innovation that some call temporary products. Develop a stream of new items that differ very little from those now on the market, insert them into the market without great fanfare, and watch which ones end users rebuy. Drop those that don’t find favor. Japanese and Korean makers of electronics goods do this regularly, with Sony and Samsung introducing several hundred new items in a year; there even are cities in Japan where firms introduce their food and since consumers know this, marketing costs can be kept low. Go Ahead with Sound Forecasts But Prepare to Handle the Risks This strategy especially appeals to managers who feel business is suffering from “paralysis by analysis.” There are lots of ways to put risk back into product innovation while managing it well. One approach is to isolate or neutralize the in-house critics (a strong reason for setting up project matrixes and spinouts). Chapter Eleven Sales Forecasting and Financial Analysis 277 FIGURE 11.7 Percent return Calculating the New Product’s Required Rate of Return Required rate of return Cost of capital Risk Average risk of the firm Risk on a particular proposed product Explanation: Required rate of return (hurdle) = Cost of capital + Risk premium for the new product. Another approach defers financial analysis until later in the development process. One firm realized it was consistently killing off good new product ideas by demanding precise financial analyses at the time of screening. It didn’t have the data. Another strategy is to use market testing rollouts (see Chapter 18). If a financial analysis looks weak, but the idea seems sound, try it out on a limited scale to see where the solution might lie. This thinking may violate several popular management theories (e.g., empowered teams), but it may be necessary at times. Managing risk is a major field in itself today, since we know business needs risk as a source of profits. Figure 11.7 shows the risk situation new product managers face in their evaluations—they know their product will bring more risk than the average risk of the firm, but how much? Conceptually, Figure 11.7 suggests that the riskier the new project is expected to be, the higher the required rate of return should be, but in practice it can be difficult to put real numbers into the diagram. Product managers can borrow from options-pricing theory to make early decisions on product concepts. Real-options analysis may be used to estimate the net present value of a new product when it is still in the concept stage. It accounts for the fact that there still are unknowns at this early stage, and that the firm may need to abandon the project at some time in the future as more information is obtained, and uncertainty is reduced.16 Consider the example detailed in 16 For a good discussion of real-options analysis in financial evaluation of product concepts, see Edward Nelling, “Options and the Analysis of Technology Projects,” in V. K. Narayanan and Gina C. O’Connor (eds.), Encyclopedia of Technology & Innovation Management (Chichester, UK: John Wiley, 2010), Chapter 8. 278 Part Three FIGURE 11.8 Real-Options Analysis of a Product Concept Concept/Project Evaluation Data: Startup costs in Year 0: $70,000. The cash flows for Years 1 through 4 are estimated to be $40,000 in a high-demand scenario, or $10,000 in a low-demand scenario. The probabilities of a high- or low-demand scenario are both 50 percent. The product concept could be abandoned after Year 1, and the equipment could be sold for $38,000. Discount rate 5 12%. Procedure: Begin by assessing cash flow in Year 1 for each demand scenario. Demand Year 1 Year 2 Year 3 Year 4 Total High 40,000 40,000/(1.12) 5 35,714 40,000/(1.12)2 5 31,888 40,000/(1.12)3 5 28,471 $136,073 Low 10,000 10,000/(1.12) 5 8,929 10,000/(1.12)2 5 7,972 10,000/(1.12)3 5 7,118 $34,018 Next, assess cash flow for Year 1 if the option to abandon the project is taken and the equipment is sold: Demand Year 1 Take Option to Abandon and Sell Equipment Low 10,000 38,000 Total $48,000 Since $48,000 . $34,018, management will choose to abandon the project after Year 1. Next, go back to the present (Year 0) and assess NPV for each demand scenario, with the knowledge that management will choose to abandon the project after Year 1 if demand is low. Demand Year 0 Year 1 High 270,000 40,000/(1.12) 5 35,714 Low 270,000 48,000/(1.12) 5 42,857 Year 2 Year 3 2 40,000/(1.12) 5 31,888 40,000/(1.12) 5 28,471 Year 4 3 Total 4 40,000/(1.12) 5 25,421 $51,494 2$27,143 Since each scenario is equally likely to occur, the expected value of the investment is: (0.5)($51,494) 1 (0.5)(227,143) 5 $12,176, and since this expected value is greater than zero, the firm should make the investment. Source: Edward Nelling, “Options and the Analysis of Technology Projects,” in V. K. Narayanan and Gina C. O’Connor (eds.), Encyclopedia of Technology & Innovation Management, Chichester, UK: John Wiley, 2010, Chapter 8. Figure 11.8. A product concept under consideration would incur startup costs of $70,000. Demand is still uncertain; let’s assume a 50-50 chance of generating a cash flow of either $40,000 or $10,000 per year for the next four years, depending on whether demand turns out to be high or low. In the case of low demand, the firm has the option of abandoning the project at the end of the first year and selling the equipment for an estimated $38,000. Assume a discount rate of 12 percent. Chapter Eleven Sales Forecasting and Financial Analysis 279 As shown in Figure 11.8, the way to begin is to calculate net present values as of the end of Year 1, which is when the option would be exercised. The figure shows that, as of the end of Year 1, the NPV is over $136,000 if demand is high, but is only about $34,000 if demand is low. If the firm exercises the option and abandons the project, the NPV increases to $48,000, so the firm will indeed choose to abandon the project one year from now if demand turns out to be low. With this information, we can now go back and calculate the expected value of the concept to the firm. The last part of Figure 11.8 shows that there is a 50 percent chance that demand will be high, and the product will generate an NPV of over $51,000. There is also a 50 percent chance that the demand will be low, in which case the project will be abandoned and the current NPV would be a loss of about $27,000. The expected value of the product concept’s NPV is therefore a little over $12,000. Since this is positive, the firm should go ahead with the investment. The possibility that the project loses money is offset by the firm’s ability to recover some of the investment should the project be abandoned. Use Different Methods of Financial Analysis on New Products, Depending on the Situation Most product innovation (in terms of sheer numbers of items) is singles, not home runs—product improvements and close line extensions. This innovation is managed deep within the ongoing operation, no empowered teams, no huge technical breakthroughs. The item is often demanded by a key customer or key channel, and the decision to develop it is not based on item profitability at all. The risks are relatively small; sometimes the development is in a partnership with a customer who will provide profitable volume. But home runs are something else entirely. They involve big risks and potentially big gains. They need much attention and cannot be handled easily with methods such as those in the previous section. Here, the best approach is to have two systems of financial analysis, one for singles and one for home runs. Alternatively, some firms have no standard system at all, but develop a financial analysis for each project, keying the information to those issues where the risks really lie and unknowns prevail. Improve Current Financial Forecasting Methods For example, marketing people sometimes make use of mathematical sales forecasting models (such as A-T-A-R or something similar). Although many of these models were developed for use on consumer packaged goods, efforts continue to make them work better on durable goods.17 Some firms analyze their own past efforts as well. More progress will come when firms systematically study their most recent 50 (or 25, or whatever) new products to summarize what financial methods were used and how well they forecast the actual outcomes. This is what we now call success/failure analysis, and it leads to best practices. It is rather common 17 See Glen L. Urban, John S. Hulland, and Bruce D. Weinberg, “Premarket Forecasting for New Consumer Durable Goods; Modeling Categorization, Elimination, and Consideration Phenomena,” Journal of Marketing, April 1993, pp. 47–63. 280 Part Three Concept/Project Evaluation FIGURE 11.9 Hurdle Rates on Returns and Other Measures Hurdle Rates Product A B C Strategic Role or Purpose Combat competitive entry Establish foothold in new market Capitalize on existing markets Sales Return on Investment Market Share Increase $3,000,000 $2,000,000 $1,000,000 10% 17% 12% 0 Points 15 Points 1 Point Explanation: This array shows that hurdles should reflect a product’s purpose, or assignment. For example, combating a competitive entry will require more sales than would establishing a toehold in a new market. Also, we might accept a very low share increase for an item that simply capitalized on our existing market position. in other phases of new product work—for example, recall the NewProd screening model of Cooper in Chapter 10. There have also been some improvements in accounting methods. Lastly, some new products managers make a general plea that all financial analysis should be advisory—not fixed hurdles and mandates but flags that warn of potential problems. Of course, hurdle rates can be managed in the sense of being situational (see Figure 11.9). Return to the PIC So far in this chapter, we have focused on financial analysis for a new product project. Before leaving this topic, we should note that many of the very successful product developing firms have realized that financial analysis is not enough. One must also reconsider the PIC and the strategic criteria it implies: For example, does the new product project technology create a new market opportunity or reshape an existing one?18 Firms are increasingly using a combination of financial analysis and PIC considerations when making the tough decisions on which new product projects to commit to. That is, projects need to be considered on how well they fit the firm’s strategy for innovation. As noted above, many firms report that too many new product projects get approved, and the human and financial resources end up getting spread too thin. This can happen for several reasons. Too many projects clear simple financial hurdle rates (such as minimum NPV), and all get approved; resource constraints are not included in the NPV calculations so trade-offs are not made; or low-quality work at the fuzzy front end reduces the quality of information available to managers making Go/No Go decisions. Furthermore, the wrong mix of projects may be undertaken: Management approves several small, quick-hit projects while passing up the opportunity to develop a significant new product platform or technology.19 18 Edward U. Bond, III and Mark B. Houston, “Barriers to Matching New Technologies and Market Opportunities in Established Firms,” Journal of Product Innovation Management, 20(2), 2003, pp. 120–135. 19 See Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, “New Products, New Solutions: Making Portfolio Management More Effective,” Research-Technology Management, March–April 2000, pp. 18–33. Chapter Eleven Sales Forecasting and Financial Analysis 281 These problems may stem from the firm’s reliance on only financial projections when selecting projects. These projections may be unreliable (especially at this early stage in product development) and obviously do not provide any information about how well the project fits the firm’s product innovation charter.20 Recall from earlier discussions that firms that use both strategic and financial criteria in project selection tend to outperform those relying mostly on financial projections. In Chapter 3, the strategic portfolio model for portfolio management was presented. It is typical of a top-down strategic approach—that is, the firm or SBU lays out its strategy first, then allocates funds across different kinds of projects. This approach can clearly be used in project selection. For example, if the firm is already involved in plenty of quick-hit projects, strategic portfolio considerations would indicate that new funding would be better routed to a long-term, major technology development. Management can also take a bottom-up approach to strategy development by building strategic criteria into their project selection tools. The top-performing firms, in fact, often use a combination of top-down and bottom-up approaches and consider strategic as well as financial criteria when selecting projects, while the worst performers tend to rely only on financial criteria.21 Robert Cooper and his colleagues use Hoechst-U.S.’s scoring model as an example of how to balance strategic and financial concerns (see Figure 11.10). Of the five factors shown in the figure, two are clearly full-screen feasibility factors similar to those in Figure 10.5 (Probability of Technical and Commercial Success), one is a financial criterion (Reward), and two are strategic factors related to the firm’s PIC (Business-Strategy Fit and Strategic Leverage). Similarly, Specialty FIGURE 11.10 Key Factors HoechstU.S. Scoring Model Probability of Technical Success Probability of Commercial Success Reward Business-Strategy Fit Strategic Leverage Rating Scale (from 1–10) 1 ... ,20% probability 4 ,25% probability Small R&D independent of business strategy “One-of-a-kind”/ dead end ... 7 ... 10 .90% probability .90% probability Payback , 3 years R&D strongly supports business strategy Many proprietary opportunities Source: Adapted from Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, Portfolio Management for New Products, McMaster University, Hamilton, Ontario, Canada, 1977, pp. 24–28. Reprinted with permission. 20 Randall L. Englund and Robert J. Graham, “From Experience: Linking Projects to Strategy,” Journal of Product Innovation Management, 16(1), January 1999, pp. 52–64. 21 See Robert G. Cooper, “Portfolio Management: Results of New Product Portfolio Management Best Practices Study,” in L. W. Murray (ed.), Maximizing the Return on Product Development, Proceedings of the 1997 PDMA Research Conference, Monterey, CA, pp. 331–358. 282 Part Three Concept/Project Evaluation Minerals, a spin-off of Pfizer, uses a 7-point scoring model that shows a similar combination of financial and strategic considerations: • • • • • • • Management interest Customer interest Sustainability of competitive advantage Technical feasibility Business case strength Fit with core competencies Profitability and impact As another example, the screening criteria used by a real manufacturing company (whose identity was not revealed) included: • • • • Net Present Value Internal Rate of Return Strategic Importance of Project (how the project aligns with business strategy) Probability of Technical Success Again, one criterion (the third) is clearly a measure of fit with PIC (albeit using slightly different terms), while the others are related to technical feasibility or financial expectations.22 Research into this subject continues, but so far the results are suggesting that consideration of strategic as well as financial criteria is important when assessing new product projects. Finally, Figure 11.11 presents the advice of Erika Seamon of Kuczmarski and Associates, a well-respected consultant group. This consultancy clearly also advocates FIGURE 11.11 A Tool for Concept Evaluation Dimension Sample Questions Strategic Fit Does the concept fit with corporate vision? Does the concept fit with our sales force? Does the concept allow the customer to better meet consumer needs? Does the concept have a good value as perceived by the customer? Does the concept satisfy an unmet or latent consumer need? Will consumer loyalty be increased? Is the concept unique relative to the competition? Could our firm be a Number 1 or Number 2 competitor? Is the concept feasible? Is the concept protectable? Will the project break even soon? Will the project achieve required earnings in the desired time? Customer Fit Consumer Fit Market Attractiveness Technical Feasibility Financial Returns Source: From Erika B. Seamon, “Achieving Growth through an Innovative Culture,” in P. Belliveau, A. Griffin, and S. M. Somermeyer, The PDMA Toolbook 2 for New Product Development, John Wiley & Sons, Inc., 2004, Ch. 1. 22 The Hoechst, Pfizer, and manufacturing firm examples are from Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, Portfolio Management for New Products (Hamilton, Ontario: McMaster University, 1997), pp. 22–29. Chapter Eleven Sales Forecasting and Financial Analysis 283 that firms should consider both strategic and financial criteria when deciding which concepts to move into prototype development: their concept evaluation tool includes several ways to assess strategic fit and market attractiveness, and also considers financial performance. Summary This chapter has dealt with the matter of how to make judgments on the financial merits of new products. It has also explored in depth the issue of sales forecasting, since this is one area where the new product team usually relies heavily on the expertise brought in by the marketing representative. There are good basic methods for doing financial analysis (net present value calculations using discounted cash flow) and for doing sales forecasting. Most firms use these daily. However, new products managers know they often do not have the data these sophisticated methods require. So they may also need to use “risk-reducers”—actions that give them nonquantitative guides to probable success. The method of making financial analyses is given in the Bay City case, which comes after the Applications section. The case offers data for a new electronics product and gives the opportunity as well to look at some nondata issues involved in financial analysis. At this point in the new product development process, we are ready to begin Phase IV (development). Applications 1. “You’re still a student, but when you tell me about all the problems new product managers have putting together financial worksheets, you sound like the people we have around here. They complain that all finance numbers are unreliable, estimates, guesses, etc. What they really want is no financial appraisal at all—just leave them alone and they’ll eventually bring back the bacon—big slabs of it. That’s simply not true—our financial evaluations have a weak figure or two, sure, but how else can we keep reasonable managerial control over the use of sometimes very great corporate resources?” 2. “One thing I know for certain—I don’t want any sales managers or technical research people making new product forecasts. I’ve never seen such lousy forecasting as we get from these people. Sales managers either love a new item so much they think it will outsell everyone, or they think it is a dud and underforecast just as badly. Absolutely no objectivity in them. And the technical people, well, they become so enamored with their inventions that they lose all objectivity too. What I like is forecasting done by independent people—project managers or new products managers in separate departments. Have you run into any good ways of keeping sales managers and technical researchers out of forecasting? You agree that they should be excluded, don’t you?” 284 Part Three Concept/Project Evaluation 3. “Actually, I agree with one thing you said a while ago, and that related to the desirability of making financial analyses on a threshold basis. I realize how many unknowns there are in the new products business. As a president, I realize too that most of the financial projections I read are just air. If a new products group can convince me that they can sell at least X volume, and at that volume their costs will be Y, or lower, then I am inclined to go along with them. But, deep in my heart, I don’t like it—those thresholds are just as much subject to manipulation as are the more structured NPV projections. You agree?” Case: Bay City Electronics23 Financial analysis of new products at Bay City Electronics had always been rather informal. Bill Roberts, who founded the firm in 1970, knew residential electronics because he had worked for almost seven years for another firm specializing in home security systems. But he had never been trained in financial analysis. In fact, all he knew was what the bank had asked for every time he went to discuss his line of credit. Bay City had about 45 full-time employees (plus a seasonal factory workforce) and did in the neighborhood of $18 million in sales. His products all related to home security and were sold by his sales manager, who worked with a group of manufacturers’ reps, who in turn called on wholesalers, hardware and department store chains, and other large retailers. He did some consumer advertising, but not much. Bill was inventive, however, and had built the business primarily by coming up with new techniques. His latest device was a remote-controlled electronic closure for any door in the home. The closure was effected by a special ringing of the telephone: For example, if a user wanted to leave a back door open until 9:00 p.m., it was simple to call the house at 9:00 and wait for 10 rings, after which the electronic device would switch the door to a locked position. A similar call would reopen the door. The bank liked the idea but wanted Bill to do a better job of financial analysis, so the loan officer asked him to use the forms shown in the Bay City Appendix as Figure 11.12 and Figure 11.13. After some effort, Bill was able to fill out the key data form, Figure 11.12, and his work is reproduced here. To date, Bay City had spent $85,000 in expense money for supplies and labor developing the closure and had invested $15,000 in a machine (asset). If the company decided to go ahead, it would have to invest $50,000 more in a new facility, continue R&D to validate and improve the product, and—if things went according to expectations—invest another $45,000 in year 3 to expand production capability. He also had to fill out the financial worksheet, Figure 11.13; for this he used a friend of the family who had studied financial analysis in college. The friend had relied on a summary of how to do this, and this summary is attached. He also warned Bill that there were lots of judgment calls in that calculation, “so don’t get into an argument with the people at the bank about details.” 