Datasets:

pdf2dataset
/

d71d21b22639cf7caa8199fd302e9753

Modalities:

Text

Formats:

Size:

Libraries:

Dataset card Data Studio Files Files and versions

xet

Community

Dataset Viewer

Auto-converted to Parquet Duplicate

Split (1)

train · 869 rows

text stringlengths 0 3.43k

Database Marketing
Series Editor Jehoshua Eliashberg The Wharton School University of Pennsylvania Philadelphia, Pennsylvania USA Books in the series Blattberg, R., Kim, B., Neslin, S. Database Marketing: Analyzing and Managing Customers Ingene, C. A. and Parry, M. E. Mathematical Models of Distribution Channels Chakravarty, A. and Eliashberg, J. Managing Business Interfaces: Marketing, Engineering, and Manufacturing Perspectives Jorgensen, S. and Zaccour, G. Diﬀerential Games in Marketing Wind, Yoram (Jerry) and Green, Paul E. Marketing Research and Modeling: Progress and Prospects Erickson, Gary M. Dynamic Models of Advertising Competition, 2 nd Ed Hanssens, D., Parsons, L., and Schultz, R. Market Response Models: Econometric and Time Series Analysis, 2 nd Ed Mahajan, V., Muller, E. and Wind, Y. New-Product Diﬀusion Models Wierenga, B. and van Bruggen, G. Marketing Management Support Systems: Principles, Tools, and Implementation Leeﬂang, P., Wittink, D., Wedel, M. and Naert, P. Building Models for Marketing Decisions Wedel, M. and Kamakura, W. G. Market Segmentation, 2nd Ed Wedel, M. and Kamakura, W. G. Market Segmentation Nguyen, D. Marketing Decisions Under Uncertainty Laurent, G., Lilien, G. L., Pras, B. Research Traditions in Marketing Erickson, G. Dynamic Models of Advertising Competition Mc Cann, J. and Gallagher, J. Expert Systems for Scanner Data Environments Hanssens, D., Parsons, L., and Schultz, R. Market Response Models: Econometric and Time Series Analysis Cooper, L. and Nakanishi, M. Market Share Analysis
Robert C. Blattberg, Byung-Do Kim and Scott A. Neslin Database Marketing Analyzing and Managing Customers 123
Robert C. Blattberg Byung-Do Kim Kellogg School of Management Graduate School of Business Northwestern University Seoul National University Evanston, Illinois, USA Seoul, Korea and Tepper School of Business Carnegie-Mellon University Pittsburgh, Pennsylvania, USA Scott A. Neslin Tuck School of Business Dartmouth College Hanover, New Hampshire, USA Series Editor: Jehoshua Eliashberg The Wharton School University of Pennsylvania Philadelphia, Pennsylvania, USA Library of Congress Control Number: 2007936366 ISBN-13: 978-0-387-72578-9 e-ISBN-13: 978-0-387-72579-6 Printed on acid-free paper. ©2008 by Springer Science+Business Media, LLC All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval,electronic adaptation, computer software, or by similar or dissimilar methodology nowknow or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks and similar terms, even if the are not identiﬁed as such, is not to be taken as an expression of opinion as towhether or not they are subject to proprietary rights. 987654321 springer. com
To Our Spouses and Families
Preface The conﬂuence of more powerful information technology, advances in method-ology, and management's demand for an approach to marketing that is both eﬀective and accountable, has fueled explosive growth in the application of database marketing. In order to position the ﬁeld for future advances, we believe this is an opportune time to take stock of what we know about database marketingand identify where the knowledge gaps are. To do so, we have drawn on the rich and voluminous repository of research on database marketing. Our emphasis on research-academic, practitioner, and joint research-is driven by three factors. First, as we hope the book demonstrates, research hasproduced a great deal of knowledge about database marketing, which until now has not been collected and examined in one volume. Second, research isfundamentally a search for truth, and to enable future advances in the ﬁeld, we think it is crucial to separate what is known from what is conjectured. Third, the overlap between research and practice is particularly seamless inthis ﬁeld. Database marketing is a meritocracy-if a researcher can ﬁnd a method that oﬀers promise, a company can easily test it versus their current practice, and adopt the new method if it proves itself better. We have thus attempted to produce a research-based synthesis of the ﬁeld-a uniﬁed and comprehensive treatment of what research has taught usabout the methods and tools of database marketing. Our goals are to enhance research, teaching, and the practice of database marketing. Accordingly, this book potentially serves several audiences: Researchers : Researchers should be able to use the book to assess what is known about a particular topic, develop a list of research questions, anddraw on previous research along with newly developed methods to answerthese questions. Teachers : Teachers should ﬁnd this book useful to educate themselves about the ﬁeld and decide what content they need to teach. We trust thisbook will enable teachers to keep one step ahead of their students! vii
viii Preface Ph. D. Students : Ph. D. students should utilize this book to gain the re-quired background needed to conduct thesis research in the ﬁeld of database marketing. Advanced Business Students : By “advanced” business students, we mean undergraduate and MBA students who need a resource book that goes intodepth about a particular topic. We have found in teaching database marketing that it is very easy for the curious student to ask a question about topicssuch as predictive modeling, cross-selling, collaborative ﬁltering, or churn management that takes them beyond the depth that can be covered in class. This book is intended to provide that depth. Database Marketing Practitioners : This group encompasses those working in, working with, and managing marketing analytics groups in companiesand consulting ﬁrms. An IT specialist needs to understand for what pur-pose the data are to be used. A retention manager needs to know what is “out there” in terms of methods for decreasing customer churn. A senior manager may need insights on how to allocate funds to acquisition versusretention of customers. A statistician may need to understand how to con-struct a database marketing model that can be used to develop a customer-personalized cross-selling eﬀort. An analyst simply may need to understand what neural networks, Bayesian networks, and support vector machines are. We endeavor to provide answers to these and other relevant issues in thisbook. While it is true that database marketing has experienced explosive growth in the last decade, we have no doubt that the forces that produced thisgrowth-IT, methods and managerial imperatives-will continue. This book is based on the premise that research can contribute to this growth, and as a result, that database marketing's best days are ahead of it. We hope thisbook provides a platform that can be used to realize this potential. One of the most important aspects of database marketing is the interplay between method and application. Our goal is to provide an in-depth treat-ment of both of these elements of database marketing. Accordingly, there is a natural sectioning of the book in terms of method and application. Parts II-IV are mostly methodological chapters; Parts I, V, and IV cover application. Speciﬁcally, we structure the book as follows: Part I: Strategic Issues-We deﬁne the scope of the ﬁeld and the process of conducting database marketing (Chapter 1). That process begins with adatabase marketing strategy, which in turn leads to the question, what is the purpose and role of database marketing (Chapter 2)? We discuss this question in depth as well as two crucial factors that provide the backdrop for successful DBM: organizational structure and customer privacy (Chapters 3 and 4). Part II: Customer Lifetime Value (LTV)-Customer lifetime value is one of the pillars, along with predictive modeling and testing, upon whichdatabase marketing rests. We discuss methods for calculating LTV, includingproviding detailed coverage of the “thorny” issues such as cost accounting