23 This is a realistic, but hypothetical, situation. Chapter Eleven Sales Forecasting and Financial Analysis 285 FIGURE 11.12 Key Data Form for Financial Analysis, Part A Financial Analysis Proposal: Bay City Electronics Closure* Date of this analysis: Previous analysis: 1. Economic conditions, if relevant: Corporate scenario OK 2. The market (category): Stable–5% growth 4. List price: $90 Distributor discounts: $36 Net to factory: $54 5. Production costs: Explanation of any unique costing procedures being used: None. Experience curve effect. 3. Product life 5 years Other discounts: Promotion: $1 Quantity: $1 Average dollars per unit sold: $52 Applicable rate for indirect manufacturing costs: 20% of direct costs 6. Future expenditures, other capital investments, or extraordinary expenditures: Build production facilities: $50,000 Ongoing R&D: $15,000; $10,000; $15,000; $10,000 for first four years after intro Special UL test during the 2nd year will cost $5,000 Expand facilities in 3rd year for $45,000 7. Working capital: 35 % of sales 10% inventory; recover 80% in period 5 15% receivables; all recovered in period 5 10% cash, all recovered 8. Applicable overheads: Corp.: 10 % of sales Division: – % of sales 9. Net loss on cannibalized sales, if any, expressed as a percent of the new product's sales: 10 % 10. Future costs/revenues of project abandonment, if that were done instead of marketing: Abort now would net $3,000 from sale of machine. 11. Tax credits, if any, on new assets or expenditures: 1% of taxes due to state and federal, based on positive environmental effect. 12. Applicable depreciation rate(s) on depreciable assets: 25% on orig. plant and machines; 33 1/ 3% on expansion facilities 13. Federal and state income tax rate applicable: 34 % Comments: 14. Applicable cost of capital: 16 % ± Premiums or penalties: high-risk project 8 % Any change in cost of capital anticipated over life of product? No *This key data form is filled in with demonstration data for the Bay City Electronics case. 286 Part Three Concept/Project Evaluation FIGURE 11.12 (CONCLUDED) Key Data Form for Financial Analysis, Part A 15. Basic overall risk curve applicable to the NPV: 16. Key elements to be given sensitivity testing (e.g., sales, price cuts): Standard OK 17. Sunk costs: Expenses to date: Ignore Capital invested to date: $15,000 (see below) 18. Elements of new product strategy that are especially relevant on this proposal: (e.g., diversification mandate or cash risk): Strategy calls for us to strengthen company in diversified markets, which this product will do. 19. Basic sales and cost: Year Unit sales Direct production cost per unit Marketing expenses 1 2 3 4 5 14,000 10,000 18,000 24,000 15,000 $16 $12 $11 $09 $14 $100,000 $$80,000 $$50,000 $$60,000 $$10,000 20. Hurdle rates: Must have 40% gross margin after production costs. 21. Any mandatory contingencies: None 22. Other special assumptions or guidelines: (1) The total $110,000 of facilities and machines will salvage for $10,000 when production is finished. (2) The firm has other income to absorb any tax loss on this project. (3) Ignore investment tax credit. Sensitivity testing (Calculate the effect on NPV of the following): (1) We may have to cut the price to $34 net at start of third year. (2) Our direct manufacturing cost estimate may be overly optimistic. What if we never get the cost below the original $16? (3) Competition may force much higher marketing costs–what if starting in year 2 the level we have to spend at is just twice what we forecasted above? (4) How about a worst-case outcome, in which all of the above three contingencies are tested at one time? Chapter Eleven Sales Forecasting and Financial Analysis 287 FIGURE 11.13 Financial Worksheet, Bay City Electronics Product Proposal: Electronic Closure Date: Years of the Market Unit sales Revenue per unit Dollar sales Production costs: Direct Indirect Total Gross profit Direct marketing costs Profit contribution Overheads (excluding R&D): Division Corporate Total Other expenses: Depreciation Cannibalization R&D to be incurred Extraordinary expense Project abandonment Total Overheads and expenses Income before taxes Tax effect: Taxes on income Tax credits Total effect Cash flow: Income after taxes Depreciation Production facilities Working capital: Cash Working capital: Inventories Working capital: Acc. Rec. Net cash flows Discounted flows Net present value Internal rate of return Payback Test 1: NPV 5 $88,885 Test 2: NPV 5 $149,453 Test 3: NPV 5 $196,013 All 3: NPV 5 ($99,699) 0 0 0 0 1 4,000 52 208,000 2 10,000 52 520,000 3 18,000 52 936,000 4 24,000 52 1,248,000 5 5,000 52 266,000 0 0 0 0 0 0 64,000 12,800 76,800 131,200 100,000 31,200 120,000 24,000 144,000 376,000 80,000 296,000 198,000 39,600 237,600 648,400 50,000 648,400 216,000 43,200 259,200 988,000 60,000 928,800 70,000 14,000 84,000 176,000 10,000 166,000 0 0 0 0 20,800 20,800 0 52,000 52,000 0 93,600 93,600 0 124,800 124,800 0 26,000 26,000 16,250 0 0 0 3,000 19,250 19,250 (19,250) 16,250 20,800 15,000 0 0 52,050 72,850 (41,650) 16,250 52,000 10,000 5,000 0 83,250 135,250 160,750 31,250 93,600 15,000 0 0 139,850 233,450 414,950 15,000 124,800 10,000 0 0 149,800 274,600 654,200 15,000 26,000 0 0 0 41,000 67,000 99,000 (6,545) (65) (6,480) (14,161) (142) (14,019) 54,655 547 54,108 141,083 1,411 139,672 222,428 2,224 220,204 33,660 337 33,323 (12,770) 16,250 50,000 0 0 0 (46,520) (46,520) $267,025 73.7 Nov., Year 3 (27,631) 16,250 0 20,800 20,800 31,200 (84,181) (67,888) 106,642 16,250 0 31,200 31,200 46,800 13,692 8,904 275,278 31,250 45,000 41,600 41,600 62,400 115,928 60,803 433,996 15,000 0 31,200 31,200 46,800 339,796 143,725 65,677 15,000 0 (124,800) (99,840) (187,200) 492,517 168,001 Worst case is very undesirable, even here where indirect effects, sunk costs, and salvage were omitted. 288 Part Three Concept/Project Evaluation While waiting for his appointment at the bank, Bill spent some time just thinking about his situation. Did the numbers look good? Where were the shaky parts that the banker might give him trouble on? Most of all, he was curious about whether a friend of his at the LazyBoy chair firm in Monroe had to do the same thing, and would 3M require the same type of form from his daughter, who now worked for them? Frankly, he didn’t feel he personally had learned much about his situation from the exercise and was already wondering whether there weren’t better ways for him to go about reassuring the bank that their loan was a good proposition. BAY CITY APPENDIX: FINANCIAL ANALYSIS FOR NEW PRODUCTS New products financial analysis requires two separate activities: (1) gathering the full set of data and other “givens” in the situation, and (2) using them in calculations to derive whatever final figure is sought. These two tasks are shown in Figures 11.12 (the key data form) and 11.13 (the financial worksheet). COMPILING THE KEY DATA Economic Conditions. Most firms have ongoing economic forecasts, but sometimes a team wishes to differ. If so, the difference should be noted here. The Market or Category. The “market” for the new product is defined carefully, and the growth rate assumption is noted. Also, the current total market unit and dollar volumes are recorded. Product Life. The number of years used in the economic analysis of new products is usually set by company policy, but any particular project may be an exception. Pricing. Start with the end-user list price, work back through the various trade discounts to get a factory net, then deduct any planned special discounts and allowances. The average dollars per unit sold is the price used in worksheet calculations. Production Costs. Is anything unusual being done on this project? Actual anticipated cost goes directly onto the financial worksheet. Cite factory burden percentage rate.24 Future Special Expenditures. These typically include factory facilities, licensing rights, the one-time introductory marketing cost, up-front payments to suppliers, further R&D on improvements and line extensions, and plant expansions as volume grows. These are all investment outflows. Working Capital. This estimates cash, inventories, and receivables needed to support the sales volumes. How are they to be recovered? 24 Factory overheads are often assigned using an activity-based costing (ABC) system. If adopted, new items have a greater chance of realistic allocations. See Bernard C. Reimann, “Challenging Conventional Wisdom: Corporate Strategies That Work,” Planning Review, November/December 1991, pp. 36–39. Chapter Eleven Sales Forecasting and Financial Analysis 289 Applicable Overheads. Some firms assign only “direct” overheads—those caused by the new product (such as an expanded sales force or a new quality function). Other firms believe overheads tend to grow as functions of volume and should be included. Net Loss on Cannibalized Sales. These are dollar sales lost as the new product steals sales from current products. This is to be deducted from revenue. Some experts believe if we don’t do this a competitor will, so they omit it. Future Costs/Revenues of Project Abandonment. Along the way, the project may have accumulated facilities, people, patent rights, inventories, and so on. If abandoned now, disposal of these will produce revenue, money that is actually a cost of abandoning the project. But disposing of radioactive chemicals may be expensive, thus a revenue of going ahead. Tax Credits. Federal or state incentives for activity in the public interest. Applicable Depreciation Rate. Policy question, set by management. Federal and State Income Tax Rate. Company figure, provided. Required Rate of Return. This one tells us the cash flow discount rate to be used and can be complex and political. Theoretically, the figure to use is the weighted average cost of capital, including the three sources of capital—debt, preferred stock, and retained earnings. Often it is simply the firm’s current borrowing rate.25 It may be the rate of earnings from current operations. New product managers want it low, conservative financial people may want it high. The actual rate to be used is often an arbitrary decision. Whatever the rate, the next step is to decide how the riskiness of this project compares with the rest of the firm’s activities. Look at Figure 11.7, which shows that a relationship between risk and rate of return exists for every business, as discussed in the chapter. Given the current average cost of capital and the level and slope of the line, the manager can mark off the risk of the particular new product, go up to the risk/return line, and then read off the required rate of return. Except in unusual circumstances, that required level will represent a premium over the current cost of capital. The premium is entered in section 14 of the key data form. Risk Curve. Figure 11.9 shows the typical risk curve of possible profit outcomes from a given new product project, as discussed in the chapter. In the B pattern, for example, chances are the project will have a lower payout, but a very high payout is also possible. Imitative competition is expected; but, if it doesn’t come, the profit will be high. This risk pattern information is good to keep in mind when making the financial analysis, though few firms undertake the probability-adjusted risk analysis it permits. Sensitivity Testing. After an analysis has been completed using original data, the analyst goes back and recalculates the profit using other figures for especially sensitive factors. 25 A variation on this is to use the current market risk-free cost of capital (interest rate on Treasuries, for example). We then add a premium reflecting the general level of risk in the industry at hand. 290 Part Three Concept/Project Evaluation Elements of Strategy. When evaluating new product proposals, it is important to remember the strategy that prompted them. Less-profitable products may well be warranted under certain strategies. Basic Sales and Cost Forecasts. This section gives the primary data inputs— the number of units to be sold, the direct production cost per unit, and the total marketing expenditures. Hurdle Rates. A company sometimes has hurdle rates on variables other than rate of return. Mandatory Contingencies. A firm may want one or more contingencies worked into the analysis every time, not left optional. Other Special Assumptions or Guidelines. This is the typical miscellaneous section, totally situational. Beyond the Key Data Form: Sunk Costs. Sunk costs should not enter into this analysis. Sunk money is just that—sunk. It stays sunk whether we go ahead at this time or abandon the project. Salvage. NPV forms sometimes call for the dollars obtained at the end of the product’s life from sale of salvaged equipment. The amounts are usually small and are best omitted. Portfolio. If the new item is playing a special role as part of an overall portfolio of projects, the value of that role should be mentioned. The new project may be high risk but still worthwhile to balance a large number of low-risk projects—or the reverse. C H A P T E R T W E L V E Product Protocol Setting When a new products group finishes the full screen and the associated financial analysis, they have reached what many feel is the most critical single step in the new product’s life—more critical than the market introduction and more critical than even the building of manufacturing capacity. This is the point where very important things all around the firm begin to happen. Granted, some managements still use a relay race system, where one department does its work, passes the product concept to the next department, which does its work, and so on. The leading product innovators do not follow the relay race model: they use some type of concurrent system, one in which all of the players begin working, doing as much as they can at any time as the project rolls along. When technical work begins, process engineers are not sitting around waiting for the final prototype to be tossed to them. When process engineers are laying out the manufacturing system, procurement people are not waiting for final word about when certain components are going to be built. And while all of this technical/operations work is continuing, marketing people are not idly waiting for a hand-off that will trigger their thoughts about advertising and customer technical service. At the most innovative firms, all product team members begin work at the same time, and in fact many have been watching the concept testing and screening to see how positive the early word is. If a concept looks like a winner, even if financial screening won’t take place for a few months, these down-the-line people are already starting to do what they will eventually have to do. Some workers actually may be a year ahead of need, especially if there is some built-in delay in what they do. For example, while process engineers are waiting for product specs, so they can begin ordering cost-effective components and developing appropriate manufacturing systems, packaging people have been thinking about the concept. Many products require packaging—durable, value-producing packaging, or impressive, shelf-talking promotion packaging. Packages, in turn, require product names. So purchasing cannot order new packages until brands are settled, and brands cannot be settled until product content is known and marketing strategy is settled. Marketing strategy involves price decisions, which must await costs, which must await final manufacturing systems and component costs, which is where we started at the 292 Part Three Concept/Project Evaluation beginning of this paragraph! Clearly, coordination of all parties involved will be critical to the success of the new product project, otherwise things can go very wrong. The Product Protocol What do we do? We do it all, side by side, doing what we can, when we can, making minor commitments at some risk, holding on costly commitments. All of these efforts are risky and will never work well without something that keeps the team together, something that allows them to make reasonable speculations. That something currently has no standard form, no accepted name, and no established practice. But most firms are doing part of the task, a few all of it, waiting for the activity to gel. In this book we will call the activity protocol preparation, and the output is a product protocol. Other names that it goes by are product requirements, product definition, and deliverables. All terms mean the same thing—what is the final package of output from the development system, what benefits or performance will the product deliver to the customer, and what changes will the marketing program bring in the marketplace. A protocol is, by definition, a signed agreement between negotiating parties. In a product protocol, the negotiating parties are the functions—marketing, technical, operations, and others. Signed agreement is a bit formal, perhaps, but the financial analysis that triggered this phase depended on certain assumptions— product qualities and costs, certain support facilities, certain patents, and certain marketplace accomplishments. If they are not delivered, all bets with management are off. Since most projects today involve some form of multifunctional team, the whole group is responsible for writing a protocol. Although new products do indeed require trade-offs, they are negotiated in a very positive use of the term. Even if the multifunctional team works well together, technical limitations may emerge that may make quick agreement difficult. A humorous view of the kind of challenges that can crop up at this phase is presented in Figure 12.1. FIGURE 12.1 A MarketingR&D Conversation MKTG: We’re going to be needing a solar-powered version of our standard garage door opener, soon. R&D: How reliable should it be? Should it be controllable from inside the house? Should we use new electronics technology? Should it be separate from the collector system already installed? MKTG: Well, you’re the technical people, make some recommendations. R&D: In other words, you don’t know what you want. MKTG: Cripes, do we have to tell you everything? What do you do for a living? How should we know where the collectors should be located? R&D: If we go electronic, you’ll say it’s too expensive. If we go electric, you’ll say we’re living in the ’30s. Wherever we put the collectors you will say we are wrong. If we guess, you second-guess. MKTG: OK. Put the collectors on the garage roof. R&D: That probably can’t be done. Chapter Twelve FIGURE 12.2 The Oobeya Room Product Protocol 293 One tool used by Toyota to speed up its product development is the Oobeya Room. “Oobeya” (from the word for “room” in Japanese, pronounced oh-beya) is, indeed, a big room, set up to accommodate the entire team for one new product project (usually containing people from marketing and sales, engineering, logistics, planning, design, and production). In the center of the Oobeya room is the prototype (a model, mock-up, or drawing) that encourages communication among team members, and helps the team visualize the product and identify potential problems early. All around the room are boards that guide discussion of the product project. These would include: An objective board (containing Toyota’s version of a PIC: background, objectives, technical specifications, and project organization) A metrics board (showing current project status and allowing for team members to determine where they are at, or behind, target) An action board (showing the activities of all members on the team that are required to reach the objective and indicating which activities are already completed) A decomposition board (indicating which sub-projects need the most attention) An issue board (showing the most critical problems that have arisen, used to stimulate discussion between team leader and team members and to assign accountability) An important part of the Oobeya Room concept is the roles of all the participants. The team leader is responsible for setting targets, assessing team member plans, negotiating with the team or with company management when the goals are not realistic, and keeping meetings under control. Team members’ responsibilities include providing solutions that help the team attain desired goals, providing status reports (on target versus behind target), suggesting how to overcome obstacles, understanding the activities of other team members, and resolving key issues. Generally, at each meeting, team members are expected to make a short presentation about their areas. The more experience they have, the easier it is for team members to keep their reports under three minutes. Total meeting time, including review of main and issue boards, usually is under one hour. The Oobeya Room concept seems rather simple, but it is in fact a powerful tool. The payoff comes from the fact that the process requires the team members to integrate