Preface ix that are tempting to ignore, but whose resolution can have a crucial impact on practice (Chapters 5-7). Part III: Database Marketing Tools: The Basics-DBM has one ab-solute requirement-customer data. We discuss the sources and types ofcustomer data companies use (Chapter 8). We provide in-depth treatment of two other pillars of database marketing-testing and predictive modeling (Chapters 9-10). Part IV: Database Marketing Tools: Statistical Techniques-H e r ew ed i s-cuss the several statistical methods, both traditional and cutting edge, thatare used to produce predictive models (Chapters 11-19). This is a valuablesection for anyone wanting to know, “How is a decision tree produced,” or “What are the detailed considerations in using logistic regression,” or “Why is a neural net potentially better than a decision tree,” or “What is machinelearning all about?” Part V: Customer Management-Here we focus our attention squarely on application. We review the conceptual issues, what is known about them, andthe tools available to tackle customer management activities including acqui-sition, cross-and up-selling, churn management, frequency reward programs, customer tier programs, multichannel customer management, and acquisition and retention spending (Chapters 20-26). Part VI: Managing the Marketing Mix-We concentrate on communica-tions and pricing. We provide a thorough treatment of what we predict will bethe hallmark of the next generation of database marketing, namely “optimal contact models,” where the emphasis is on taking into account-in quanti-tative fashion-the future ramiﬁcations of current decisions, truly managing the long-term value of a customer (Chapter 28). We also discuss the design of DBM communications copy (Chapter 27) and several critical issues in pric-ing, including acquisition versus retention pricing, and the coordination of the two (Chapter 29). Our initial outline for this book took shape at the beginning of the mil-lennium, in May 2000. The irony of taking 7 years to write a book abouttechniques that often work in a matter of seconds does not escape us. In-deed, writing this book has been a matter of trying to hit a moving target. However, this eﬀort has been the proverbial “labor of love,” and its length and gestation period are products of the depth and scope we were aiming for. This book is the outcome of the debates we have had on issues such as how totreat ﬁxed costs in calculating customer lifetime value, which methods merit our attention and how exactly do they work, and why the multichannel cus-tomer is a higher-value customer. Writing this book has truly been a process, as is database marketing. Along the way, we have become indebted to numerous colleagues in both academia and business without whom this book would be a shadow of its current self. These people have provided working papers and references, ex-changed e-mails with us, talked with us, and ultimately, taught us a greatdeal about various aspects of database marketing. Included are: Kusum
x Preface Ailawadi, Eric Anderson, Kenneth Baker, Anand Bodapati, Bruce Hardie, Wai-Ki Ching, Kristoﬀ Coussement, Preyas Desai, Ravi Dhar, Jehoshua Eliashberg, Peter Fader, Doug Faherty, Helen Fanucci, Fred Feinberg, Edward Fox, Frances Frei, Steve Fuller, Bikram Prak Ghosh, Scott Gillum, William Greene, Abbie Griﬃn, John Hauser, Dick Hodges, Donna Hoﬀman, Eric J. Johnson, Wagner Kamakura, Gary King, George Knox, Praveen Kopalle, V. Kumar, Donald Lehmann, Peter Liberatore, Junxiang Lu, Charlotte Ma-son, Carl Mela, Prasad Naik, Koen Pauwels, Margaret Peteraf, Phil Pfeifer, Joseph Pych, Werner Reinartz, Richard Sansing, David Schmittlein, Robert Shumsky, K. Sudhir, Baohong Sun, Anant Sundaram, Jacquelyn Thomas,Glen Urban, Christophe Van den Bulte, Rajkumar Venkatesan, Julian Vil-lanueva, Florian von Wangenheim, Michel Wedel, Birger Wernerfeldt, and John Zhang. We are extremely grateful for research assistance provided by Carmen Maria Navarro (customer privacy practices), Jungho Bae and Ji Hong Min(data analysis), Qing-Lin Zhu and Paul Wolfson (simulation programming),and Karen Sluzenski (library references), and for manuscript preparation support tirelessly provided by Mary Biathrow, Deborah Gibbs, Patricia Hunt, and Carol Millay. We beneﬁted from two excellent reviews provided by Peter Verhoef and Ed Malthouse, which supplied insights on both the forest andthe trees that signiﬁcantly improved the ﬁnal product. The Springer publishing team was tremendously supportive, helpful, and extremely patient with our ﬁnal assembly of the book. We owe our deepgratitude to Deborah Doherty, Josh Eliashberg, Gillian Greenough, and Nick Philipson. While people write and support the book, we also want to acknowledge signiﬁcant institutional support that provided us with funding, facilities, anda stimulating environment in which to work. These include the Teradata Center for CRM at Fuqua Business School, Duke University, which hosted Scott Neslin during 2002, and our home institutions: the Kellogg School of Management, Northwestern; Seoul National University; and the Tuck School of Business, Dartmouth College. Finally, we owe our profound and deepest gratitude simply to our spouses and families, who provided the support, enduring patience, and companion-ship without which this book would never have materialized. By showing usthat family is what really matters, they enabled us to survive the ups and downs of putting together an eﬀort of this magnitude. It is to our spouses and families that we dedicate this book. R. Blattberg B. Kim S. Neslin
Contents Preface....................................................... v i i Part I Strategic Issues 1 Introduction.............................................. 3 1. 1 W h a t I s D a t a b a s e M a r k e t i n g ?............................ 3 1. 1. 1 D e ﬁ n i n g D a t a b a s e M a r k e t i n g...................... 4 1. 1. 2 Database Marketing, Direct Marketing, and Customer R e l a t i o n s h i p M a n a g e m e n t......................... 5 1. 2 Why Is Database Marketing Becoming More Important?.... 61. 3 The Database Marketing Process......................... 8 1. 4 O r g a n i z a t i o no ft h e B o o k................................ 1 2 2 Why Database Marketing?............................... 1 3 2. 1 E n h a n c i n g M a r k e t i n g P r o d u c t i v i t y....................... 1 3 2. 1. 1 T h e B a s i c A r g u m e n t.............................. 1 32. 1. 2 The Marketing Productivity Argument in Depth..... 15 2. 1. 3 Evidence for the Marketing Productivity Argument... 192. 1. 4 Assessment...................................... 22 2. 2 Creating and Enhancing Customer Relationships........... 23 2. 2. 1 T h e B a s i c A r g u m e n t.............................. 2 3 2. 2. 2 Customer Relationships and the Role of Database M a r k e t i n g....................................... 2 3 2. 2. 3 Evidence for the Argument that Database Marketing Enhances Customer Relationships.................. 28 2. 2. 4 Assessment...................................... 31 2. 3 Creating Sustainable Competitive Advantage.............. 32 2. 3. 1 T h e B a s i c A r g u m e n t.............................. 3 2 2. 3. 2 Evolution of the Sustainable Competitive Advantage A r g u m e n t....................................... 3 2 xi