their behaviors, and to work in a very efficient and structured manner. That way, more information is generated and less time is needed. Problem detection and resolution speeds up and the value of each individual meeting is increased. No one can “loaf,” read reports, or send e-mails, since there are strict time constraints. To get the work done well and on time requires a real commitment to collaboration and interaction. Source: Toshi Horikiri, Don Kieffer, Takashi Tanaka, and Craig Flynn, “A Toyota Secret Revealed: The Oobeya Room—How Toyota Uses This Concept to Speed Up Product Development,” Visions, 33(2), July 2009, pp. 9–13. One technique used by Toyota to get cooperation across functional areas, to speed up integration, and to focus the team is the “Oobeya Room,” described in detail in Figure 12.2. The Oobeya Room is conceptually very close to the idea of the product protocol: It very effectively overcomes the challenges shown in Figure 12.1 by giving the team members little choice but to work together. Protocol preparation is the subject of this chapter. In prior chapters of this book you had a chance to see the new product process from an overall perspective— how it goes from strategy to market success, how the strategy gives the process focus, how concepts are created and gathered, how concepts are then tested and 294 Part Three Concept/Project Evaluation FIGURE 12.3 The Integrating and Focusing Role of Protocol Full screen Other projects Protocol Development (Chapters 13–15) Other projects Launch (Chapters 16–20) Core benefit Formal product Augmentation evaluated, and how the evaluation process comes to a temporary conclusion with the full screen and financial analysis. Figure 12.3 shows what happens now. In the middle of the figure lies a bull’seye-like circle representing the augmented product concept. This shows that at the core of a product is end-user benefit, the real purpose for which the product was created. Purposes of the Protocol This can vary from market segment to market segment and from time to time. What the customer actually buys, however, consists of one or more core benefits, a formal product presentation (physical form or service sequence), and an augmentation of things from presale technical service to a money-back guarantee. The point here is that customers and end users buy fully augmented products, and their core benefit may partly come from the augmentations. New products managers cannot focus only on the formal product. All three of the concentric rings of the bull’s-eye must be designed and executed, and two functional groups play a role in all of them, as shown by the arrows leading into the augmented circles. Figure 12.3 also shows that the technical departments (with help from manufacturing, quality, procurement, and others) work essentially as a unit, and marketing Chapter Twelve Product Protocol 295 (with help from its allies in sales, market research, promotion, channel management, and others) does the same on the right side of the diagram. Both groups keep in close touch with each other. The issue is: What do these two groups need to do their work? The answer differs by firm and industry and situation, but whatever, it should be consolidated into a protocol statement. The protocol is, in fact, one step in the evolution of a concept, as you saw in Figure 2.3 of Chapter 2. It is more than the simple statement approved in the screening, and less than what will exist when the first prototype appears. But it is what we need now, what all departments need to begin their work. This idea of how others use the protocol is what gave it the name product deliverables. In fact, the first general purpose of the protocol is to specify what each department will deliver to the final product that the customer buys. For a new type of golf footwear, a deliverable from technical might be “Can be used in all types of weather and on all turf conditions.” A deliverable from marketing might be personal trial use by at least 80 percent of the golf professionals in Europe, the United States, Australia, and South Africa. A deliverable from information technology might be “800 number service with less than five minutes’ waiting time, covering the needs of 80 percent of callers from the United States this year and from the other markets by end of second year.” Not all deliverables are known at this time, of course, but the critical ones should be. Otherwise, we are not ready for release into a system of parallel (or concurrent) development. If, on that golf footwear, we don’t know the importance of bad weather and turf conditions, the golf pro’s influence on affluent golfers (what we are producing we can see will have to be expensive), the criticalness of trial (key benefits will be hidden), and the certainty of technical questions on a complex product like this, then we haven’t done our homework. The fact is, protocol (like many things in use today) states requirements that force us to do what we should be doing anyway, such as good market research! In Chapter 3, on the product innovation charter, you read that the PIC is used to provide clarity of direction to the new product team. A second general purpose of the protocol statement is the same for the participants in new product development. It communicates essentials to all of the players, helps lead them into integrated actions, helps direct outcomes that are consistent with the full screen and financials, and gives all players their targets to shoot for. Some new products people think the mere call for the document leads to early customer contacts that should always be made, but often aren’t. A third purpose of the protocol relates to time through the process, or cycle time. As seen in Chapter 2, many firms place high priority on accelerated time to market, and better product definition can help cut development time. Consider how much development time would be wasted, and how many costly steps would have to be redone, if a new portable MP3 player, intended to take market share away from the iPod, were fully planned, designed, and in prototype production when someone noticed it was too heavy for normal use! It would have been better to specify the desired and maximum weight before undertaking development. Seemingly small decisions like this at product definition, if done wrong, can result in extremely costly fixups late in the process. 296 Part Three Concept/Project Evaluation Fourth, if done right, the protocol gives requirements in words that can usually be measured. It thus permits a development process to be managed. It tells what is to be done, when and why, the how (if that is required by some power beyond our control), the who, and perhaps most important, the whether. That is, we know at any time whether the requirements have been met; this will automatically caution that we are not ready to market an item if there is still an open requirement, unless specifically waived. Many of the techniques we have learned in preceding chapters (perceptual gap analysis, preference mapping, conjoint analysis) will provide us with information that can be used as inputs to various “requirements” of the protocol. Protocol’s Specific Contents You have just read what is in a protocol, in general terms. The details can vary greatly and will for some time until our practice on this new step tells just how to do it. But we should specify guidelines for product development. This might include some physical characteristics, but will at the very least include customer attributes or benefits sought (sometimes called an “I want” list). Do you recall the Hapifork from Chapter 4, which was designed to make people eat slower? The “I want” list would include: • Same shape and weight as a regular fork; handle should not be much bigger than normal. • Emits a signal if the person is eating too quickly. • Washable and dishwasher safe. • Resistant, not fragile. • Comes in a wide variety of colors to match any décor. • Can connect via Bluetooth or USB to other devices to record and track progress. • Easy to use. • Comes with information on diet and exercise.1 Note that some of those customer needs seem simple to say, but might require much work on the technical side. What’s the power source for the fork? If AA batteries are mounted in the handle, the handle then becomes too large. What kind of signal would be best, a beeping sound or vibration, or both? And making it dishwasher safe poses its own challenges in terms of waterproofing. These are challenging but required tasks for R&D when designing the new fork. But R&D is not guessing at the specifications—they know that if they can achieve these objectives, it will result in a product customers will want. Also, not everything that we call for must be delivered. Some firms use the terms Musts and Wants—that is, some requirements we must have, and some are simply what we would like to have if feasible and practical within technology, cost, and time frames. The following sections list items often found in protocols. An abbreviated version of a simple protocol for a new home trash disposal system is given in Figure 12.4. 1 See www.hapifork.com. Chapter Twelve Product Protocol 297 FIGURE 12.4 Simplified Protocol for a Home Trash Disposal/Recycling System 1. Target market: Ultimate: Top 30% of income group, in cities of over 100,000, with upscale lifestyle. Intermediate: Stakeholders in building industry for homes over $300,000, especially developers, architects, builders, bankers, and regulators. 2. Product positioning: A convenient, mess-free method for recycling items in the home. 3. Product attributes (benefits if possible): • The system must automate trash disposal in a home environment with recycling (separating trash, compacting, placing bags outside, and rebagging the empty bins and notifying user when the bag supply is running out) at a factory cost not to exceed $800. • The system must be clean, ventilated, and odor-free. The user will want an easy-to-clean appliance. Rodents, pets, and angry neighbors could become a problem if odors exist. • Installation must be simple. Distributors and other installation personnel must have favorable experience in installations. • The system must be safe enough for operation by children of school age. • The entire working unit must not be larger in cubic feet than twice a 22 cubic foot refrigerator. 4. Competitive comparison: None: First of a kind. 5. Augmentation dimensions: Financing arrangeable with us, if necessary. Generous warranty. Competent installation service, and fast/ competent post-installation service. Education about recycling and about the product will be difficult and essential. 6. Timing: Being right overrides getting to market fast. But the window will not be open more than two years. 7. Marketing requirements: • Marketing announcement must be made at national builders shows and environment/ecology shows. • A new channel structure will be needed for the intermediate target market, but it will eventually be collapsed into our regular channel. • We will need a small, select sales force for this introduction. • To capitalize on announcement value, we need 50 installations during the first four months. 8. Financial requirements: • Development and intro period losses will not exceed $20,000,000. Break-even is expected by end of second full year on the market. • Ultimately, this project must achieve a five-year net present value of zero, based on 35% cost of capital. 9. Production requirements: • Once we announce, there must be no interruption of supply. • Quality standards simply must be met, without exception. 10. Regulatory requirements: Regulations are from many sources and vary by states and localities. There are various substakeholders here; we need to know them well. A surprise, significant holdup (after launch) cannot be allowed on this development. 11. Corporate strategy requirements: Corporate strategy is driving this project, and has personal leadership at the corporate general management level. We seek diversification of markets, enhanced reputation for innovativeness, and sustainable margins higher than those in our major markets today. 12. Potholes: This project has massive pothole potentials, because of its newness. The most worrisome ones are (1) regulatory approval of health issues, (2) accomplishing the $800 cost constraint, and (3) getting fast market approvals for early installations. 298 Part Three Concept/Project Evaluation This figure shows the targeted customer benefits or attributes for the new system, as well as marketing, technical, financial, and corporate requirements and other items usually specified in the protocol. Target Market Most firms manage most of their new product projects using techniques we have been presenting: PICs, concept testing, screening models, protocols, and so on. Other projects are wildcatting—betting on a technology that hasn’t yet been shown to work, betting on a new application where some end-user will partner with us to see what works, or just betting on a scientist with a good track record for coming up with saleable new products. None of these is appropriate for a protocol; we just don’t have the knowledge to write one, and its only effect would be to bother the developers, who, actually, will ignore it completely. In most cases, however, we know the target market very well—first in finding their problems to solve, later in asking if our new product concept meets their need and seems reasonable to them, and still later in screening factors (e.g., do we have a sales force that can reach them or will we have to build a new one?). Perceptual and preference mapping techniques we discussed in earlier chapters can be very helpful in developing this part of the protocol, as benefit segments will have been identified and their specific needs will be understood. The target market needs to be spelled out here, quite specifically. Some firms like to have a primary target market, selected perhaps due to size, growth rate, urgency of need, buying power, perceived ease of making competitive inroads, and so on. Typically, one or more secondary target markets will also be selected to move to after successful introduction, and at least one fall-back target market if the primary does not work out due to technical failure, regulation, competition, or other reasons. Positioning This is a real challenge for many firms. Product positioning is the concept that came out of the advertising world in the early 1970s. Essentially, it says, “Product X is better for your use than other products because . . . .” It announces the item as new and gives the end user a real reason for trying it. In the process, it shows the end user what problem it attacks and what about it makes it better than whatever they are using now. This concept will be developed more completely when we get to Chapter 16, but for now it is usually enough to state the target market and complete that sentence above. Fortunately, this should be easy because joint space mapping and other concept testing activities will have provided key information on desirable positioning options for our product. In effect, the concept test assures us stakeholders will be interested in trying an item and a positioning claim. Technical people are often not told what the positioning of a new item will be. It’s almost as though we say, develop a new item and do it in a way the customer will like. That’s not management; that’s abdication. Even in large packaged goods firms today, with their excellent staffs, products a bit off the beaten track often get neglected; many of these firms’ R&D staffs have had to build market research departments to do concept testing on items they are originating. Misunderstandings Chapter Twelve Product Protocol 299 on positioning have probably been the cause of more technical/marketing fights than anything else. Product Attributes As discussed earlier, product attributes define the product. They are of three types—features, functions, and benefits. Benefits include uses. Protocols can list any of these, and do. “The new bulk laxative will dissolve completely in a four-ounce glass of water in 10 seconds” (Merrill-Dow). This is function—how the item will work, not what it is (feature) or what the benefit is of fast dissolving. Note that by being asked for speed, technical people were allowed to select any chemical they wished (and did—it is now second only to Metamucil in this market). Benefits Benefits are the most desirable form for a protocol to use—better than functions or features. Information obtained from conjoint (trade-off) analysis and other concept testing techniques can be extremely useful in determining what combinations of features, functions, and specifications ought to be built into the product. An advantage of specifying the protocol in terms of benefits is that it places no (or very few) constraints on the R&D staff: They are given free rein to figure out how best to design the product so that it provides the desired benefit. Consider Built NY, which is a small design firm. A friend of the firm’s owners suggested an idea for a new product: a convenient carrier for two wine bottles, which could be brought to a BYOB restaurant. The designers quickly developed a list of customer benefits for the ideal two-bottle wine carrier: protective (so the bottles wouldn’t break), insulating (to maintain temperature), ergonomic (easy to carry), lightweight, reusable, inexpensive, maybe also flexible (easy to store when not in use). The challenge then was to select the material that could best deliver all these benefits. They hit upon neoprene, a synthetic material most commonly used for wetsuits. It offered all of the customer benefits, and also turned out to be easy to cut into shape and to dye into designer colors. The “Two-Bottle Tote” won awards for product design and also inspired a range of similar products, such as beer carriers and baby-bottle carriers.2 Functions Function attributes sometimes cause confusion. Marketers tend to use them a lot, and they are often called performance specs, or performance parameters, or design parameters. One everyone knows is: “The car must accelerate from 0 to 60 miles per hour in 8 seconds.” This requirement does not tell us what features will yield that performance. What it does is answer the question of how the customer achieves the benefits of exciting (or safe) start-ups. Some people feel a performance parameter (a function) may come to be expressed as a design parameter. For example, on the matter of the car pickup above, the statement might be “Use the new German 11-Z4 engine.” Such a new engine would be a technology but clearly might be a solution to a need, not a description 2 The firm’s Web site is www.builtny.com. 300 Part Three Concept/Project Evaluation of it; there are probably many other ways rapid pickup could be achieved. Car platforms are heavily laced with such statements. Protocols for services are especially likely to be in performance terms since the production of a service is a performance, not a good. But protocols are also much less necessary on services because of the smaller investment in technical development. These producers can, in many cases, get to prototype very quickly, so that prototype concept testing or even product use testing can easily gain confirmation of customer need fulfillment. Features Features are also a problem. Technical people often come up with features first, based on technologies they have. Some scientist at a glass manufacturing firm might figure out a way to make a boat deck out of finely ground glass left over from some production operation. The thought is pursued for several months only to be knocked out by a shipbuilder’s need for reduced weight. A full protocol statement might have avoided that waste of time. In another case, a scientist did in fact figure out a solution to a certain worm infestation in children, only to be told that this infestation occurs only on scattered Pacific islands and could never constitute a viable market for a pharmaceutical firm. For this reason, firms often ask scientists to keep others informed and to seek input about markets being worked on. The bigger problem with features is that they deprive the firm’s most creative and inventive people of the freedom to use their skills. A large computer firm many years ago was known for having a strong technical research staff. They originated some useful technology. But the firm never achieved much success in reacting to changing needs in the marketplace. Some insiders said it was the result of a system that had a central engineering staff take each situation and spell out the features and characteristics their research staff were to produce. One such spec sheet ran for 13 pages, and the scientist getting it said he felt like a beginning law clerk. He left the firm as soon as he could. An extreme version of a protocol was reported by a pharmaceutical firm in which a new products manager sent a comprehensive advertising layout to his technical counterpart in R&D with an attached note, “Please prepare an item that will back up this ad.” The first reaction was negative, until technical realized they were given carte blanche to do whatever they wanted, so long as the result met the listed claims. Occasionally, a firm knows from long-time market contact what features are associated with what functions (performance) and benefits. They occasionally will put through a work request that calls for “A new pump with electronic valves that give faster reaction to down-line stoppages and thus prevent blowouts.” If the valves are standards, this protocol statement gives feature, function, and benefit. Detailed Specifications On occasions, customers make such decisions and call for products with specific features. This is dangerous. If the customers are qualified and have reason to know better than we do what features will do for them, we are wise to listen. In Chapter 4 Chapter Twelve Product Protocol 301 we talked about getting finished product concepts from lead users (sometimes even a finished prototype). Another case where features may be needed is where a firm is benchmarking competitive products. One strategy is to have the Best of the Best. Take the best features in the market, all products combined, and assemble them in your new product. This sounds great, but it means our product design is being led by competitors, not end users.3 Still other situations where features will appear in protocols are (1) where regulations stipulate a particular feature (e.g., prescription containers), (2) where end users own major items of equipment that impose limitations (e.g., under-dash space limitations for disk players), (3) where established practice in a customer industry is too strong for one supplier to change (e.g., for many years software makers had no choice but to put MS-DOS as a feature requirement), and regrettably (4) where upper managements have personal preferences. In general, as a conclusion to this section on attributes, it is still the best policy to write protocols in terms of benefits, using performance or specific features if that helps explain and doesn’t inhibit too much. For the Hapifork, the customer probably doesn’t care if the fork is made of metal or high-strength plastic, as long as it is durable and can be tossed in the dishwasher after use. Competitive Comparisons and Augmentation Dimensions Benchmarking has been mentioned, but there are many other competitive standards that can be put into a protocol—matching some important policy, the degree of differentiation we have to meet, and many aspects of the marketing plan (e.g., size of sales force, price, distribution availability, and more). Information on competitive comparison can be derived from perceptual maps, and the gaps appearing on the perceptual map can provide guidance on selecting an appropriate competitive position. Just as the product itself was described in attributes above, the augmentation ring of the product can also be cited. Sometimes the product itself may be “metoo,” but still is a legitimate competitive offering as it may offer the customer a new level of service, a better warranty, or better distributor support. Recall that there are three rings in the fully augmented product—ring one (core benefit) is covered in the positioning statement, ring two (the formal product) is covered in the attribute requirements, and ring three (augmentations) is covered here. Other Components of the Product Protocol There are several other components of the protocol that we will handle here very briefly. These are probably best illustrated through example, such as in Figure 12.3. Timing: Most new products today must come out faster, but not all do. Some involve major technical breakthroughs that cannot be put on the clock. The distinction needs to be clear to all. And if there is a date to meet, it should be right here. 3 This is explained by Milton D. Rosenau Jr. in “Avoiding Marketing’s Best-of-the-Best Specification Trap,” Journal of Product Innovation Management, 9(4), December 1992, pp. 300–302. 302 Part Three Concept/Project Evaluation Financials: Typically, the protocol will include price level, discounts, sales volume, sales dollars, market share, profits, net present value, and many of the other financial data introduced in the previous chapter. Production: This one is much like marketing requirements, some focusing on what the function will prepare to do and what that will accomplish—thus, plants to be built, volumes, and quality to be achieved. Regulatory Requirements: These are highly varied, but managements today understand the need to have advanced understanding on them. Corporate Strategy Requirements: Key ideas (such as core competencies) will have already been captured in the product innovation charter. Also, at this time, the assurance of upper management support is important. Potholes: As we have seen before, there are potholes in product innovation, just as they are on that stretch of highway as you drive at night—and they are capable of bringing a new product down. Management that doesn’t take a good look ahead deserves to hit one. We don’t usually drive into known potholes, so listing them here helps. Protocol and the Voice of the Customer Hearing the Voice of the Customer4 Back in Chapter 2, you were introduced to the concept of the voice of the customer (VOC). We return to it here, as it plays such an important role of the development of the product protocol. VOC has been defined as a “complete set of customer wants and needs, expressed in the customer’s own language, organized the way the customer thinks about, uses, and interacts with the product . . . , and prioritized by the customer in terms of both importance and performance—in other words, current satisfaction with existing alternatives.”5 In this definition, “customer’s own language” means exactly that—no scientific jargon. Printer users don’t generally think in terms of edge resolution or number of pixels; rather, they think in terms of how well the letters come out, or how nice the pictures look. Just because the terms don’t sound scientific doesn’t mean the opinions are unimportant! Also, the customers must organize and prioritize their needs in their own way, as they see fit; this is likely to be different from the way the firm sees it. Recall from Chapter 5 that we have several ways to access the voice of the customer: through direct interviewing, for example, or by conducting focus groups. Interviewing customers individually can provide very rich and detailed information, but might be time-consuming and costly. So, how many interviews should be conducted before one is relatively confident the VOC has been captured? Research by Abbie Griffin and John Hauser suggests a reasonable ground rule for 4 Much of this section derives from Gerald M. Katz, op. cit. Gerald M. Katz, “The Voice of the Customer,” in P. Belliveau, A. Griffin, and S. M. Somermeyer, The PDMA Toolbook 2 for New Product Development (New York: John Wiley, 2002), Ch. 7. 5 Chapter Twelve Product Protocol 303 interviews: about 30 individual interviews, each lasting about three-quarters of an hour, produce close to 100 percent of all customer needs; 20 interviews produce about 90 percent of the needs. A VOC process should be audio-recorded with verbatim transcription; this is far more accurate and detailed than having a human note-taker. Respondents should be asked for permission to be recorded; almost everyone complies, and soon forgets the recorder is on.6 If the VOC process was successful, the new product team should have obtained about 70–140 customer needs statements from these interviews. The customer needs statements should then be organized into 15 to 25 groups, called affinity groupings (preferably by the customers themselves as they will generally have a different way of organizing these than will the market researchers). These groupings can then be prioritized by importance to the customer. While this sounds like a long and involved process, cutting corners may compromise the richness of the insights or result in incorrect prioritizations of customer needs.7 The interviewer should be prepared with the right questions. One of the worst ways to elicit the VOC is to ask “What are your needs?” or “What are your requirements?” Customers are all too willing to provide a wish list of “musthaves,” almost certainly taken from existing, available solutions. Innovation expert Guy Kawasaki says, “If you ask customers what they want, they will tell you, ‘better, faster, and cheaper,’ that is, better sameness, not revolutionary change.” Apple co-founder Steve Jobs famously said, “It’s really hard to design products by focus groups. A lot of times, people don’t know what they want until you show it to them. . . . You can’t just ask customers what they want and then try to give that to them. By the time you get it built, they’ll want something new.”8 What should we learn from these comments? Should we forget about using focus groups or interviews to hear the voice of the customer? The real issue here is, what kind of information can you expect to get from focus groups or interviews? To avoid “better sameness,” a better way to proceed is to focus on experiences or desired outcomes, for example, by asking “What are the most difficult tasks you are trying to accomplish with the product?” “What do you like, and what do you dislike?” or “What is the best, and the worst, experience you have ever had with this product?” Consider staying overnight at a hotel. If you were asked to state your needs, what would you say? Probably a clean room, a nice bed, shower, TV, and Internet connection. But if asked what your worst experience was, what would it be? Couldn’t find the plug for your razor? Bumped yourself on the shower head? Towels weren’t clean? Front desk personnel were rude? The interviewer will get many more ideas for product or service improvements this way. 6 Abbie Griffin and John Hauser, “The Voice of the Customer,” Marketing Science, 12(1), Winter 1993, pp. 1–27. 7 Part of this section is adapted from Gerry Katz, “Nine Myths about the Voice of the Customer,” Visions, 35(3), 2011, pp. 34–35. 8 The Guy Kawasaki and Steve Jobs quotes are from Brad Barbera, “Steve Jobs: A Product Developer’s Perspective,” Visions, 36(1), 2012, pp. 10–15. 304 Part Three Concept/Project Evaluation To go back to Steve Jobs’s comment, Apple was indeed asking the right questions of customers and was listening to them. This is what he said about marketing the iPod Touch: Originally we weren’t exactly sure how to market [it]. . . . Was it an iPhone without the phone? Was it a pocket computer? . . . What customers told us was, they started to see it as a game machine. We started to market it that way, and it just took off. And now what we really see is it’s the lowest-cost way to the App Store, and that’s the big draw. So . . . we were focused on [getting] the price down where everyone can afford it.9 So Apple was listening, absolutely: to customer’s desired outcomes and how the product would best provide those outcomes! There is no question that the long list of successful Apple products are primarily technology-push; as Guy Kawasaki said, “The richest vein for tech startups is creating the product that you want to use . . . that’s what [Apple] did.”10 But recall from the PIC that the technology dimension must match with a viable, high-potential market dimension. The iPod Touch example nicely reminds us that the value of the VOC does not come from asking customers to tell you what they want. While perhaps obvious, this point is easy to miss, and lack of success with the VOC may well be due to poor implementation. In obtaining the VOC, it is not enough to get the generalities, such as “I need my smartphone to be flexible,” or “I need my Internet access provider to be consistent.” The obvious follow-ups here are “What do you mean by flexible?” and “What do you mean by consistent?” This ensures that the VOC is clearly heard and not misinterpreted. A good rule of thumb is to keep asking why: “Why did you say that?” “Why do you feel like that?” “Why would it be better that way?” The goal, remember, is not to get technical solutions to the problems. That comes later. Rather, the customer’s wants, needs, likes, dislikes, and so on must be articulated as well as possible at this point. Market researcher and consultant Gerry Katz summarizes the misconceptions about VOC that can lead to its misuse and should be avoided. 1. Many companies treat VOC as qualitative research only, whereas the real value of it comes from organizing and clustering the stated needs and prioritizing them into their relative importance. This is a quantitative process and is often overlooked. 2. Firms often focus on getting the VOC only from major customers, while much important information can be obtained from noncustomers, average customers, and customers who favor the competitor’s product. 3. Managers may believe that customers don’t know what they want. In fact, they are quite good at stating their needs. Not being professional engineers or R&D personnel, they are usually not able to tell what new technology needs to be developed to address those needs. It is up to the firm to match customer need to engineering characteristics. This may be accomplished through Quality Function Deployment, which we explore later in this chapter. 9 Brad Barbera (2012), op. cit. Brad Barbera (2012), op. cit. 10 Chapter Twelve Product Protocol 305 4. Finally, to repeat, it is tempting just to ask customers what they want and need, but that usually provides few new insights. It is better to ask what they like and dislike about current products and what outcomes they would like to see in the future. Protocol and Quality Function Deployment (QFD) QFD and the House of Quality To understand the role of the product protocol in the new products process, think of this process as shown in Figure 12.5. This figure emphasizes the role of the customer, as it shows that we begin with the voice of the customer and end up with a product that satisfies the customer’s needs. Through market research, sales calls, and other forms of customer contact, we are able to identify what the customer desires. The next step is a tricky one, but absolutely essential. We need to convert those customer desires into some kind of blueprint, perhaps an engineering schematic or a detailed service plan, that offers the customer those desired benefits in a format that is useful for the product development team. Once we get to this point, we can take it back to the customer and conduct the appropriate tests to fine-tune the product. It is indeed the product protocol, the subject of this chapter, that helps the firm get “around the bend” of Figure 12.5, because when carefully planned, the protocol allows the firm to translate customer desires into the appropriate product form. This next section describes a technique, originating in Japan but now commonly used worldwide, that allows the VOC to become a driver of all later steps in the new products process. Quality function deployment (QFD) was invented in the Japanese automobile industry years ago as a tool of project control in an industry with incredibly complicated projects. It can lead to reduced design time and costs, and more efficient FIGURE 12.5 The Role of Protocol in the New Products Process End User → Market Contact → New Product Group → R&D Contact → Engineers Unmet Needs Inventory of and Problems Needs Statement of Benefits to How to Deliver the Needs to Be Deliver Requested Benefits Fulfilled by Product PROTOCOL Benefit to Feature Conversion (Specs) Finished Prototype Evaluate Prototype; R&D Delivers Product Confirmed Further Development Prototype Features Delivered; Lab Assesses Performance End User ← Market Contact ← New Product Group ← R&D Contact ← Engineers 306 Part Three Concept/Project Evaluation communication between project team members from functional areas.11 In fact, QFD has been credited with a major contribution to the U.S. automobile industry’s comeback against Japanese competition. We present it here, as it is one way in which many firms have fostered the kind of cross-functional interaction mandated by the product protocol. QFD has also been successfully used earlier in the new products process, very early in the fuzzy front end in concept generation, because it can help the new products team think of novel new concepts that will satisfy customer needs.12 In theory, QFD is designed to ensure that customer needs are focused on all through the new product project: product engineering, parts deployment, process planning, and production. In practice, the first step of QFD has received the most attention and has been useful to the largest number of firms, and that is the so-called house of quality (HOQ). The value of the HOQ to firms is in the way it summarizes multiple product aspects simultaneously and in relationship to one another. Figure 12.6 shows a sample HOQ for the development of a new computer printer. The HOQ requires inputs from marketing and technical personnel and encourages communication and cooperation across these functional areas. Down the lefthand side of the figure appear the customer attributes (CAs), variously called needs, whats, or requirements. This is a critical marketing input into the HOQ. Compatibility, print quality, ease of use, and productivity were identified in this case as the most important CAs for a printer. CAs are identified through market research: focus groups, interviews, and the like. This section of the HOQ corresponds to the part of protocol relating to what the end user will get from the product. It is usually filled with benefits, though occasionally (as above), features or functions (functional benefits) are so mandatory that they are put there. The CAs in this example seem to be primary attributes; in a more complex application there may be secondary or even tertiary attributes under each. For example, ease of use might include “easy to learn how to operate,” “easy to connect,” “easy to replace the paper,” and so on. CAs are also frequently weighted in terms of importance. At the far right of the HOQ are the ratings of the proposed new product and its main competitors on each of the CAs, where 0 5 “poor” and 5 5 “excellent.” This section can be interpreted much like a snake plot of customer perceptions, as we have previously seen in Chapter 6. It identifies the strong points and areas for improvement of our new product. The upper section of the HOQ shows engineering characteristics (ECs): edge sharpness, resolution, and so on. ECs are often technologies, but can also be stated in terms of performance or design parameters. This is where the customer’s needs 11 John R. Hauser and Don Clausing, “The House of Quality,” Harvard Business Review, 66(3), 1988, pp. 63–73; Abbie Griffin and John R. Hauser, “Patterns of Communication among Marketing, Engineering and Manufacturing: A Comparison between Two Product Teams,” Management Science, 38(3), March 1992, pp. 360–373; and Abbie Griffin, “Evaluating QFD’s Use in U.S. Firms as a Process for Developing Products,” Journal of Product Innovation Management, 9(3), September 1992, pp. 171–187. For more information on QFD, try the Web site for the QFD Institute, www.qfdi.org. 12 Gerald M. Katz, “Practitioner Note: A Response to Pullman, et al.’s (2002) Comparison of Quality Function Deployment versus Conjoint Analysis,” Journal of Product Innovation Management, 21(1), 2004, pp. 61–63. Chapter Twelve Product Protocol FIGURE 12.6 QFD and Its House of Quality x x x x Customer rating PostScript compatible Resolution Edge sharpness Duplex printing Hours training required Speed of text Speed of graphics HOWs (technologies) WHATs (benefits) Our new product Product A Product B 0 1 2 3 4 5 Compatibility Print quality Ease of use Yes 400 dpi. .01mm variation Automatic duplex 16 hrs. maximum 10 ppm 5 ppm Productivity (Engin. ests.) Product A Product B Our new product targets Roof Strong positive Positive Negative Strong negative 5 4 3 2 1 0 Matrix x Strong Medium Weak Note: This example shows how the results of a product use test, upon completion, can be compared with what was intended. Under Customer Rating, intended users say the firm did not achieve its objectives. Source: From Milton D. Rosenau Jr and John J. Moran, Managing the Development of New Products, John Wiley & Sons, Inc., 1993, p. 231. Reprinted with permission of John Wiley & Sons, Inc. 307 308 Part Three Concept/Project Evaluation are translated into technical specifications. The project team goes through the central grid of the HOQ, identifying those ECs that will affect one or more CAs either positively or negatively. In this case, “hours of training required” is positively related to both ease of use (strongly) and productivity (less strongly); “Speed of text” is strongly related to productivity. Obviously, this step requires real cooperation between marketing and technical personnel. Objective measures are then set for each EC (usually by engineers), and the team can now begin setting target values for the ECs based on customer need and competitive offerings. For example, speed of text can be objectively measured in pages per minute (ppm); and in this case, 10 ppm was set as the objective. In the automobile example earlier of fast pickup speed, one CA might be “teenage pride among peers.” Related ECs might be a new engine (a technology), the 0–60 time (a performance parameter), or a weight switch putting more load at the point of drive-wheel contact (a design parameter). Practice varies such that we can’t give instruction here, but there are other sources.13 Finally, the top part of the house (the peaked “roof”) shows the trade-offs between ECs that the technical personnel must consider. Each diamond in the roof represents the interaction between a pair of ECs, and the technical staff must identify each significant interaction. The “strong negative” crosshatch (pound sign) at the crossing of “resolution” and “speed of graphics,” for example, indicates that if the printer’s resolution quality is boosted, it is likely to slow down speed of graphics printing. Some of these interactions are positive: A single design change may boost both speed of text printing and of graphics printing.14 As noted above, the HOQ is really only the first part of the full QFD procedure. Figure 12.7 shows what comes next. The HOQ, which translates CAs into ECs, is linked to a parts deployment house, which takes the ECs as inputs and converts them into parts characteristics. Subsequent houses specify the key process operations and production requirements. Nevertheless, experienced QFD practitioners will often find that 80 percent of the value of QFD can be obtained in the first HOQ matrix; consequently, few QFD projects go all the way through the process.15 In a very simple illustrative example, suppose we had decided on the concept of extra-hot, thick, green salsa on the basis of our conjoint analysis in Chapter 7. The CA of extra-hot might be translated to an EC such as hotness on a 10-point scale (a kind of simplified Scoville pepper scale) where habaneros, the hottest peppers, are rated 10. We might aim at no more than 7 or 8 on this scale (as only the most 13 See Hauser and Clausing, “The House of Quality,” for a general introduction. For applications, see John R. Hauser, “Puritan-Bennett, The Renaissance Spirometry System: Listening to the Voice of the Customer,” Sloan Management Review 34, 1993, pp. 61–70; and Milton D. Rosenau and John J. Moran, Managing the Development of New Products (New York: Van Nostrand Reinhold, 1993), pp. 225–237. 