xii Contents 2. 3. 3 Assessment...................................... 44 2. 4 S u m m a r y.............................................. 4 5 3 Organizing for Database Marketing....................... 4 7 3. 1 The Customer-Centric Organization...................... 47 3. 2 D a t a b a s e M a r k e t i n g S t r a t e g y............................ 4 8 3. 2. 1 S t r a t e g i e sf o r I m p l e m e n t i n g D B M.................. 4 93. 2. 2 Generating a Competitive Advantage............... 51 3. 2. 3 S u m m a r y........................................ 5 1 3. 3 Customer Management: The Structural Foundation of the Customer-Centric Organization.......................... 523. 3. 1 W h a t I s C u s t o m e r M a n a g e m e n t ?................... 5 2 3. 3. 2 The Motivation for Customer Management.......... 53 3. 3. 3 F o r m i n g C u s t o m e r P o r t f o l i o s...................... 5 4 3. 3. 4 Is Customer Management the Wave of the Future?... 55 3. 3. 5 Acquisition and Retention Departmentalization...... 56 3. 4 Processes for Managing Information: Knowledge Management 57 3. 4. 1 T h e C o n c e p t..................................... 5 73. 4. 2 Does Eﬀective Knowledge Management Enhance P e r f o r m a n c e ?.................................... 5 8 3. 4. 3 C r e a t i n g K n o w l e d g e.............................. 5 9 3. 4. 4 C o d i f y i n g K n o w l e d g e............................. 6 0 3. 4. 5 Transferring Knowledge........................... 613. 4. 6 U s i n g K n o w l e d g e................................. 6 2 3. 4. 7 Designing a Knowledge Management System......... 63 3. 4. 8 Issues and Challenges............................. 65 3. 5 C o m p e n s a t i o na n d I n c e n t i v e s............................ 6 5 3. 5. 1 T h e o r y.......................................... 6 63. 5. 2 Empirical Findings............................... 67 3. 5. 3 S u m m a r y........................................ 6 9 3. 6 P e o p l e................................................ 6 9 3. 6. 1 Providing Appropriate Support.................... 69 3. 6. 2 Intra-Firm Coordination.......................... 70 4 Customer Privacy and Database Marketing............... 7 5 4. 1 B a c k g r o u n d............................................ 7 5 4. 1. 1 Customer Privacy Concerns and Their Consequences for Database Marketers........................... 75 4. 1. 2 H i s t o r i c a l P e r s p e c t i v e............................. 7 8 4. 2 Customer Attitudes Toward Privacy...................... 79 4. 2. 1 S e g m e n t a t i o n S c h e m e s............................ 7 94. 2. 2 Impact of Attitudes on Database Marketing Behaviors 81 4. 2. 3 International Diﬀerences in Privacy Concerns........ 82 4. 3 C u r r e n t P r a c t i c e s R e g a r d i n g P r i v a c y...................... 8 5 4. 3. 1 Privacy Policies.................................. 85
Contents xiii 4. 3. 2 Collecting Data.................................. 87 4. 3. 3 The Legal Environment........................... 88 4. 4 P o t e n t i a l S o l u t i o n st o P r i v a c y C o n c e r n s................... 9 1 4. 4. 1 S o f t w a r e S o l u t i o n s................................ 9 14. 4. 2 R e g u l a t i o n...................................... 9 14. 4. 3 Permission Marketing............................. 94 4. 4. 4 C u s t o m e r D a t a O w n e r s h i p........................ 9 6 4. 4. 5 F o c u so n T r u s t................................... 9 74. 4. 6 Top Management Support......................... 984. 4. 7 Privacy as Proﬁt Maximization.................... 99 4. 5 S u m m a r ya n d A v e n u e sf o r R e s e a r c h...................... 1 0 0 Part II Customer Lifetime Value (LTV) 5 Customer Lifetime Value: Fundamentals.................. 1 0 5 5. 1 I n t r o d u c t i o n........................................... 1 0 5 5. 1. 1 Deﬁnition of Lifetime Value of a Customer.......... 1065. 1. 2 A Simple Example of Calculating C u s t o m e r L i f e t i m e V a l u e.......................... 1 0 6 5. 2 M a t h e m a t i c a l F o r m u l a t i o no f L T V........................ 1 0 85. 3 The Two Primary LTV Models: Simple R e t e n t i o na n d M i g r a t i o n................................ 1 0 95. 3. 1 S i m p l e R e t e n t i o n M o d e l s.......................... 1 0 95. 3. 2 M i g r a t i o n M o d e l s................................ 1 1 4 5. 4 LTV Models that Include Unobserved Customer Attrition..................................... 121 5. 5 E s t i m a t i n g R e v e n u e s.................................... 1 3 0 5. 5. 1 C o n s t a n t R e v e n u ep e r P e r i o d M o d e l................ 1 3 0 5. 5. 2 T r e n d M o d e l s.................................... 1 3 0 5. 5. 3 C a u s a l M o d e l s................................... 1 3 05. 5. 4 Stochastic Models of Purchase Rates and Volume..... 131 6 Issues in Computing Customer Lifetime Value............ 1 3 3 6. 1 I n t r o d u c t i o n........................................... 1 3 3 6. 2 D i s c o u n t R a t ea n d T i m e H o r i z o n......................... 1 3 4 6. 2. 1 Opportunity Cost of Capital Approach.............. 134 6. 2. 2 Discount Rate Based on the S o u r c e-o f-R i s k A p p r o a c h.......................... 1 4 0 6. 3 C u s t o m e r P o r t f o l i o M a n a g e m e n t......................... 1 4 26. 4 C o s t A c c o u n t i n g I s s u e s.................................. 1 4 5 6. 4. 1 Activity-Based Costing (ABC)..................... 145 6. 4. 2 Variable Costs and Allocating Fixed Overhead....... 148 6. 5 Incorporating Marketing Response........................ 154 6. 6 Incorporating Externalities.............................. 158
xiv Contents 7 Customer Lifetime Value Applications.................... 1 6 1 7. 1 Using LTV to Target Customer Acquisition................ 161 7. 2 Using LTV to Guide Customer Reactivation Strategies...... 1637. 3 U s i n g S M C ' s M o d e lt o V a l u e C u s t o m e r s................... 1 6 47. 4 A Case Example of Applying LTV Modeling............... 1687. 5 S e g m e n t a t i o n M e t h o d s U s i n g V a r i a n t so f L T V............. 1 7 2 7. 5. 1 Customer Pyramids.............................. 172 7. 5. 2 Creating Customer Portfolios Using LTV Measures... 174 7. 6 Drivers of the Components of LTV....................... 175 7. 7 Forcasting Potential LTV................................ 1767. 8 Valuing a Firm's Customer Base......................... 178 Part III Database Marketing Tools: The Basics 8 Sources of Data........................................... 1 8 3 8. 1 I n t r o d u c t i o n........................................... 1 8 38. 2 T y p e so f D a t af o r D e s c r i b i n g C u s t o m e r s................... 1 8 4 8. 2. 1 Customer Identiﬁcation Data...................... 1848. 2. 2 Demographic Data............................... 1868. 2. 3 P s y c h o g r a p h i co r L i f e s t y l e D a t a.................... 1 8 68. 2. 4 T r a n s a c t i o n D a t a................................. 1 8 88. 2. 5 M a r k e t i n g A c t i o n D a t a........................... 1 9 08. 2. 6 Other Types of Data.............................. 191 8. 3 S o u r c e so f C u s t o m e r I n f o r m a t i o n......................... 1 9 1 8. 3. 1 I n t e r n a l( S e c o n d a r y )D a t a......................... 1 9 2 8. 3. 2 E x t e r n a l( S e c o n d a r y )D a t a........................ 1 9 3 8. 3. 3 Primary Data.................................... 211 8. 4 T h e D e s t i n a t i o n M a r k e t i n g C o m p a n y..................... 2 1 3 9 Test Design and Analysis................................. 2 1 5 9. 1 T h e I m p o r t a n c eo f T e s t i n g............................... 2 1 59. 2 T o T e s to r N o tt o T e s t.................................. 2 1 6 9. 2. 1 V a l u eo f I n f o r m a t i o n.............................. 2 1 69. 2. 2 Assessing Mistargeting Costs...................... 221 9. 3 S a m p l i n g T e c h n i q u e s.................................... 2 2 3 9. 3. 1 Probability Versus Nonprobability Sampling......... 2249. 3. 2 Simple Random Sampling......................... 2249. 3. 3 Systematic Random Sampling...................... 2259. 3. 4 O t h e r S a m p l i n g T e c h n i q u e s........................ 2 2 6 9. 4 D e t e r m i n i n gt h e S a m p l e S i z e............................. 2 2 7 9. 4. 1 S t a t i s t i c a l A p p r o a c h.............................. 2 2 79. 4. 2 D e c i s i o n T h e o r e t i c A p p r o a c h...................... 2 2 9 9. 5 T e s t D e s i g n s........................................... 2 3 5 9. 5. 1 Single Factor Experiments......................... 235
Contents xv 9. 5. 2 Multifactor Experiments: Full Factorials............. 238 9. 5. 3 Multifactor Experiments: Orthogonal Designs........ 241 9. 5. 4 Q u a s i-E x p e r i m e n t s............................... 2 4 3 10 The Predictive Modeling Process......................... 2 4 5 10. 1 Predictive Modelling and the Quest for M a r k e t i n g P r o d u c t i v i t y................................. 2 4 5 10. 2 The Predictive Modeling Process: Overview................ 248 10. 3 The Process in Detail................................... 248 1 0. 3. 1 D e ﬁ n et h e P r o b l e m............................... 2 4 8 1 0. 3. 2 P r e p a r et h e D a t a................................. 2 5 0 1 0. 3. 3 E s t i m a t et h e M o d e l............................... 2 5 61 0. 3. 4 E v a l u a t et h e M o d e l............................... 2 5 91 0. 3. 5 S e l e c t C u s t o m e r st o T a r g e t........................ 2 6 7 10. 4 A Predictive Modeling Example.......................... 2751 0. 5 L o n g-T e r m C o n s i d e r a t i o n s............................... 2 8 0 10. 5. 1 Preaching to the Choir............................ 28010. 5. 2 Model Shelf Life and Selectivity Bias............... 280 10. 5. 3 Learning from the Interpretation of P r e d i c t i v e M o d e l s................................ 2 8 4 10. 5. 4 Predictive Modeling Is a Process t o B e M a n a g e d................................... 2 8 5 10. 6 Future Research........................................ 286 Part IV Database Marketing Tools: Statistical Techniques 11 Statistical Issues in Predictive Modeling.................. 2 9 1 11. 1 Economic Justiﬁcation for Building a Statistical Model...... 292 1 1. 2 S e l e c t i o no f V a r i a b l e sa n d M o d e l s........................ 2 9 3 1 1. 2. 1 V a r i a b l e S e l e c t i o n................................ 2 9 31 1. 2. 2 V a r i a b l e T r a n s f o r m a t i o n s.......................... 2 9 9 1 1. 3 T r e a t m e n to f M i s s i n g V a r i a b l e s........................... 3 0 1 1 1. 3. 1 C a s e w i s e D e l e t i o n................................ 3 0 21 1. 3. 2 P a i r w i s e D e l e t i o n................................. 3 0 211. 3. 3 Single Imputation................................ 30211. 3. 4 Multiple Imputation.............................. 303 1 1. 3. 5 D a t a F u s i o n..................................... 3 0 5 1 1. 3. 6 M i s s i n g V a r i a b l e D u m m i e s........................ 3 0 7 1 1. 4 E v a l u a t i o no f S t a t i s t i c a l M o d e l s.......................... 3 0 8 11. 4. 1 Dividing the Sample into the Calibration and Validation Sample................................ 309 11. 4. 2 Evaluation Criteria............................... 312 11. 5 Concluding Note: Evolutionary Model-Building............. 321
xvi Contents 12 RFM Analysis............................................ 3 2 3 1 2. 1 I n t r o d u c t i o n........................................... 3 2 3 1 2. 2 T h e B a s i c so ft h e R F MM o d e l........................... 3 2 4 12. 2. 1 Deﬁnition of Recency, Frequency, and M o n e t a r y V a l u e.................................. 3 2 4 1 2. 2. 2 R F Mf o r S e g m e n t-L e v e l P r e d i c t i o n................. 3 2 6 12. 3 Breakeven Analysis: Determining the Cutoﬀ Point.......... 327 12. 3. 1 Proﬁt Maximizing Cutoﬀ Response Probability....... 3281 2. 3. 2 H e t e r o g e n e o u s O r d e r A m o u n t s..................... 3 2 9 12. 4 Extending the RFM Model.............................. 331 1 2. 4. 1 T r e a t i n gt h e R F MM o d e la s A N O V A............... 3 3 1 12. 4. 2 Alternative Response Models Without Discretization.. 334 12. 4. 3 A Stochastic RFM Model by Colombo and Jiang (1999)..................................... 336 13 Market Basket Analysis.................................. 3 3 9 1 3. 1 I n t r o d u c t i o n........................................... 3 3 9 13. 2 Beneﬁts for Marketers................................... 340 13. 3 Deriving Market Basket Association Rules................. 341 1 3. 3. 1 S e t u po fa M a r k e t B a s k e t P r o b l e m................. 3 4 113. 3. 2 Deriving “Interesting” Association Rules............ 342 13. 3. 3 Zhang (2000) Measures of Association a n d D i s s o c i a t i o n................................. 3 4 5 13. 4 Issues in Market Basket Analysis......................... 346 13. 4. 1 Using Taxonomies to Overcome the Dimensionality P r o b l e m......................................... 3 4 6 13. 4. 2 Association Rules for More than Two Items......... 347 13. 4. 3 Adding Virtual Items to Enrich the Quality of the M a r k e t B a s k e t A n a l y s i s........................... 3 4 8 13. 4. 4 Adding Temporal Component to the Market Basket A n a l y s i s......................................... 3 4 9 1 3. 5 C o n c l u s i o n............................................ 3 5 0 14 Collaborative Filtering.................................... 3 5 3 1 4. 1 I n t r o d u c t i o n........................................... 3 5 31 4. 2 M e m o r y-B a s e d M e t h o d s................................. 3 5 4 1 4. 2. 1 C o m p u t i n g S i m i l a r i t y B e t w e e n U s e r s................ 3 5 614. 2. 2 Evaluation Metrics............................... 360 14. 3 Model-Based Methods.................................. 363 1 4. 3. 1 T h e C l u s t e r M o d e l............................... 3 6 414. 3. 2 Item-Based Collaborative Filtering................. 36414. 3. 3 A Bayesian Mixture Model by Chien and George (1999).......................... 366 14. 3. 4 A Hierarchical Bayesian Approach by Ansari et al. (2000).......................................... 366