14 In a real-life application (iron ore products), increasing a metal’s hardness reduces its malleability (how easily it can be formed into shapes). See Magnus Tottie and Thomas Lager, “QFD: Linking the Customer to the Product Development Process as a Part of the TQM Concept,” Research-Technology Management, July 1995, pp. 257–267. 15 Gerald M. Katz, “Quality Function Deployment and the House of Quality,” in A. Griffin and S. M. Somermeyer, The PDMA Toolbook 3 for New Product Development (New York: John Wiley, 2007), Ch. 7. Chapter Twelve Product Protocol 309 FIGURE 12.7 Later Stages of QFD House of Quality Converted to: Customer attributes Engineering characteristics Parts Deployment Converted to: Engineering characteristics Parts characteristics Process Planning Converted to: Parts characteristics Process operations Production Planning Converted to: Process operations Production requirements Source: Reprinted by permission of Harvard Business Review. Adapted from “The House of Quality,” by John R. Hauser and Don Clausing, May–June 1998. Copyright © 1998 by the Harvard Business School Publishing Corporation; All rights reserved. daring would want salsa to be hotter!). Thickness could be translated into a viscosity measure, and we might aim at a score of between 4 and 6 on a 10-point thickness scale (where 7 and higher would be too thick, and 3 and lower would not be thick enough). The ECs in turn suggest which parts—or, in this case, ingredients— to use: which types of hot chile peppers, how much tomato and garlic, and so on. Process requirements might specify what kind of food processing (chopping, boiling, etc.) will be involved. Related production requirements would be the settings of the food processing equipment that give the desired consistency and appearance. Using the puree setting on the chopper might make the salsa too runny. Outcomes of QFD There are several benefits of applying QFD. For one thing, everything—from product engineering to designing the production process—is driven by customer needs (or, more specifically, by the stated customer attributes). The likelihood that the product about to be developed is one of those better mousetraps that doesn’t have a market is minimized. Furthermore, to get the benefits out of QFD, the various functional areas really do have to work together. This is especially an issue in the development of some industrial products. While consumer-good firms may routinely collect the market data used in the HOQ, industrial product developers often question why they need to do customer needs assessment (or even talk to the folks in marketing)—after all, they say they know the market! QFD has been useful in such firms in promoting dialogue between disparate groups and in encouraging product developers with technical backgrounds to see the advantages of assessing customer needs. In sum, QFD encourages cross-functional dialogue and interaction throughout the technical development process—which is precisely the kind of agreement called for by the product protocol. 310 Part Three Concept/Project Evaluation When QFD was first used extensively in the United States, mixed but generally favorable results were reported. Over 80 percent of teams using QFD reported a long-term strategic benefit and an improvement in cross-functional teaming.16 A more recent survey of QFD use in the United States and Japan finds firms in both countries having success with QFD, but in somewhat different ways. U.S. firms tend to concentrate on the HOQ matrix and collect new primary data from their customers (for example, through focus groups). Japanese firms use more of the downstream matrices and rely more on existing product data (such as complaint information and warranty data). Interestingly, U.S. firms report greater benefits in cross-functional integration and decision making through QFD than do Japanese firms, possibly because the U.S. firms had the most to learn about listening to customer needs!17 QFD has had only mixed results in some applications. It’s expressive, in both cost and employee time, due to the extensive data collection at the VOC phase. It probably is best suited to major projects such as new platform development or major process reengineering.18 Use of QFD by firms tends to be related to better financial performance and greater customer satisfaction. Many firms, however, use it occasionally rather than consistently, and especially for exploratory products (that is, one that will be dropped unless a customer will support it). Besides, the data requirements can be overwhelming. The term matrix hell has been used to describe its application, and highly trained technical personnel may not be able to resolve conflicts that arise. In some cases, the customer firm may not know what it wants, so specifying the “whats” can be difficult. Nevertheless, QFD has been experiencing a resurgence lately, probably because it is viewed as one of the most thorough and objective ways to translate customer needs to engineering specifications.19 Its proponents say it is the best way to uncover customer wants and boost cross-functionality, while its detractors call it overly lengthy and boring, leading some participants to wonder why they are doing it.20 In general, the better the team, the more efficient the QFD; Figure 12.8 provides some guidelines in team selection. The efficiency of QFD can also be improved by doing one or more of the following: • Concentrate on only some of the engineering characteristics: either the apparently most critical ones or some others where improvements might be easy to accomplish. • Organize the engineering characteristics into groups and designate responsibility for these to specific functional areas (i.e., manufacturing, product design, even marketing). 16 Abbie Griffin, “Evaluating Development Processes, QFD as an Example,” Marketing Science Institute, Report No. 91–121, August 1991. 17 John J. Cristiano, Jeffrey K. Liker, and Chelsea C. White III, “Customer-Driven Product Development through Quality Function Deployment in the U.S. and Japan,” Journal of Product Innovation Management, 17(4), July 2000, pp. 286–308. 18 Gerald M. Katz, “Quality Function Deployment and the House of Quality,” in A. Griffin and S. M. Somermeyer, The PDMA Toolbook 3 for New Product Development (New York: John Wiley, 2007), Ch. 7. 19 Gerald M. Katz, “Is QFD Making a Comeback?” Visions, 27(2), April 2003. 20 Gerald M. Katz, “Quality Function Deployment and the House of Quality,” op. cit. Chapter Twelve FIGURE 12.8 Product Protocol 311 Make sure the team is cross-functional. This means design, manufacturing, R&D, marketing, finance, technical support, and anyone else that might have a stake in the success of the product. Criteria for Good Team Selection Appoint an administrator and an advocate for the Voice of the Customer. One person should be well informed on all customer details, and able to explain exactly what customers mean when they express their needs. Team members should have ultimate responsibility to act on the results. If key line managers are on the team, it eliminates the need to convince them of the correctness of the analysis and the need to act. Other criteria: Team members should have knowledge of current practice and also a historical perspective; team members should be respected by their peers; include some top-level executives; include people from a range of levels within the firm; don’t shy away from those who will try some “creative abrasion” to stretch team thinking (but keep the disrupters off the team). Source: From Gerald M. Katz, “Quality Function Development and the House of Quality,” in A. Griffin and S. M. Somermeyer, The PDMA Toolbook 3 for New Product Development, John Wiley & Sons, Inc., 2007, Chapter 7. • Do a cost-benefit analysis on each engineering characteristic to identify which ones provide the greatest benefit relative to associated cost of improvement on that characteristic.21 Some Warnings about the Difficulty of the Protocol Process The protocol process is very complicated. For one thing, it is fraught with politics. The departments are all in natural competition for power and budget. Key individuals are as different as night and day, being scientist, marketer, accountant, and factory manager. The situation itself is fluid and changing, seemingly never nailed down. Management senses the importance of the various projects and puts heavy pressure on them. A big winner on the product frontier can make a career, exonerate a general manager’s other disappointments, award very large bonuses; and of course, a major failure can make a mess of everything close to it. This means that people have their own agenda for incorporating into a protocol (or not incorporating into it). Most want the other people nailed down to specific accomplishment requirements with dollar signs and dates clearly attached, but with no such promises from themselves. Given that a protocol is needed early, just prior to broad-scale work being started, many people are not yet on the scene. They have more pressing, near-term problems, so they delay the process or weaken it by their absence. But, beyond the politics and pressures, we also see a hardening of the requirements in a protocol. People think they were all wise when developing the document and presume the contents are all set in concrete. But it shouldn’t be seen that 21 For good practical discussions of QFD, see Gerald M. Katz, “Is QFD Making a Comeback?” Visions, October 2001; Gerald M. Katz, “After QFD: Now What?” Visions, 25(2), April 2001, pp. 22–24; and Carey C. Curtis and Lynn W. Ellis, “Satisfying Customers while Speeding R&D and Staying Profitable,” Research-Technology Management, September–October 1998, pp. 23–27. For a perspective on how to overcome QFD difficulties, see Rick W. Purcell, “Should the IV House Be a Duplex?” Visions, 27(2), April 2003. 312 Part Three Concept/Project Evaluation FIGURE 12.9 Protocol Accomplishment Requirement 1. 2. 3. 4. 5. 6. Reduce setup time Lower initial cost Easier replacement during manufacturing process Safety in customer’s plan Easier federal approval on finished item Lower cost disposal of trim Company Call Customer Call OK OK OK Doubtful ? Vendor OK Not needed OK Later Not needed Later Date: Explanation: A form such as this, listing all protocol requirements, can serve as a good exercise for the team: How are we going to measure each of the requirements? Must we go outside? When do we do all this? Is a judgment call enough or do we need data? way. It is an aid to management, not a substitute for thinking. All protocols have to change, some of them many times. But the burden of proof is on those who want to change a requirement. Ironically, in some situations the protocol is ignored, so a smart new products team manager will prepare something like the protocol accomplishment form shown in Figure 12.9. It is needed, especially, for product requirements (such as customer benefits), and there should be agreement in advance about who is going to make the call on each. Some can be made by the team, but others must be made by the person the product is being made for. Along the way, bureaucracy sneaks in. One leading computer firm recently made a presentation on product requirements that must have contained at least 25 acronyms; that presentation sounded as if it was right out of government. Last, most of these problems go away if preparation of a protocol is assigned to a multifunctional new products team. Technical doesn’t write one, and neither does marketing. Most assuredly, top management does not write one. Summary This chapter has dealt with a powerful concept—protocol. As an agreement among the functions about the required output or deliverables from a specific new product program, it sets the standards for it. The purpose is to communicate the required outputs as product benefits and other dimensions, integrate the team onto the same frequency, make clear the timing importance, and make it easier to manage the process against specific targets. You saw a simplified version of a typical protocol. At this time we are ready to blow the whistle and charge into the development activity. As seen in the new products process of Chapter 1, action will now take place in marketing and technical in parallel, so we are going to need excellent communication among marketing, R&D, production, design, and other functional areas to get us through the next phase. Applications 1. “Let’s cut right to the quick on this one. I understand the theory of having benefits rather than features, but to me it is just that, theory. I knew one of the top people at that computer company your book talked about—the one where a Chapter Twelve Product Protocol 313 corporate new product engineering group spelled out the specifications of each new product before technical work was funded. I heard the same criticism your author did, so I called this woman and asked her about it. She said the facts were right, but the implication was wrong—corporate staff did indeed spell out most of the features, but only to get the project moving. She said if they just gave their research people the benefits or needs of the customer, those dreamers would never reach a prototype. Every item would be a Taj Mahal. You know, I think she had a point. What do you think?” 2. “I really don’t think you understand what parallel or concurrent new product development is all about. You said you had studied in your course that all of the functions get involved. No, concurrent development means just that—technical development phases—design engineering, etc. They are all doing work very much alike, they work with each other, they can feel how things are going and when they can take a chance and make a premature commitment. Marketing people can’t do that. Even production people (process engineering) have trouble on this score.” 3. “I heard a funny one two weeks ago that might interest you. Seems one of our R&D people went to a new products management seminar and heard about a thing called the protocol. They told him it was the device whereby the overall manager of new products communicated to R&D exactly what was wanted from the technical group. R&D even had to ‘sign on the dotted line,’ swearing that they thought it could be done. He was really steamed—said no one could tell R&D what they should come up with, not in advance, anyway. And R&D is responsible only to top management, not new products managers, so they don’t have to promise anything. He said he considered that concept the most stifling single action imaginable. How would you answer that scientist, or would you?” Case: Fisher & Paykel22 Fisher & Paykel (F&P) is a New Zealand company with an appliances division and a health care division. They are well known in their native country and globally for their radically new appliance designs as well as for several pioneering health care products. Fisher & Paykel Healthcare Corporation (New Zealand) had noticed an opportunity for a new medical product for home use, namely, a continuous positive airway pressure (CPAP) device to treat obstructive sleep apnea. The development of the new CPAP machine illustrates an underlying concern in designing medical devices for home use in general. These devices are often very complicated; a medical professional is skilled in the use of such a device in the hospital, while a patient is likely to be confused, make errors, and create safety problems that could lead to injury or death. Hence, user interface is a critical design consideration, and health care professionals must be included in the design and testing stages. User ability and lack of medical training are only one aspect to consider in the user interface. Lighting, humidity, noise, and other ambient factors need to be taken into account. For example, the CPAP machine will often be used in the dark, with the user lying on a bed; controls will need to be easy to find 22 This case is adapted from Aruna Shekar and Andrew Salmon, “From the Hospital to the Home: Innovations in Developing Medical Devices for Residential Use,” Visions, 35(3), 2011, pp. 40–43. 314 Part Three Concept/Project Evaluation and operate in the dark and easy to reach by a reclining person. Furthermore, a medical professional may have no difficulty navigating dozens of switches and displays; operation will need to be simplified for the nonprofessional user. Needless to say, product testing is required to be sure the product does not have any overlooked flaws, such as a hard-to-reach on-off switch. CPAP machines look like most medical appliances, that is, they are big and bulky. While they might be fine for hospital use, this design would be unsuitable for a CPAP machine to be used in the home environment. For one thing, the CPAP machine is located by the user’s bed, and its base would take up the whole bedside table. An equally important consideration is that users tend not to like home medical devices that look “medical” as this is a constant reminder of their illness. Based on these insights, F&P set out to develop a home use CPAP machine that lets the user feel that things are “back to normal” at home, and that in fact would be a pleasure to have in the home. Early in the process, a list of product specifications based on customer input was drawn up. The list included the following: • • • • • • • • Ornamental and decorative aesthetic Usability—easy to operate the device Appropriate for bedroom use (in the dark, user lying down) Small footprint, easy to set on a bedside table Colors appropriate for a bedroom Use materials that are consistent with the aesthetics Design tube for better delivery of humidity to user and less condensation Built-in alarm clock and MP3 player; can wake up to music The first characteristic tackled by F&P was unit size (existing products were too large for the bedside table) and design (they are hard to operate, because the on-off switch is located on top and hard for a person lying in bed to see or reach). Users also did not like the tubes sticking out of the machine, or the “medical” appearance. F&P spent several months in the front end of the new products process (three to six times the normal length in this industry), including substantial time with users in sleep labs and hospitals, to get these and other insights regarding product usability and desirability. F&P R&D personnel then built nonworking prototypes (cardboard models) to get the form that users liked and that fit neatly on the bedside table. Further prototype testing led F&P to select large dials for controls: these would be easy for the reclining user to locate and to operate, even in the dark. Designers created simulated environments and role-played sleeping users to ensure that the controls were as accessible as possible. Based on this preliminary research, computer-aided design (CAD) was used to generate full-size “block models” of CPAP machines to see how they would fit into the user’s environment. During the design phase, three major shape redesigns were made, as well as 80 smaller detailing iterations. Then, CAD was used to finalize features using rapid prototyping. The final designs were then used in user surveys conducted in New Zealand, Australia, and the United States, and the results used to make final changes to shape, usability, and color. These were 30-minute Chapter Twelve Product Protocol 315 in-depth interviews using real, nonworking, physical prototypes to generate discussion of preferences for colors and shapes as well as display placements. Since the sleep environment is such an important consideration in the use of this product, the surveys were conducted in sleep labs to simulate the home bedroom experience as much as possible. Once design was close to finalized, the product went into development. Mechanical, electrical, and software engineering was done right away so as to allow real-life in-house testing as soon as possible. Once the design was firm, small batch production began and final testing of product requirements began. Because manufacturing and design functions were located side by side, it was easy and quick to make design iterations, communications were simple, and the functional groups shared joint project ownership. F&P worked with their suppliers to introduce new technologies, and the intellectual property attorneys were constantly in the loop in order to protect ideas fresh off the drawing board. The result of this product development was the ICON CPAP product line, considered to be revolutionary in terms of its design. It resembles a clock-radio and actually functions as one. Despite its unobtrusive appearance, it contains an internal humidifier, resulting in less noise and higher efficiency. The number of buttons was kept to a minimum, and their large size simplified use and maintenance. The success of the ICON line can be measured in terms of higher patient compliance as well as fewer user errors. Comment on how the list of product characteristics was developed and used. How is it similar to, or different from, the simplified product protocol, or list of “musts and wants,” as discussed in the chapter? What other principles of good product development are apparent in the development of the ICON CPAP line? Case: DuPont23 DuPont is a chemical company with a long and successful history of product development. Founded in 1802, it is one of the world’s leading chemical companies and produces products for the food, construction, communications, and transportation industries, among others. DuPont operates in over 70 countries and employs over 58,000 people. DuPont’s success with new products over the years is well documented: some of the most familiar include nylon, Teflon, Kevlar, and Lycra. In 2007, however, DuPont senior management felt that something was amiss in the firm’s new products process. After lengthy study, the company identified several problems, which are really not that uncommon. First, there was little deep customer insight, in particular at the customer segment level. DuPont was wrestling with a familiar segmentation issue: develop a product for a single customer and there is no guarantee of sales in the wider market, but a one-size-fits-all product will end up pleasing no one. Insights at the target segment level were lacking. Second, on paper the company had a voice-of-the-customer process, but it seemed to have devolved into “chat with the customer” sessions that were not generating breakthrough ideas. Finally, DuPont representatives were meeting with direct 23 This case is adapted from Dan Edgar and Dan Adams, “How DuPont Uses New Product Blueprinting at the Front End—And Implemented a New e-Learning Model to Teach It,” Visions, 34(3), 2010, pp. 12–17. 