Contents xvii 14. 4 Current Issues in Collaborative Filtering.................. 368 14. 4. 1 Combining Content-Based Information Filtering with Collaborative Filtering............................ 368 14. 4. 2 Implicit Ratings.................................. 372 1 4. 4. 3 S e l e c t i o n B i a s.................................... 3 7 4 1 4. 4. 4 R e c o m m e n d a t i o n s A c r o s s C a t e g o r i e s................ 3 7 5 15 Discrete Dependent Variables and Duration Models...... 3 7 7 15. 1 Binary Response Model................................. 378 15. 1. 1 Linear Probability Model.......................... 37815. 1. 2 Binary Logit (or Logistic Regression) and Probit M o d e l s.......................................... 3 7 9 15. 1. 3 Logistic Regression with Rare Events Data.......... 3821 5. 1. 4 D i s c r i m i n a n t A n a l y s i s............................. 3 8 5 15. 2 Multinomial Response Model............................ 386 1 5. 3 M o d e l sf o r C o u n t D a t a.................................. 3 8 8 15. 3. 1 Poisson Regression............................... 388 15. 3. 2 Negative Binomial Regression...................... 389 15. 4 Censored Regression (Tobit) Models and Extensions........ 3901 5. 5 T i m e D u r a t i o n( H a z a r d )M o d e l s.......................... 3 9 2 1 5. 5. 1 C h a r a c t e r i s t i c so f D u r a t i o n D a t a................... 3 9 3 15. 5. 2 Analysis of Duration Data Using a Classical Linear Regression....................................... 3 9 4 1 5. 5. 3 H a z a r d M o d e l s................................... 3 9 5 15. 5. 4 Incorporating Covariates into the Hazard Function... 398 16 Cluster Analysis.......................................... 4 0 1 1 6. 1 I n t r o d u c t i o n........................................... 4 0 116. 2 The Clustering Process.................................. 402 1 6. 2. 1 S e l e c t i n g C l u s t e r i n g V a r i a b l e s...................... 4 0 31 6. 2. 2 S i m i l a r i t y M e a s u r e s............................... 4 0 41 6. 2. 3 C l u s t e r i n g M e t h o d s............................... 4 0 8 1 6. 2. 4 T h e N u m b e ro f C l u s t e r s........................... 4 1 8 16. 3 Applying Cluster Analysis............................... 419 1 6. 3. 1 I n t e r p r e t i n gt h e R e s u l t s........................... 4 1 91 6. 3. 2 T a r g e t i n gt h e D e s i r e d C l u s t e r...................... 4 2 0 17 Decision Trees............................................ 4 2 3 1 7. 1 I n t r o d u c t i o n........................................... 4 2 3 17. 2 Fundamentals of Decision Trees.......................... 42417. 3 Finding the Best Splitting Rule.......................... 427 1 7. 3. 1 G i n i I n d e xo f D i v e r s i t y............................ 4 2 717. 3. 2 Entropy and Information Theoretic Measures........ 4291 7. 3. 3 C h i-S q u a r e T e s t.................................. 4 3 0 17. 3. 4 Other Splitting Rules............................. 432
xviii Contents 1 7. 4 F i n d i n gt h e R i g h t S i z e d T r e e............................. 4 3 2 1 7. 4. 1 P r u n i n g......................................... 4 3 2 17. 4. 2 Other Methods for Finding the Right Sized Tree..... 434 17. 5 Other Issues in Decision Trees............................ 435 17. 5. 1 Multivariate Splits................................ 4361 7. 5. 2 C o s t C o n s i d e r a t i o n s.............................. 4 3 6 17. 5. 3 Finding an Optimal Tree.......................... 436 1 7. 6 A p p l i c a t i o nt oa D i r e c t M a i l O ﬀ e r........................ 4 3 7 17. 7 Strengths and Weaknesses of Decision Trees............... 438 18 Artiﬁcial Neural Networks................................ 4 4 3 1 8. 1 I n t r o d u c t i o n........................................... 4 4 3 1 8. 1. 1 H i s t o r i c a l R e m a r k s............................... 4 4 318. 1. 2 ANN Applications in Database Marketing........... 44418. 1. 3 Strengths and Weaknesses......................... 445 1 8. 2 M o d e l so f N e u r o n s...................................... 4 4 718. 3 Multilayer Perceptrons.................................. 450 1 8. 3. 1 N e t w o r k A r c h i t e c t u r e............................. 4 5 118. 3. 2 Back-Propagation Algorithm....................... 4541 8. 3. 3 A p p l i c a t i o nt o C r e d i t S c o r i n g...................... 4 5 5 18. 3. 4 Optimal Number of Units in the Hidden Layer, Learning-Rate, and Momentum Parameters.......... 457 18. 3. 5 Stopping Criteria................................. 457 18. 3. 6 Feature (Input Variable) Selection.................. 458 18. 3. 7 Assessing the Importance of the Input Variables...... 459 1 8. 4 R a d i a l-B a s i s F u n c t i o n N e t w o r k s.......................... 4 6 0 1 8. 4. 1 B a c k g r o u n d..................................... 4 6 018. 4. 2 A Curve-Fitting (Approximation) Problem.......... 461 18. 4. 3 Application Example............................. 463 19 Machine Learning........................................ 4 6 5 1 9. 1 I n t r o d u c t i o n........................................... 4 6 51 9. 2 1-R u l e................................................ 4 6 6 19. 3 Rule Induction by Covering Algorithms................... 468 19. 3. 1 Covering Algorithms and Decision Trees............. 4691 9. 3. 2 P R I S M......................................... 4 7 0 19. 3. 3 A Probability Measure for Rule Evaluation and the INDUCT Algorithm....................... 474 19. 4 Instance-Based Learning................................ 477 19. 4. 1 Strengths and Limitations......................... 478 19. 4. 2 A Brief Description of an Instance-Based Learning Algorithm....................................... 478 1 9. 4. 3 S e l e c t i o no f E x e m p l a r s............................ 4 7 9 19. 4. 4 Attribute Weights................................ 481 19. 5 Genetic Algorithms..................................... 481
Contents xix 1 9. 6 B a y e s i a n N e t w o r k s..................................... 4 8 4 19. 7 Support Vector Machines................................ 48619. 8 Combining Multiple Models: Committee Machines.......... 489 1 9. 8. 1 B a g g i n g......................................... 4 9 01 9. 8. 2 B o o s t i n g........................................ 4 9 119. 8. 3 Other Committee Machines........................ 492 Part V Customer Management 20 Acquiring Customers..................................... 4 9 5 2 0. 1 I n t r o d u c t i o n........................................... 4 9 520. 2 The Fundamental Equation of Customer Equity............ 496 20. 3 Acquisition Costs....................................... 497 20. 4 Strategies for Increasing Number of C u s t o m e r s A c q u i r e d.................................... 4 9 9 20. 4. 1 Increasing Market Size............................ 499 20. 4. 2 Increasing Marketing Acquisition Expenditures....... 50020. 4. 3 Changing the Shape of the Acquisition Curve........ 5012 0. 4. 4 U s i n g L e a d P r o d u c t s............................. 5 0 3 20. 4. 5 Acquisition Pricing and Promotions................ 504 20. 5 Developing a Customer Acquisition Program............... 505 2 0. 5. 1 F r a m e w o r k...................................... 5 0 5 20. 5. 2 Segmentation, Targeting and Positioning (STP)...... 506 20. 5. 3 Product/Service Oﬀering.......................... 507 20. 5. 4 Acquisition Targeting............................. 50820. 5. 5 Targeting Methods for Customer Acquisition......... 510 20. 6 Research Issues in Acquisition Marketing.................. 514 21 Cross-Selling and Up-Selling.............................. 5 1 5 21. 1 The Strategy.......................................... 51521. 2 Cross-Selling Models.................................... 516 2 1. 2. 1 N e x t-P r o d u c t-t o-B u y M o d e l s...................... 5 1 721. 2. 2 Next-Product-to-Buy Models with Explicit C o n s i d e r a t i o no f P u r c h a s e T i m i n g.................. 5 2 9 21. 2. 3 Next-Product-to-Buy with Timing and Response..... 534 21. 3 Up-Selling............................................. 537 2 1. 3. 1 AD a t a E n v e l o p e A n a l y s i s M o d e l................... 5 3 8 2 1. 3. 2 AS t o c h a s t i c F r o n t i e r M o d e l....................... 5 4 0 21. 4 Developing an Ongoing Cross-Selling Eﬀort................ 541 21. 4. 1 Process Overview................................. 54121. 4. 2 Strategy......................................... 541 2 1. 4. 3 D a t a C o l l e c t i o n.................................. 5 4 421. 4. 4 Analytics........................................ 5442 1. 4. 5 I m p l e m e n t a t i o n.................................. 5 4 6
xx Contents 2 1. 4. 6 E v a l u a t i o n...................................... 5 4 6 21. 5 Research Needs........................................ 547 22 Frequency Reward Programs............................. 5 4 9 22. 1 Deﬁnition and Motivation............................... 549 22. 2 How Frequency Reward Programs Inﬂuence Customer B e h a v i o r.............................................. 5 5 0 2 2. 2. 1 M e c h a n i s m sf o r I n c r e a s i n g S a l e s.................... 5 5 0 22. 2. 2 What We Know About How Customers Respond to R e w a r d P r o g r a m s................................ 5 5 2 22. 3 Do Frequency Reward Programs Increase Proﬁts in a Competitive Environment?.............................. 562 22. 4 Frequency Reward Program Design....................... 565 2 2. 4. 1 D e s i g n D e c i s i o n s................................. 5 6 52 2. 4. 2 I n f r a s t r u c t u r e.................................... 5 6 52 2. 4. 3 E n r o l l m e n t P r o c e d u r e s............................ 5 6 62 2. 4. 4 R e w a r d S c h e d u l e................................. 5 6 62 2. 4. 5 T h e R e w a r d..................................... 5 6 92 2. 4. 6 P e r s o n a l i z e d M a r k e t i n g........................... 5 7 12 2. 4. 7 P a r t n e r i n g....................................... 5 7 222. 4. 8 Monitor and Evaluate............................. 573 2 2. 5 F r e q u e n c y R e w a r d P r o g r a m E x a m p l e s.................... 5 7 3 22. 5. 1 Harrah's Entertainment 1.......................... 5 7 3 22. 5. 2 The UK Supermarket Industry: Nectar V e r s u s C l u b c a r d................................. 5 7 4 2 2. 5. 3 C i n g u l a r R o l l o v e r M i n u t e s......................... 5 7 622. 5. 4 Hilton Hotels.................................... 576 22. 6 Research Needs........................................ 578 23 Customer Tier Programs................................. 5 7 9 23. 1 Deﬁnition and Motivation............................... 5792 3. 2 D e s i g n i n g C u s t o m e r T i e r P r o g r a m s....................... 5 8 1 23. 2. 1 Overview........................................ 5812 3. 2. 2 R e v i e w O b j e c t i v e s................................ 5 8 22 3. 2. 3 C r e a t et h e C u s t o m e r D a t a b a s e..................... 5 8 2 2 3. 2. 4 D e ﬁ n e T i e r s..................................... 5 8 2 23. 2. 5 Determine Acquisition Potential for Each Tier....... 58423. 2. 6 Determine Development Potential for Each Tier...... 58523. 2. 7 Allocate Funds to Tiers........................... 5882 3. 2. 8 D e s i g n T i e r-S p e c i ﬁ c P r o g r a m s..................... 5 9 52 3. 2. 9 I m p l e m e n ta n d E v a l u a t e.......................... 5 9 6 2 3. 3 E x a m p l e so f C u s t o m e r T i e r P r o g r a m s..................... 5 9 7 23. 3. 1 Bank One (Hartfeil 1996).......................... 59723. 3. 2 Royal Bank of Canada (Rasmusson 1999)............ 59823. 3. 3 Thomas Cook Travel (Rasmusson 1999)............. 598
Contents xxi 23. 3. 4 Canadian Grocery Store Chain (Grant and Schlesinger 1995)................................. 598 23. 3. 5 Major US Bank (Rust et al. 2000).................. 599 23. 3. 6 Viking Oﬃce Products (Miller 2001)................ 600 23. 3. 7 Swedbank (Storbacka and Luukinen 1994, see also Storbacka 1993).................................. 600 23. 4 Risks in Implementing Customer Tier Programs............ 601 23. 5 Future Research Requirements........................... 604 24 Churn Management...................................... 6 0 7 2 4. 1 T h e P r o b l e m........................................... 6 0 72 4. 2 F a c t o r st h a t C a u s e C h u r n............................... 6 1 12 4. 3 P r e d i c t i n g C u s t o m e r C h u r n.............................. 6 1 5 2 4. 3. 1 S i n g l e F u t u r e P e r i o d M o d e l s....................... 6 1 6 2 4. 3. 2 T i m e S e r i e s M o d e l s............................... 6 2 2 2 4. 4 M a n a g e r i a l A p p r o a c h e st o R e d u c i n g C h u r n................ 6 2 5 24. 4. 1 Overview........................................ 62524. 4. 2 A Framework for Proactive Churn Management...... 627 24. 4. 3 Implementing a Proactive Churn M a n a g e m e n t P r o g r a m............................ 6 3 1 24. 5 Future Research........................................ 633 25 Multichannel Customer Management..................... 6 3 5 25. 1 The Emergence of Multichannel C u s t o m e r M a n a g e m e n t.................................. 6 3 625. 1. 1 The Push Toward Multichannel.................... 636 25. 1. 2 The Pull of Multichannel.......................... 636 25. 2 The Multichannel Customer............................. 637 25. 2. 1 A Framework for Studying the Customer's Channel C h o i c e D e c i s i o n.................................. 6 3 7 25. 2. 2 Characteristics of Multichannel Customers........... 6382 5. 2. 3 D e t e r m i n a n t so f C h a n n e l C h o i c e................... 6 4 1 2 5. 2. 4 M o d e l so f C u s t o m e r C h a n n e l M i g r a t i o n............. 6 4 725. 2. 5 Research Shopping............................... 652 25. 2. 6 Channel Usage and Customer Loyalty............... 655 25. 2. 7 The Impact of Acquisition Channel o n C u s t o m e r B e h a v i o r............................ 6 5 5 25. 2. 8 The Impact of Channel Introduction o n F i r m P e r f o r m a n c e............................. 6 5 7 25. 3 Developing Multichannel Strategies....................... 659 25. 3. 1 Framework for the Multichannel Design Process...... 659 2 5. 3. 2 A n a l y z e C u s t o m e r s............................... 6 5 9 2 5. 3. 3 D e s i g n C h a n n e l s................................. 6 6 12 5. 3. 4 I m p l e m e n t a t i o n.................................. 6 6 7 2 5. 3. 5 E v a l u a t i o n...................................... 6 6 8
xxii Contents 25. 4 Industry Examples..................................... 672 25. 4. 1 Retail “Best Practice” (Crawford 2002)............. 672 25. 4. 2 Waters Corporation (CRM ROI Review 2003)........ 67225. 4. 3 The Pharmaceutical Industry (Boehm 2002)......... 67325. 4. 4 Circuit City (Smith 2006; Wolf 2006)............... 6742 5. 4. 5 S u m m a r y........................................ 6 7 4 26 Acquisition and Retention Management.................. 6 7 5 2 6. 1 I n t r o d u c t i o n........................................... 6 7 5 26. 2 Modeling Acquisition and Retention...................... 676 26. 2. 1 The Blattberg and Deighton (1996) Model........... 676 2 6. 2. 2 C o h o r t M o d e l s................................... 6 8 22 6. 2. 3 T y p e I IT o b i t M o d e l s............................. 6 8 226. 2. 4 Competitive Models.............................. 68726. 2. 5 Summary: Lessons on How to Model Acquisition and R e t e n t i o n....................................... 6 8 9 26. 3 Optimal Acquisition and Retention Spending.............. 690 26. 3. 1 Optimizing the Blattberg/Deighton Model with No B u d g e t C o n s t r a i n t................................ 6 9 1 26. 3. 2 The Relationship Among Acquisition and Retention Costs, LTV, and Optimal Spending: If Acquisition “Costs” Exceed Retention “Costs”, Should the Firm F o c u so n R e t e n t i o n ?.............................. 