316 Part Three Concept/Project Evaluation customers, but not with their customer’s customers. While it is the direct customer that actually pays the bills, it is the customer’s customer who has the important needs that must be met by new products. As a result, the value propositions developed by DuPont were often based on wishful thinking, not hard customer-needs data, and therefore did not inspire customers. Recognizing the need for a novel, innovative approach to the front end of the new products process, DuPont turned to the Institute for the Study of Business Markets (ISBM), where they met with the consulting firm AIM and discovered New Product Blueprinting, an approach especially suitable for business-to-business (B2B) product development. There are several characteristics of B2B product development that distinguish it from consumer product development, and which drive New Product Blueprinting. First, B2B customers are knowledgeable, having the education and experience to be able to provide new product guidance to their suppliers (like DuPont). Second, B2B customers want to help their suppliers because it can help them as well, in terms of lower costs or higher performance for their own products. Third, B2B customers make rational, stable decisions, and fourth, there are relatively fewer B2B customers and their decision making can be influenced by supplier actions. New Product Blueprinting emphasizes “going deep” with customers. The mode of contact is by interview. Questionnaires are not used, since it is felt that “you will seldom get more than you ask for.” It is deemed better instead to run an ideageneration interview, taking down “digital sticky notes” and displaying them on a big screen to fully engage customers and obtain their comments and feedback. In this synchronous process, supplier and customer interact, the customer can comment on and correct the supplier, and the supplier can probe the customer’s comments deeply. In New Product Blueprinting, the focus is on finding the customer’s desired outcomes, which allows the supplier to search for solutions. New Product Blueprinting also uses some quantitative metrics, in particular the market satisfaction gap (MSG). MSG is calculated as Importance x (10–Satisfaction), where Importance and Satisfaction (with current products) are measured on 1–10 scales. An MSG is greater than 30 signals a significant opportunity. Initial trials of New Product Blueprinting at DuPont were successful. Three projects were initiated around this time. One was a product for the displays market in Asia; interaction with customers determined that there was little desire for any new materials and that the market segment held little value. The second was a product for the global electronics market for which much internal development was already completed. After properly specifying the desired customer outcomes, it was determined that at least five more years of development would have been required. The third was a global solar energy product for which strong customer needs were identified that met well with DuPont’s capabilities. The result of New Product Blueprinting: the first two projects were terminated (thus avoiding two very costly failures), and the third was approved and became financially successful. How is New Product Blueprinting different from traditional voice of the customer analysis? How does it complement VOC? Do you think New Product Blueprinting might work for a consumer packaged goods company such as Campbell Soup? What about for a service such as a hospital or bank? Why or why not? What Chapter Twelve Product Protocol 317 might be drawbacks of New Product Blueprinting for a high-tech manufacturer like DuPont or for other kinds of firms? Case: Logitech (B)24 Think back to the Logitech case (A) at the end of Chapter 10. The clock has rolled forward a few months, the full screen has now been conducted, and the inexpensive digital video surveillance product concept has passed easily. Evan and Andrew are very excited about the prospects so far. They have even come up with a name for the new product: LukWerks, pronounced “look works.” Here is what is known so far on the technical side. The digital technology is definitely there, and it works. It is possible to make a video camera that can be used for motion detection; in fact, it is possible to put the software required for motion detection, digitization, and data compression right into the camera itself. The data can be transported to any regular PC via Ethernet, thus eliminating the need for a dedicated computer for surveillance purposes as well as the need to drill holes in the walls and feed new wires through them. The user can download LukWerks software on his or her regular computer, and video images would be stored there. The camera described above could be built relatively easily and could safely be used indoors. A few additional kinks would have to be ironed out to make an outdoor version (as noted before, protection against the elements and infrared capability for nighttime use become concerns for the external camera). These are not insurmountable, but would require additional technology and possibly some additional development time. Finally, there is the issue of price. At the current time, Evan and Andrew think the product they have in mind, if commercialized, could be brought to market at a retail price of about $300 for one camera plus software. They are exploring the idea of selling add-on cameras for the user who wants multiple cameras, at perhaps about $250 each. First, think about whether the protocol idea would fit the situation of LukWerks. Then, write five lines of benefits that customers would probably stress if they were interviewed. Decide how you would actually measure whether the benefits were being achieved when LukWerks was used. Second, refer to the list of contents in a protocol and see if there are any other points that could be added to the benefits you just wrote. There won’t be many in a simple situation like this, but there will probably be some. Look especially at the marketing requirements. Finally, though you are (probably) not in the digital camera video surveillance business, there is enough technical information given in this case for you to try a very basic house of quality. Take each of the customer benefits and try to convert it into an engineering characteristic. As a simple example: The customer wants ease of installation. One possible engineering characteristic would be to eliminate any need for special skill or tools to install the camera. If it could be placed on a table or hung on a wall like a picture, it would require no skill beyond that needed to hang a picture. 24 See Logitech (A) for list of references. 318 Part One Overview and Opportunity Identification/Selection FIGURE IV.1 Development From Figure III.1 (Concept/Project Evaluation) Gather resources Create appropriate environment Assemble the team Culture, style, rewards Outline allied resource network Designate project leader Sketch informal roles Technical development Create necessary information systems Install market scan system Staged updates of financial analysis Project review system Marketing development Basic research and equivalents Product design and evaluation Release early prototypes Protocol check Conduct alpha tests Gradual development of (and continuing revisions in) the marketing plan Repeat end-user concept screen Develop revised prototypes Conduct alpha/beta tests Prepare product specs Product design and product cost forecasts Produce pilot product Target marketing and positioning Major marketing mix: Promotion Pricing Channel Product line changes Tentative sales forecasts Early expense budgets Conduct alpha/beta tests Plan product augmentations Packaging, customer service, distribution logistics, technical service, warranty requirements Update cost forecasts Update sales forecasts Prepare comprehensive financial analysis To Figure V.1 (Launch) P A R T F O U R Development Somewhere during the preceding process of creation and early evaluation, a decision was made to develop the concept being considered. The decision may have come quickly (a key customer wanted the item and was ready to help develop it) or slowly, after concept testing and extensive review of capital and operational expenditures required. A product protocol was written, and an early financial plan released funds for the development. The question now is fulfillment of that protocol. There may be extensive technical search (for example, for a new pharmaceutical) or none (for a new service). The key problem may be in industrial design or in the very technical characteristics of the newest computer chip. Fulfillment may consist of nothing more than confirming a recipe that was used to produce new cookies for the concept testing. In cases such as the development of a new blade by Gillette, years of technical development may be necessary. This is a point of high creativity, and there is usually a strong art form, even when dealing in scientific areas. Progress in the development phase has the attention of managers in all functions, not just the R&D people. Once the product protocol is written, the rest of the team does not sit back and wait for the engineers to produce a finished prototype. It is better to think of development as a phase that includes the creation of everything needed to market the product, including funding, distribution, promotion, and technical service. Look at Figure IV.1. The technical work (including design, engineering, and manufacturing) is displayed down the left side of the stream. Testing, marketing, and legal, among others, are displayed on the right. Both continue through the launch. Several things about that figure may surprise you. • First, note that what we commonly think is the technical creation task is just one box on a page with 15 boxes in the technical (left-hand) stream. 320 Part Four One Overview Development and Opportunity Identification/Selection That one box, in practice, is broken down into literally thousands of other boxes. Many firms use a project control system called a Program Evaluation Review Technique (PERT) chart, or network diagram, originally developed for the first nuclear submarine, the Polaris, in the late 1950s. A network diagram uses boxes (“nodes”) and connecting lines to indicate the flows of tasks in a project and how they all interrelate. In the automobile industry, a network diagram for just one assembly (e.g., the dashboard) is so complicated that it cannot even be printed out on paper. • Note, too, the large box at the top of the diagram. Getting ready to do technical development sometimes takes months—finding the people, acquiring the rights to certain materials, creating a particular culture, training the team, and, so important today, creating the information system to support the complex of activities. • Typically there is not just one prototype. Sometimes there will be dozens or even hundreds, depending on how lucky the team is. Granted, a new Frisbee with an edge shaped to be easily grabbed by a dog in its teeth may be real progress for competitors in that sport, but hardly an afternoon’s work for the designer. Edison was said to have tried hundreds of materials for the filament in the first electric light bulb before settling on the right one. • Developers must stop frequently to have their work checked—note terms like evaluation, check, screen, test, and clearance. Generally this is good, because to advance a design with a flaw is wasteful, yet to stop at every possible turn grinds things to a halt, including morale. • The technical side is a rolling evolution. Even when an early prototype looks good, it must evolve into a tested prototype, then into a process, then into a pilot product, then into a production scale-up product, then into a marketed product. This follows the same route as the life cycle of a concept that was diagrammed and explained in Chapter 2 on process. We don’t really develop a thing so much as we evolve one. There are fewer “eureka moments” than people think. It’s hard work, step by step. • Note, too, how items on the right-hand stream associate with items on the left-hand. Thus, producing a prototype may start design on a package; producing a scale-up product stimulates start on a technical customer service activity; producing a marketable product means a distribution network must be in place. But rather than try to discuss both streams simultaneously, we cover marketing’s role in the technical work here in Part IV and marketing’s role in the other stream in Part V. Actually, we have been working on the marketing stream from Chapter 3 on—for example, target market is usually known at PIC time and new product positioning statements are used in concept testing. Chapter Part FourOneDevelopment The Menu 321 Thus Chapter 13 talks about the players involved, the essence of design, and productivity in the development process. Chapter 14 covers the creation and management of today’s cross-functional teams, and Chapter 15 tells how the team finds whether the latest prototype is indeed, ready to launch—a subject we will take up in Part V of the book. Before we get too far, let’s be sure we know just what marketing’s role is in the work that takes place on the technical side. There are nine important dimensions. 1. To make absolutely clear to everyone what the protocol calls for. What is the end point? How can technical groups know when they are finished? 2. To make sure that this protocol task is technically feasible and doable within the time and dollars imposed by the development budget. That is, do all technical people agree? 3. To provide an open window for industrial and systems designers to all influential forces in the marketplace. Marketing should not be a gatekeeper, but rather an enthusiastic tour guide. It is truly in their best interests, and in the firm’s as well, that all development effort (technical and marketing) be based on market knowledge. 4. To provide a continuous interim of opportunity to pretest various versions of the new product. This means to cooperate in early in-house testing and in later customer use testing. 5. To be available to technical people at all reasonable times. Some marketing people seem to forget that technical work is going on. A common joke in the labs is the scientist who left for lunch with the request to an aide: “If my product manager calls, get their name.” 6. To stay informed about technical progress via team meetings, lab visits, social contacts, and so forth. This is not spying. It is seeking an opportunity to pass along some market information technical people didn’t know about. Well-led teams today soften this problem, but marketers have to learn how to be good team members. 7. To involve technical people in the decision making on the marketing side of the development stream—especially any changes in the givens at the start of development, target market for example. Again, teams help, but just as marketers can get distracted, so can technical people. We have to show them why we need their input on matters they may not feel are as important as the technical ones they are busy on. 8. To stay continuously alert to the project’s progress and to be creative in finding ways to help. For example, in Chapter 14 you will see the benefits of cross-functional teams, one of which is to speed up a new product’s development. Saving a day in marketing may be as good as saving a day in technical. 322 Part Four One Overview Development and Opportunity Identification/Selection 9. To flag the various ways that work in nonmarketing departments impacts marketing plans directly. This action, often called internal marketing, involves technical departments (for example, technical information for sales brochures), manufacturing (for example, cost reductions and standby production capability), packaging (for example, promotion claims made on front panel), and human resources (for example, selection of new personnel needed in the launch effort). It is the purpose of the material in Part IV to help you perform those roles, but be aware—the technical side of the development stream is immensely more complicated than most outsiders realize. Don’t take the roles lightly. C H A P T E R T H I R T E E N Design Setting Part IV of this book explores all aspects of the development phase, which includes product design, product architecture and prototype development, and product use testing, as well as organizational and team management issues. Here in Chapter 13, we examine just what this development phase means to different companies, and we introduce design and its use as a strategic resource. We also examine the role of the product designer and the interface between design and other functions involved in the new products process. As consumers, we have all been frustrated by poorly designed products and wonder how they ever got to market: • Too-bulky or underpowered vacuum cleaners. • Cereal boxes with protective packaging that rips when first opened and thus no longer protects. • Oddly shaped spatulas that are useless for flipping pancakes. • A coffee vending machine that does not indicate that you have to provide the cup: you learn this the first time you get hot coffee spilled on your pants. • A combination CD-tape player, with the tape controls located near the CD drive and the CD controls near the tape drive. Yet we recognize and appreciate outstanding designs—a new car, revolutionary office furniture, or even a universal screwdriver that really works—and reward the product manufacturers. The design and appearance of the Apple iPod certainly adds to its appeal; likewise James Dyson’s vacuum cleaner. In a day and age of “don’t sweat the small things,” it may be those very small things that determine brand preferences and that the manufacturers should focus on!1 1 Laurence P. Feldman, “But Have You Tried the Product?” Visions, October 1999. The examples are from this article, as well as Laurence P. Feldman, “Is Your Product ‘Utility Challenged’?” Visions, April 2000, and from the Bad Designs Web site, www.baddesigns.com. This site features dozens of poorly or oddly designed products and includes ideas on how the design could have been easily improved. 324 Part Four Development What Is Design? One writer defines design as “the synthesis of technology and human needs into manufacturable products.”2 In practice, however, design as a term has many uses. To the car companies, it can mean the styling department. To a container company it means their customer’s packaging people. To a manufacturing department it most likely means the engineers who set final product specifications. Excellence in design also benefits the bottom line. Firms that are judged to be higher in design effectiveness outperform other firms in return on sales and assets, net incomes, and cash flow, as well as higher stock market returns.3 Consider, for example, the role of design at Apple. Over the years, Apple has received much praise for the sleek, modernistic designs of its iPads, iPhones, and other devices. The clean, simple lines of these products can be directly traced to the 1960s-era record players and radios designed by famed German designer Dieter Rams. In any case, design should not be considered an afterthought where industrial designers are asked to pretty up a product that is about ready to be manufactured. This narrow view of design causes managers to miss the potential that design has to occasionally innovate within the organization. Design-Driven Innovation4 Some writers have suggested that the traditional, dual-drive product innovation strategy (technology-driven or market-driven) neglects the potentially powerful role of design. In both technology-driven and market-driven innovation, design plays a secondary role. Technology-driven innovation starts with the technology; the role of design is to modify the product so that it can accommodate the performance characteristics. Market-driven innovation starts with the customer; here, design modifies the product so that it meets customer expectations. Design academic Roberto Verganti suggests considering a third way: design-driven innovation, in which it is design itself that takes on the leadership role. In his words, Design introduces a bold new way of competing. Design-driven innovations do not come from the market; they create new markets. They don’t push new technologies; they push new meanings. Customers had not asked for these new meanings, but once they had experienced them, it was love at first sight.5 2 See Michael Evamy, “Call Yourself a Designer?” Design, March 1994, pp. 14–16. This article was part of a series in this publication, all on the matter of design definition. Useful also is Karl T. Ulrich and Steven D. Eppinger, Product Design and Development (New York: McGraw-Hill, 1995). 