6 9 5 26. 3. 3 Optimizing the Budget-Constrained B l a t t b e r g / D e i g h t o n M o d e l......................... 6 9 8 26. 3. 4 Optimizing a Multi-Period, Budget-Constrained C o h o r t M o d e l.................................... 7 0 2 26. 3. 5 Optimizing the Reinartz et al. (2005) T o b i t M o d e l..................................... 7 0 5 26. 3. 6 Summary: When Should We Spend More on Acquisition or Retention?......................... 706 26. 4 Acquisition and Retention Budget Planning................ 708 26. 4. 1 The Customer Management Marketing Budget ( C M M B )........................................ 7 0 8 2 6. 4. 2 I m p l e m e n t a t i o n I s s u e s............................ 7 0 9 26. 5 Acquisition and Retention Strategy: An Overall Framework.. 710 Part VI Managing the Marketing Mix 27 Designing Database Marketing Communications.......... 7 1 5 27. 1 The Planning Process................................... 715 27. 2 Setting the Overall Plan................................. 716 2 7. 2. 1 O b j e c t i v e s....................................... 7 1 6 27. 2. 2 Strategy......................................... 717
Contents xxiii 2 7. 2. 3 B u d g e t.......................................... 7 1 7 2 7. 2. 4 S u m m a r y........................................ 7 1 8 27. 3 Developing Copy....................................... 719 27. 3. 1 Creative Strategy................................ 7192 7. 3. 2 T h e O ﬀ e r....................................... 7 2 32 7. 3. 3 T h e P r o d u c t..................................... 7 2 627. 3. 4 Personalizing Multiple Components of the Communication.................................. 736 2 7. 4 S e l e c t i n g M e d i a........................................ 7 3 7 27. 4. 1 Optimization.................................... 7372 7. 4. 2 I n t e g r a t e d M a r k e t i n g C o m m u n i c a t i o n s.............. 7 3 9 27. 5 Evaluating Communications Programs.................... 739 28 Multiple Campaign Management......................... 7 4 3 28. 1 Overview.............................................. 74328. 2 Dynamic Response Phenomena........................... 744 28. 2. 1 Wear-in, Wear-out, and Forgetting.................. 7442 8. 2. 2 O v e r l a p......................................... 7 4 9 28. 2. 3 Purchase Acceleration, Loyalty, and Price Sensitivity Eﬀects....................... 750 28. 2. 4 Including Wear-in, Wear-out, Forgetting, Overlap, Acceleration, and Loyalty......................... 752 28. 3 Optimal Contact Models................................ 753 28. 3. 1 A Promotions Model (Ching et al. 2004)............ 75528. 3. 2 Using a Decision Tree Response Model (Simester et al. 2006)............................. 756 28. 3. 3 Using a Hazard Response Model (G¨on¨ ul et al. 2000)............................... 758 28. 3. 4 Using a Hierarchical Bayes Model (Rust and Verhoef 2005)........................................... 760 28. 3. 5 Incorporating Customer and Firm Dynamic Rationality (G¨ on¨ul and Shi 1998)................... 763 28. 3. 6 Incorporating Inventory Management (Bitran and Mondschein 1996)................................ 765 28. 3. 7 Incorporating a Variety of Catalogs (Campbell et al. 2001)............................ 768 28. 3. 8 Multiple Catalog Mailings (Elsner et al. 2003, 2004)................................. 772 28. 3. 9 Increasing Response to Online Panel Surveys (Neslin et al. 2007)........................ 774 2 8. 4 S u m m a r y.............................................. 7 7 7 29 Pricing................................................... 7 8 1 29. 1 Overview-Customer-based Pricing....................... 781
xxiv Contents 29. 2 Customer Pricing when Customers Can Purchase Multiple O n e-T i m e P r o d u c t sf r o mt h e F i r m........................ 7 8 3 29. 2. 1 Case 1: Only Product 1 Is Purchased............... 786 29. 2. 2 Case 2: Two Product Purchase Model with Lead P r o d u c t1....................................... 7 8 6 29. 3 Pricing the Same Products/Services to Customers o v e r T w o P e r i o d s....................................... 7 8 829. 3. 1 Pessimistic Case: R<q-Expectations of Quality a r e L e s st h a n A c t u a l Q u a l i t y...................... 7 8 9 29. 3. 2 Optimistic Case: R>q-Expectations of Quality are Greater than Actual Quality............ 790 29. 3. 3 Research Issues.................................. 790 29. 4 Acquisition and Retention Pricing Using the Customer E q u i t y M o d e l.......................................... 7 9 1 29. 5 Pricing to Recapture Customers.......................... 7942 9. 6 P r i c i n g A d d-o n S a l e s.................................... 7 9 629. 7 Price Discrimination Through Database Targeting Models... 797 References.................................................... 8 0 1 Author Index................................................. 8 4 7 Subject Index................................................ 8 5 9
Chapter 1 Introduction Abstract Database marketing is “the use of customer databases to enhance marketing productivity through more eﬀective acquisition, retention, and de-velopment of customers. ” In this chapter we elaborate on this deﬁnition, pro-vide an overview of why database marketing is becoming more important, and propose a framework for the “database marketing process. ” We conclude with a discussion of how we organize the book. 1. 1 What Is Database Marketing? The purpose of marketing is to enable the ﬁrm to enhance customer value. In today's competitive, information-intensive, ROI-oriented business environ-ment, database marketing has emerged as an invaluable approach for achiev-ing this purpose. The applications of database marketing are numerous and growing exponentially. Here are a few examples: “Internet Portal, Inc. ” determines which of its customers will be most receptive to targeted eﬀorts to increase their usage of the portal. Perhaps more importantly, it determines which customers will notbe receptive to these eﬀorts. “XYZ Bank” decides which of its many ﬁnancial products should be mar-keted to which of its current customers. “ABC Wireless” develops the ability to predict which customers are mostlikely to leave when their contract runs out, and designs a “churn man-agement program” to encourage them to stay. UK Retailer Tesco develops thousands of customized promotion packages it mails to its 14 million customers (Rohwedder 2006). Best Buy has identiﬁed the major segments of customers who visit its stores. It then (1) tailors its store in a particular locality to ﬁt the repre-sentation of the segments in that locality, and (2) trains its store personnel to recognize which segment a particular customer belongs to, so the cus-tomer can be serviced appropriately (Boyle 2006). 3