3 Julie H. Hertenstein, Marjorie B. Platt, and Robert W. Veryzer, “The Impact of Industrial Design Effectiveness on Corporate Financial Performance,” Journal of Product Innovation Management, 22(1), January 2005, pp. 3–21. 4 This section is adapted from C. Anthony Di Benedetto, “Product Design: Research Trends and an Agenda for the Future,” Journal of Global Fashion Marketing, 3(3), 2012, pp. 99–107. 5 Roberto Verganti, “Radical Design and Technology Epiphanies: A New Focus for Research on Design Management,” Journal of Product Innovation Management, 28(3), 2011, pp. 384–388. Chapter Thirteen Design 325 Verganti offers a designer teapot, designed by architect Michael Graves and sold by Italian manufacturer Alessi, as an example of design-driven innovation. Most teapots are utilitarian: they boil water, quite effectively, for maybe five minutes a day, and take up space in the kitchen for the rest of the time. The Graves design was felt to be “delightful,” to the extent that it actually made the breakfast experience more pleasurable. It was attractive, its coneshaped design with the wide bottom does not rock unsteadily on the stovetop, the handle with the grip set far back eliminates burning one’s hands when pouring hot water, and a little bird on the spout whistles when the water is ready. Rather than taking up space, the teapot becomes part of the décor and is something most people would be proud to own and show off. The fact that virtually the same product was mass-produced and sold at a much lower price point through Target stores suggests the universal appeal of this high-design product. In fact, this example clearly shows that product functionality is just as important to excellent design as product appearance or aesthetics. As noted by Ken Munsch, New Product Business Development director at Herman Miller, “Sharper Image specialized in sleek, modern style and filed for bankruptcy. Beautiful is not enough. The product must be useful. Design includes the whole human interface.”6 The Role of Design in the New Products Process7 Design’s potential role in the new products process is sometimes underestimated. This may be because of a lack of understanding or appreciation of designers, design management, and the design function on the part of managers from other functional areas. Designers undergo rigorous training to learn how to design products that function well mechanically, that are durable, that are easy and safe to use, that can be made from easily available materials, and that look appealing. Clearly, many of these requirements will be in conflict, and it is up to the skillful designer to achieve all of them simultaneously. Contributions of Design to New Product Goals As proof of the importance of design, consider several ways in which design excellence can help firms achieve a broad spectrum of new product goals, as shown in Figure 13.1. 6 Quoted in Brad Barbera, “Steve Jobs: A Product Developer’s Perspective,” Visions, 36(1), 2012, pp. 10–15. 7 Much of this section is drawn from Jeneanne Marshall, “Design as a Strategic Resource: A Business Perspective,” Design Leadership Program, Corporate Design Foundation, 1991; and Eric M. Olson, Rachel Cooper, and Stanley F. Slater, “Design Strategy and Competitive Advantage,” Business Horizons, 41(2), March–April 1998, pp. 55–61. For a good view of “hot topics” among design practitioners, read the periodic newsletter @Issue. Current and back issues are available online on the Web site for the Corporate Design Foundation, www.cdf.org. 326 Part Four Development FIGURE 13.1 Contributions of Design to the New Products Process Design for Speed to Market Design for Ease of Manufacture Design for Differentiation Design to Meet Customer Needs Design to Build or Support Corporate Identity Design for the Environment Design for Speed to Market Ingersoll-Rand developed its Cyclone Grinder (an air-grinder power tool) in record time, thanks to an efficient cross-functional team and excellence in design. The team (composed of marketing, manufacturing, and engineering personnel) worked closely with Group Four Design to identify customer needs. Users of traditional grinders often complained that they were difficult to hold, and that their hands would freeze (the unit became cold during use). The new grinder was ergonomically shaped (better shaped for the human body meaning, in this case, easier to hold), lighter, and made of a new composite material that was both more durable and more comfortable to hold (since it conducted less thermal energy and thus did not get cold). Furthermore, the one-piece housing design was a cost improvement over the previous version, which required assembly of seven different components. Design for Ease of Manufacture A classic example here concerns IBM’s development of its Proprinter dot-matrix printer in the mid-1980s. At the time, the Japanese owned the worldwide market for low-end printers. It was felt, however, that the competition was vulnerable: Their printers were not well designed, and in particular had hundreds of parts including dozens of rivets and fasteners. IBM set a performance target of 200 nearletter-quality characters per second (not the current standard, but the expected standard four years into the future) and had a motto of “no fasteners”: Everything had to snap together easily. Furthermore, the development time had to be compressed from the standard four years to two-and-a-half years. All of the above was achieved: The original Proprinter had only 61 parts and could be assembled in three minutes. Similarly, Swatch watches are designed for ease of manufacture, Chapter Thirteen Design 327 having about a third of the moving parts of a traditional Swiss watch, a plastic casing without a removable back, a plastic strap incorporated into the casing, and many other design features. Swatch watches retail at a small fraction of the price of traditional Swiss watches. Design for Differentiation Haworth Inc., the office furniture designer, employs an Ideation Group, responsible for exploring and assessing customer acceptance of speculative products (highrisk products without a clear-cut market). Haworth believes that “nonstandard” product development is needed for speculative products. Few of the prototypes developed by Ideation may make it to the marketplace, and those that do (like the Crossings furniture line) may end up looking quite different. Good ideas from the Ideation Group can make their way into existing lines or other future products, and more importantly, Haworth has successfully differentiated its product offerings as being more original in design. Incidentally, excellence in design seems to be important in the office furniture industry: Steelcase Inc. is a majority owner of IDEO, the design firm we have met more than once in earlier chapters.8 Design to Meet Customer Needs Deep understanding of customer needs is required in order for the firm to translate a high-potential technology into a product that provides meaningful benefits to the customer. Collaboration with end users (seen in Chapter 4) and capturing the voice of the customer (Chapter 12) are important ways to get this depth of understanding, now sometimes referred to as user-oriented design.9 The voice of the customer was extensively used in the design of the Infiniti QX4 sport utility vehicle. In fact, marketing director Steve Kight said at the time that “the QX4 was designed expressly for [our customers].” Interviews and surveys of Infiniti drivers in Westchester County, New York, revealed their preferences in an SUV: handles like a car, easy to get into, priced below $40,000. Infiniti drivers and nondrivers within the target market (35–64 years old, over $125,000 household income, willing to purchase a luxury car) were presented with five different designs. The best of these was molded into clay and fiberglass models with the additional input of dealers. Finally, the SUV was supported with a strong promotional campaign, advertising heavily in magazines such as Smart Money. As a result, sales far exceeded expectations.10 Crown Equipment Corporation, a manufacturer of forklift trucks, developed its RC (Rider Counterbalance) lift truck and launched it in 2008. An age-old problem expressed by forklift truck drivers is their inability to see clearly in front, especially 8 Janis R. Evink and Henry H. Beam, “Just What Is an Ideation Group?” Business Horizons, January–February 1999, pp. 7–77; Bruce Nussbaum, “The Power of Design,” www.businessweek.com, May 17, 2004. 9 Robert W. Veryzer and Brigitte Borja de Mozota, “The Impact of User-Oriented Design on New Product Development: An Examination of Fundamental Relationships,” Journal of Product Innovation Management, 22(2), March 2005, pp. 128–143. 10 Constance Gustke, “Built to Last,” Sales and Marketing Management, August 1997, pp. 78–83. 328 Part Four Development if they have pallets raised on the forks. In some cases, a second person would be required to guide the driver, whose sight line was obstructed by the load carried at the front of the truck. Using an ingenious counterbalance system, the RC’s forks are located to the side so as to remove the driver’s obstruction. Additionally, the RC had extra design elements that addressed other common user complaints and appealed to the driver: a much larger than average operator compartment, a desktop area allowing the driver to keep papers and tools nearby, a newly designed shock absorption system that smoothed the ride, and a stylish and ergonomic appearance. The RC significantly grew Crown Equipment’s market share and also won several design awards.11 Universal design is the term sometimes used to mean the design of products to be usable by anyone regardless of age or ability. Principles of universal design can be used to develop products for new markets based on unmet customer needs. The designer considers the abilities of real people in real-world settings when applying universal design principles. For example, some people are visually impaired, while others have temporary vision problems due to eye fatigue, recovery from surgery, or even poor lighting. Phones with extra-large buttons address permanent or temporary vision problems and can be used by anyone. Closedcaptioned television, automatic garage-door openers, and automatically opening doors to grocery stores also exemplify universal design. Figure 13.2 illustrates the principles of universal design. Design to Build or Support Corporate Identity Many firms have established visual equity across the products they sell: a recognizable look or feel that they use consistently. Product design can thus help build or support public perception of the firm and, ultimately, its corporate identity. Apple computers and other devices have always been designed to look user-friendly. Rolex watches all have a classic, high-prestige appearance, and Braun appliances have lines and colors that convey simplicity and quality.12 Nokia phones share common design elements that make them unique, yet at the same time familiar. The company calls these commonalities “Nokia DNA.” Radically designed new BMW models, such as the Z4, still share familiar design attributes with classic BMWs of years ago, such as the distinctive grille.13 Design for the Environment Design for disassembly is the technique by which products can be taken apart after use for separate recycling of metal, glass, and plastic parts. Among other carmakers, BMW has designed disassembly and recycling into its cars. Used plastic 11 Bruce Nussbaum, “The Best Global Design of 2008,” Business Week, July 17, 2008; also see the firm’s Web page, www.crown.com. 12 Karl T. Ulrich and Steven D. Eppinger, Product Design and Development, 2nd ed. (Burr Ridge, IL: Irwin/McGraw-Hill, 2000), p. 219. 13 Anonymous, “Online Extra: A Chat with Nokia’s Alastair Curtis,” www.businessweek.com, July 17, 2006. Chapter Thirteen Design 329 FIGURE 13.2 Principles of Universal Design Principle Examples Equitable Use: The design is useful to people with varied abilities. Pay phones in public places with adjustable volume levels Powered doors to grocery stores are convenient to disabled shoppers and also people pushing carts, strollers, etc. Flexibility in Use: The design accommodates a variety of preferences and abilities. Phones with large buttons Scissors or knives that work left- or right-handed Simple and Intuitive to Use: The design is easy for anyone to understand and use. Color-coded labels on cough medicine Ikea furniture building instructions use illustrations and minimal text to avoid language barriers Newer DVD and DVR players are easier and more intuitive to program with on-screen commands Perceptible Information: The design communicates the required information effectively to the user. Plugs and jacks connecting DVD players and other electronic gadgets to televisions are color-coded Honeywell thermostats show numerical settings and also use audible click-stops when the dial is turned Tolerance for Error: The design minimizes adverse consequences of inappropriate use. Irons or coffeemakers that shut off if not used for five minutes Lawnmower handle that requires the user to squeeze a lever against the handle to keep the lawnmower running Low Physical Effort: The design can be used efficiently by anyone with minimal fatigue. Rollers and handles on luggage Angled computer keyboard easier for operator to use Size and Space for Approach and Use: Regardless of the user’s size or mobility, the product is easy to reach, manipulate, and use. Whirlpool side-by-side refrigerator-freezers with full-length handles Copco chopping knife’s handle is designed to be comfortably held in hands of any size Wide car door opening makes it easier for someone with a walker or wheelchair to get in or out Source: From James L. Mueller and Molly Follette Story, “Universal Design: Principles for Driving Growth into New Markets,” in P. Belliveau, A. Griffin, and S. Sodermeyer (Eds.), The PDMA Toolbook for New Product Development, John Wiley & Sons, Inc., 2002, pp. 297–326. Reprinted with permission of John Wiley & Sons, Inc. parts are sorted, recycled, and made into new parts. Other components are either recycled or rebuilt, while unusable parts are incinerated to create energy.14 In fact, green design is now a driving force within many firms. The carmaker Subaru provides an example. Thomas Easterday, senior vice president, Subaru of Indiana, says that Subaru has “embraced the concepts of reduce, reuse, and recycle.” He claims that Subaru has achieved zero landfill status and has attained a recycling rate of 99.8 percent (the remainder is hazardous waste that must be incinerated due to EPA regulations). Subaru works with suppliers so that they use recyclable packaging and with local companies responsible for collecting and recycling materials; the carmaker also finds markets for recycled materials. More recycling 14 Jacquelyn A. Ottman, Green Marketing: Challenges and Opportunities for the New Marketing Age (Lincolnwood, IL: NTC Business Books, 1993), p. 119. 330 Part Four Development FIGURE 13.3 Range of Leading Design Applications Purpose of Design Item Being Designed Aesthetics Ergonomics Function Manufacturability Servicing Disassembly Goods Services Architecture Graphic arts Offices Packages Comment: Design is a big term, covering many areas of human activity, especially new products. The new products field contributes to two classes of items and to all six classes of purpose. Some people hold that even the other four classes of items are really products to the organizations producing them. leads to less waste, and cost savings, at Subaru.15 Apple also makes several green claims for the iPad on its Web site, noting that the display is mercury-free, there is no PVC plastic used, and the aluminum-and-glass enclosure is recyclable.16 Figure 13.3 shows a variety of design dimensions, using only the two criteria of Purpose of Design and Item Being Designed. Design is not just a field in which artists draw pictures of new microwaves. It blends form and function, quality and style, art and engineering. In short, a good design is aesthetically pleasing, easy to make correctly, reliable, easy to use, economical to operate and service, and in line with recycling standards. Ergonomics are also an important consideration; this can be defined as studying human characteristics in order to develop appropriate designs.17 Many of the poorly designed products mentioned at the start of this chapter might have been improved with better attention to ergonomics. An excellent design can play a big role in determining how well a new product will meet the needs of customers, as well as retailers and other stakeholders, and therefore is an important determinant of success. Consider one innovatively designed product: the Cross Action toothbrush by Gillette’s Oral-B division. Researchers videotaped people using toothbrushes to determine actual brushing patterns, then built a robot arm to simulate brushing action. High-speed video cameras and computer imaging were used to test several different prototypes and to arrive at the bristle configuration that was most effective in cleaning teeth.18 The role of the design in a product’s ultimate acceptance by customers is easily understood. Consider a new car design. If the new style is not that different from existing cars, customers might find it uninteresting or overly conservative. On the other hand, if the new design looks as if it came from Mars, most customers are likely to find it too revolutionary or even ugly. Given that as much as $2 billion 15 Mary G. Wojtas, “32nd PDMA International Conference Delivers Expert Insights, Knowledge, and Tools to Enhance Innovation Success,” Visions, Vol. 32, No. 4, December 2008, pp. 22–25. 16 The specifications are at www.apple.com/ipad/specs. 17 Karl H. E. Kroemer, “Ergonomics: Definition of Ergonomics,” National Safety Commission Web site (www.nsc.org). 18 Mark Maremont, “New Toothbrush Is Big-Ticket Item,” The Wall Street Journal, October 27, 1998, p. B-1. Chapter Thirteen Design 331 may be invested in a new car design, it seems reasonable for the car companies to spend as much as $1 million on getting just the right balance of style and shape. Focus groups may be used to get initial reactions, then full-size models (or car shapes on a computer screen) may be shown to hundreds of potential buyers. Despite careful research, however, misleading results may be obtained: Customers often don’t really know what they want as far as style is concerned.19 Product Architecture20 Product architecture has been described as the process by which a customer need is developed into a product design. This is a critical step in moving toward a product design, as solid architecture improves ultimate product performance, reduces the cost of changing the product once it is in production, and can speed the product to market. To understand architecture development, consider that a product contains components (a portable CD player-recorder has a chassis, motors, disk drive, speakers, and so on) that can be combined into chunks (the base, the disk handling system, the recording system, and the sound production system). A product is also composed of functional elements (for a CD player, these might include reading disks, recording sound, producing sound, and adjusting sound quality). The product’s architecture is how the functional elements are assigned to the chunks and how the chunks are interrelated. A Process for Product Architecture A stepwise process for product architecture development can be applied to make sure the product’s design will be in keeping with customer needs and, ultimately, the product innovation charter.21 The process is illustrated in simplified form in Figure 13.4. Careless product architecture results in products such as the CD-tape player with the mismatched controls, mentioned earlier in this chapter. Although each component works perfectly well, the way the pieces are put together makes little sense from the user’s perspective, and minor rearrangement would have resulted in an intuitive, easier-to-use product. 1. Create the Product Schematic. The schematic shows the components and functional elements of the product and how they are interconnected. Several alternative schematics may be developed and explored at this point. For the CD player-recorder, one might develop a version designed to plug into a 19 Tom Moulson and George Sproles, “Styling Strategy,” Business Horizons, September–October 2000, pp. 45–52. 20 Much of this section derives from David Cutherell, “Product Architecture,” in M. D. Rosenau, A. Griffin, G. Castellion, and N. Anscheutz (eds.), The PDMA Handbook of New Product Development (New York: John Wiley, 1996), pp. 217–235. 21 The stepwise process described here is based on that of Karl T. Ulrich and Steven D. Eppinger, Product Design and Development, 2nd ed. (Homewood, IL: Irwin/McGraw-Hill, 2000), Chapter 9. 332 Part Four Development FIGURE 13.4 Product Architecture Illustration Step 1: Product schematic Component of Component of recording system disk drive Component of output system Component of power source Step 2: Cluster schematic elements Chunk 1: Chunk 2: Chunk 3: Chunk 4: Recording Disk Drive Output Power Source Steps 3 and 4: Create geometric layout and check interactions (shown as arrows) Chunk 1 Chunk 4 Chunk 2 Chunk 3 standard stereo system, a stand-alone version with miniature speakers, or another to be used only with headphones. It would contain components connected with the disk drive itself, input (recording) functions, output (playback or speaker) functions, and power supply, among other things. 2. Cluster the Schematic Elements. Here, the chunks (or modules) are defined. In the figure, input, disk, output, and power chunks are identified. Interaction among the chunks should be simple so changes can be easily effected, and one should take advantage of manufacturing capabilities wherever possible. If rapid changes are expected in some part of the product, that part should most certainly be made into a chunk. For example, if new disk drive technology is expected to permit 10 times as much content to be recorded and stored on a quarter-sized disk, one should be able to replace the current drive with this new one if desired. 3. Create Geometric Layout. Here, using simulations, computer-aided design, or other techniques, the product is arranged in several configurations to determine the “best” solutions. For example, should the disk load in the front or the side of the CD player? Where should the speakers (if there are any) be located? One possible geometric layout is shown in Figure 13.4. 4. Check Interactions between Chunks. Understand what happens at the interfaces between chunks. In the CD player, sound flows as a digital signal to the disk during recording, and also as a digital signal from the disk during playback. Chapter Thirteen Design 333 Product Architecture and Product Platforms Clearly, careful product architecture development is critical to a firm seeking to establish a product platform. As noted in Chapter 3, car manufacturers (with few exceptions) think in terms of designing platforms, not individual products. A successful platform can result in an initially successful car, and also lead to several other models in the future (for example, the New Beetle is built on an existing Volkswagen Golf platform). If the architecture permits the designers to replace chunks or modules easily, several new products can be designed as technology improves, market tastes change, and manufacturing skill increases. This is how Black and Decker was able to develop those dozens of different hand-held tools on just a couple of basic motor platforms! In the Volkswagen example, the New Beetle is referred to as a derivative product. This term refers to products based on the same platform as an existing product, but modified incrementally in terms of technology or customer need fulfillment (in this case, a classic Beetle-like appearance). Depending on how many features are added, the derivative product may cost about the same to manufacture (such as new designs of Swatch watches), or may cost more but offer greater value to the user. Features may also be stripped out to achieve a lower-cost derivative product. Additional cost savings can be incurred by using standardized components across many products. Whatever the case, the key is to be able to make changes to the modules while still operating on the same platform. Industrial Design and the Industrial Designer22 Industrial designers are, above all, creative types: Their job is to take a problem and somehow visualize a solution to it. They are concerned about how things work as well as how things look. Their university training will have included work in aesthetic design, mechanical engineering, materials and processes, and art or drawing. It is this unique set of skills and abilities that determines the special role the product designer plays in the new products process. Consider this real-life example. An industrial designer was brought in by a leading manufacturer of liquid correction fluid (white fluid brushed over mistakes made when using a typewriter). A user problem was identified: The brushes got dried-out or misshapen, and thus became difficult to use. Some obvious solutions might be: Make the bottleneck bigger, or improve the brush applicator. But better product design work results in more creative solutions. To accomplish this, designers can use techniques similar to the general creativity techniques seen in Chapter 5, such as brainstorming. Working together with the marketing and engineering personnel from the product team, the designer can sketch hundreds of thumbnail ideations for review. For the correction fluid, these ideations included sketches of 22 Much of this section derives from Walter Herbst, “How Industrial Design Fits into Product Development,” in M. D. Rosenau, A. Griffin, G. Castellion, and N. Anscheutz (eds.), The PDMA Handbook of New Product Development (New York: John Wiley, 1996), pp. 237–251. 334 Part Four Development pens holding the white fluid, variations of the pen’s tip (including different angles, a spring-loaded version, and so on), different kinds of caps for the pen’s tip—even several versions of a dispenser much like a tape dispenser. Instead of using sketches, the ideations can also be computer-generated using Photoshop or similar software. The product team assesses each ideation based on appearance and manufacturability and chooses the best ones that are then more fully rendered by the designer. No one ideation is likely to be the final design concept to be brought to prototype development. The best parts of each ideation are combined into a single design in a step called design consolidation. As much detail as possible is fleshed out at this time— including decorative graphics and brand name and logo (if known), since this is typically one of the last evaluation points before a huge amount of financial and human resources are dedicated to the product. Generally, computer-generated renderings are preferred at this point. Other members of the new product team will provide information to determine if the product is manufacturable and marketable. Using these procedures, two new correction fluid products were designed and launched. The first put the liquid into a ballpoint-pen-type dispenser which, when squeezed, emitted a smooth flow of correction fluid right on the mistake. The second, which required two years of additional development, was a tape-dispenser that put a strip of dry white tape over the mistake (thus allowing the user to make the correction right away without waiting for liquid to dry). There are several factors that can be considered by industrial designers when deciding on the appropriateness of a design. These may include quality of user interface, emotional appeal, maintenance and repair, appropriate use of resources, and product differentiation (see Figure 13.5).23 Emotional appeal could include, for example, the sound made by a cell phone when the lid is closed. A solid “thud” is more appealing than a cheap “click.” Nokia knows this and Nokia engineers worked hard on the springs and ball bearings just to get the sound right.24 Industrial designers must also consider trade-offs among these factors. Bright colors on a phone answering machine may add to its emotional appeal but diminish perceived quality. Furthermore, many of these more aesthetic factors differ among individuals, making the designer’s job more difficult.25 Prototype Development26 For most people, the word prototype conjures up the image of a fully functioning, full-size product essentially ready to be examined by potential customers. Industrial designers define the term more broadly. A comprehensive prototype would 23 This set of assessment questions comes from Karl T. Ulrich and Steven D. Eppinger, Product Design and Development, 2nd ed. (Homewood, IL: Irwin/McGraw-Hill, 2000), pp. 227–230. 24 Anonymous, “Online Extra: A Chat with Nokia’s Alastair Curtis,” www.businessweek.com, July 17, 2006. 25 Mariëlle E. H. Creusen and Jan P. L. Schoormans, “The Different Roles of Product Appearance in Consumer Choice,” Journal of Product Innovation Management, 22(1), January 2005, pp. 63–81. 26 Much of this section derives from Ulrich and Eppinger, op. cit., Chapter 12. Chapter Thirteen Design 335 FIGURE 13.5 Assessment Factors for an Industrial Design: A Car Example Quality of the user interface Will the user understand the product and its intended use? Is it safe for use? In a car dashboard design, for example, is it clear that the knobs and switches for lights, wipers, and horn are easy to locate and operate? Emotional appeal Is it an attractive, exciting design? Would the prospective owner be proud to own the product? Does the car make a satisfying “growl” when revved up? Maintenance and repair Is the procedure for maintenance obvious and easy? Can all the fluids be easily changed, and is it easy to tell which fluid goes where? Appropriate use of resources Does the product include unnecessary features, or does it lack key features? Were the best materials chosen, with regard to cost and quality? Were environmental and ecological factors considered when choosing, for example, types of body paint for the car? Product differentiation Does the design distinguish the product? Is it memorable? Does it fit with corporate identity? When prospective luxury car owners take a look in the showroom, will they say this new model really stands out? Source: From Karl T. Ulrich and Steven D. Eppinger, Product Design and Development, 2nd ed., McGraw-Hill, 2000, pp. 227–230. Reprinted with permission of The McGraw-Hill Companies. be one of these essentially complete prototypes. They also make use of what are called focused prototypes, which examine a limited number of performance attributes or features. A bicycle or car manufacturer may build focused prototypes (a nonfunctioning bicycle out of foam or wood, or a wooden “frame” that very roughly simulates the layout of the seat, steering wheel, and dashboard of a new car interior) to determine customers’ reactions to the product’s form. The bike manufacturer may go on and develop a crude working prototype to experiment with and determine how the product might work. Recall the development of the Iomega Zip Drive from Chapter 2. In that case, dozens of nonworking prototypes of the Zip Drive, including some with a flip-up top, were built before arriving at a prototype that customers liked.27 Which type, or types, of prototypes should be built? The answer is, of course, it depends: Primarily, it depends on the intended use of the prototype. Focused prototypes are used in probe-and-learn (“lickety-stick”) product development in the 27 The Zip Drive story is told in Gary S. Lynn and Richard R. Reilly, Blockbusters: The Five Keys to Developing Great New Products (New York: HarperCollins, 2002). 336 Part Four Development development of new-to-the-world products, such as the Zip Drive. Focused prototypes are also used in cases where the product is not so new to the world to learn about how the product works and how well it will satisfy customer needs. BMW designers, for example, built clay models of new car designs for the 3 Series and sent them to southern France to see what they would look like in the sunlight at a distance, and to determine if there were line or form defects. It is much cheaper to make required changes now, rather than later in the development process.28 A more comprehensive physical prototype is necessary to determine how well all the components fit together—as an additional benefit, the various members of the new product team are essentially required to cooperate to build the comprehensive prototype. Finally, more advanced prototypes can be used as milestones— the performance of the prototype can be tracked periodically to see if it has advanced to desired levels. Once a comprehensive prototype exists, of course, it can be taken to potential users to be tested in a real usage situation, and improved and refined. This is known as product-use testing and will be taken up in Chapter 15. Managing the Interfaces in the Design Process New product managers have to keep in mind that product design should not be just the responsibility of the designers! Historically, in the era of powerful functional chimneys and slow, linear, stage-based development, industrial designers dominated the action in most firms making tangible products. Today, they have to share this traditional role with several other functions. Ironically, by joining the team and seeming to lose power, design stands on the verge of winning its ultimate position of influence. But it is the new product manager’s task to bring this about. There are several participants in the product design task, some in a more direct role than others, as shown in Figure 13.6. One model of how these people participate is shown in this figure. The representation there is somewhat linear, but with substantial overlapping or parallel effort. It is easy to see how this model of operations gives people problems, particularly the designers. Industrial designers, trained to develop aesthetics (styling), structural integrity, and function (how the product works), directly overlap with the design engineers, who are technical people who convert styling into product dimensions or specifications. Technical people are not devoid of ideas on styling, and stylists are not devoid of thoughts on how the mechanics can work. This is especially true on common products (like shoes or dinnerware) where all parties have experience. The other dimension of complexity is added by some of the supportive participants in the preceding list. Suppliers usually know their materials better than their customers do. That’s why Black & Decker picked its supplier for the Snake Lite before its design was finished. Large firms like Philips have the funds to establish large central styling centers where styling skills exceed those of the typical plant stylist. 28 C. Bangle, “The Ultimate Creativity Machine: How BMW Turns Art into Profit,” Harvard Business Review, 2001, pp. 47–55. Chapter Thirteen Design FIGURE 13.6 Percent of work being done Function design/styling Detail design Manufacturing process design Product for sale Technical design Protocol Model of the Product Design Process Envelope design Development time scale The members of a core team all participate in all four stages, but leadership in the first stage is often given to industrial designers, the middle two to engineering design, and the last to process design or manufacturing design. Terms in use vary widely. In chemical and pharmaceutical industries the design and engineering functions are replaced by research and development. And in some firms the term product engineering replaces engineering design; they want to contrast product engineer and process engineer. For services, the same steps apply, but instead of a "thing" we are developing a service sequence and technical capability. Think of an investment service developed in a financial institution, or a cable TV system, or office design service. Simultaneous with development (on goods and services) is the development of the augmented aspects of the product—pre- and postsale service, warranty, image, and so on. This activity, most often led by marketing people, is called envelope design, running across the bottom of the figure. Participants in the Design Process Direct Participants Research & Development Industrial Designers and Stylists Engineering Designers/Product Designers Manufacturing Engineers and System Designers Manufacturing Operations Supportive Participants Design Consultants Marketing Personnel Resellers Vendors/Suppliers Governments Customers Company Attorneys Technical Service 337 338 Part Four Development Customers almost always have overriding ideas to contribute. Consequently, the styling function is a synthesis of many views beyond those of the direct participants. If we add all of the other company people listed as supportive, we get back to the list of functions usually represented on the teams discussed in Chapter 14. The result of all this can be chaos, and in general the problems are thought to be at the heart of why some countries’ producers are so often beaten by new products from Japan and Germany. In Japan, for example, product design means more than how a product looks and feels to the user; it often means engineering applications. To one observer, design in Japan “means the total-enterprise process of determining customer needs and converting them to concepts, detailed designs, process plans, factory design, and delivered products, together with their supporting services.”29 This merges a holistic view of end-user needs and a holistic structure to meet those needs. Design is seen as a vertical means of fulfillment, and individual skills are not central. In the United States and Europe, participants end up playing musical chairs from one project to the next as roles change. Though the industrial designer is increasingly viewed as a full-fledged member of the new product team from the earliest phases, some design purists and traditionalists resist this movement. Design and marketing operate in drastically different cultures, and cultural gaps are hard to erase.30 In some cases, designers are taking on an expanded role as a liaison from end user to top management. Greater integration with end users can lead to better information about what design changes are desired. Designers can also serve as a conduit of information from industry, for example, making recommendations to the product development team on new materials to use.31 Both the design engineer and the stylist have been accused of continually trying to make a product just a little better and refusing to release it for production. There used to be a statement around the auto industry that engineering never released anything; the new car managers had to go in and take it away. Too much design retooling can result in products that have too many engineering characteristics or gimmicks and are late onto the market. The Apple Newton (an early personal digital assistant) and the Xerox 8200 copier are products that failed to live up to expectations, partly because of their complexity; 1980s-era PCs could also fit in this category—Apple’s initial success was based on its ease of use.32 The hard feelings sometimes run deep and lead to cross-functional animosity. The Japanese showed the world how to handle this when they began freezing the specifications at an early date in the technical cycle, forcing later ideas to be put into the schedule for the next model. 29 Daniel E. Whitney, “Integrated Design and Manufacturing in Japan,” Prism, Second Quarter, 1993, pp. 75–95. 30 Matthew K. Haggerty and Brian L. Vogel, Innovation, Winter 1992, pp. 8–13. 31 See Michael Evamy, op. cit.; and Jeneanne Marshall Rae, “Setting the Tone for Design Excellence,” Innovation, Fall 1994, pp. 7–9. 32 Paul A. Herbig and Hugh Kramer, “The Effect of Information Overload on the Innovation Choice Process,” Journal of Consumer Marketing, 11(2), 1994, pp. 45–54. Chapter Thirteen Design 339 Improving the Interfaces in the Design Process Most of the problems surrounding design have to do with concurrency, or overlapping the steps in development. It is clear from the discussion of Chapter 12 that up-front product definition (product protocol and firm prototype) is important. Several techniques are currently being used to make sure that design is integrated correctly with other functions during the development phase and that the products being designed can be manufactured in a cost-efficient way. Important among these is colocation (putting the various individuals or functional areas in close proximity). The development phase can be a communications snake pit. When the different groups are not in regular contact and cooperating, there is a tendency for information to be lost (or hidden). This causes wasted work and slows the whole operation down. Further, the problems intensify in large firms with their research centers hundreds of miles from the offices of marketers and the production lines of manufacturing people. Many firms have tried colocation to shorten communication lines and increase team cohesion. Motorola, for example, colocated its development team when developing the Bandit pager, completing the project in 18 months (less than half the normal development time). Many other firms such as Ford, Honda, AT&T, and John Deere have used colocation successfully.33 Colocation helps integrate departments and improve information flow, and also allows the team members to identify and resolve product development problems quicker. It must, however, be carefully planned and handled. It is probably not a good idea to break up a center of technological excellence in order to colocate its members. Too-distant colocation (i.e., employees have to get in their cars and drive to another building rather than walk down the hall) might lead to team members letting their problems pile up rather than resolving them immediately. There may be an unintentional home court advantage (if the meetings are at the marketing facility, marketing team members may be perceived to be more powerful). And team members must be willing to tear down the functional walls and change their attitudes about working with individuals from other functions—otherwise, colocation facilitates social exchange, but doesn’t really achieve cross-functional integration.34 In many firms, the effects of colocation are achieved without actual physical proximity of team members, using communications technology such as Lotus Notes or WebEx videoconferencing. This is sometimes k