4 1 Introduction Catalogers routinely use “predictive models” to decide which customers should receive which catalogs. “E-tailer Z” uses “recommendation engines” to customize which productsit “cross-sells” to which customers. Dell Computer uses data analyses of prospects to improve its customer acquisition rate (Direct Marketing Association 2006). These are but a few examples of database marketing in action. The com-mon theme is that all of them are based on analyzing customer data andimplementing the results. 1. 1. 1 Deﬁning Database Marketing While the above examples provide an idea as to what database marketing isabout, it is useful to formally deﬁne the topic. The National Center for Data-base Marketing, quoted by Hughes (1996a, p. 4), deﬁnes database marketing as: Managing a computerized relational database, in real time, of comprehensive, up-to-date, relevant data on customers, inquiries, prospects and suspects, to identify our most respon-sive customers for the purpose of developing a high quality, long-standing relationship ofrepeat business by developing predictive models which enable us to send desired messagesat the right time in the right form to the right people-all with the result of pleasing ourcustomers, increasing our response rate per marketing dollar, lowering our cost per order,building our business, and increasing our proﬁts. While perhaps a bit long-winded, this deﬁnition in our view captures the essentials of database marketing-analyzing customer data to enhance cus-tomer value. A more succinct deﬁnition, which we advocate, is: Database marketing is the use of customer databases to enhance marketing productivity through more eﬀective acquisition, retention, and development of customers. Each phrase in this deﬁnition is carefully chosen. First, database marketing is fundamentally about using of customer databases. The “customer” can be either current customers or potential customers. Firms have data on theircurrent customers' purchase behavior and demographic and psychographic information, as well as the ﬁrm's previous marketing eﬀorts extended to these customers and their response to them. For potential customers-prospects-ﬁrms may be able to obtain data on customer demographics and psycho-graphics, as well as purchase history data, although obviously not in the same depth as available for their current customers. Second, database marketing is about marketing productivity. In today's results-oriented businesses, senior management often asks the simple ques-tion, “Do our marketing eﬀorts pay oﬀ?” Database marketing attempts to
1. 1 What Is Database Marketing? 5 quantify that eﬀectiveness and improve it. It does this through eﬀective tar-geting. The retail pioneer John Wannamaker is credited with saying, “I know half of my advertising doesn't work; I just don't know which half. ” Thinking more broadly, in terms of marketing rather than advertising, database mar-keting identiﬁes which half of the ﬁrm's marketing eﬀorts is wasted. It does this by learning which customers respond to marketing and which ones do not. The responsive customers are the ones who are then targeted. Third, database marketing is about managing customers. C u s t o m e r sm u s t be acquired, retained, and developed. Acquiring customers means getting anindividual who currently does not do business with the company to startdoing business with the company. Retention means ensuring the current cus-tomer keeps doing business with the company. Development means enhanc-ing the volume of business the retained customer does with the company. A key concept in database marketing that captures these three factors is “customer equity” (Blattberg et al. 2001), which we investigate in detail when we discuss “Acquisition and Retention Management” in Chapter 26. For now, the important point is to recognize that database marketing is con-cerned with all three elements of customer equity. The Dell example above involves customer acquisition. The ABC Telecom example involves customer retention. The XYZ Bank, Tesco, and E-tailer Z examples involve customer development. 1. 1. 2 Database Marketing, Direct Marketing, and Customer Relationship Management We can shed more light on the deﬁnition of database marketing by considering its close cousins, direct marketing and customer relationship management (CRM). Indeed, direct marketing and CRM overlap strongly with databasemarketing. While each of the three concepts has its own nuances, the key distinguishing characteristic of database marketing is its emphasis on the use of customer databases. Customer relationship management emphasizes enhancing customer re-lationships. That certainly is part of the deﬁnition of database marketing(acquisition, retention, and development). However, ﬁrms can enhance cus-tomer relationships without using data. The local clothing store's salesperson gets to know individual customers through their repeated visits to the store. The salesperson learns how to treat each customer and what their tastes are. This produces and enhances a relationship between the store and the customer. There is no formal analysis of databases. Essentially, the “data”are the experiences remembered by the salesperson. Database marketing can be viewed as an approach for large companies to develop relationships with customers, because there are so many customers and so many salespersons
6 1 Introduction that it is impossible for every salesperson to really know each customer. Para-doxically, the software and computer systems for compiling the data needed to implement database marketing to enhance customer relationships have been marketed as CRM software or technology. Direct marketing's emphasis is on “addressability,” the ability to interact with a customer one-to-one (Blattberg and Deighton 1991). Addressabilityis certainly a key aspect of database marketing, since targeting is the keyway that database marketing enhances marketing productivity. But direct marketing can directly address customers simply by purchasing lists that “make sense,” and sending customers on that list an oﬀer. Note again, thereis no formal data analysis in this example. Database marketing emphasizes the analysis of the data. In addition, while database marketing implemen-tations often involve direct one-to-one contacts, this need not be always thecase. In the Best Buy example above, the ﬁrst component of the applica-tion is that the analysis of customer data drives the design of the store. This is not direct marketing but it is database marketing. The second com-ponent of the application, training salespeople to recognize particular mar-ket segments as they shop in the store, is more along the lines of direct marketing. In summary, database marketing, direct marketing, and customer rela-tionship highly overlap. They diﬀer in points of emphasis-database market-ing emphasizes the analysis of customer data, direct marketing emphasizes addressability, and customer relationship management emphasizes the cus-tomer relationship. However, many people who call themselves direct mar-keters certainly analyze customer data. And many CRM applications soft-ware companies emphasize customer data. So customer data analysis is not the exclusive domain of database marketing-it's just database marketing'sspecialty. 1. 2 Why Is Database Marketing Becoming More Important? It is diﬃcult to ﬁnd statistics that document the size of the database mar-keting industry. Some suggestive numbers are: (1) The market for “CRM Software” is valued at $7. 773 billion in 2005 and expected to grow to $10. 940 billion by 2010 (Band 2006). (2) As of 2004, 100 of the top 376 companies in the Fortune 500 list of US corporations are members of the Direct Mar-keting Association, the trade association for direct marketing (Direct Mar-keting Association 2004, pp. 22-23). (3) In 2004, 39. 153 million US adultsbought products through the mail (Direct Marketing Association 2004, p. 29). (4) Business-to-business direct marketing advertising expenditures totaled $107 billion in 2003, and are expected to increase to $135 billion by 2007
1. 2 Why Is Database Marketing Becoming More Important? 7 (Direct Marketing Association 2004, p. 167). These numbers provide indica-tions of the size of the industry, but do not include budgets for marketing analytics groups that analyze the data, for campaigns that implement data-base marketing programs, or for the multitude of service ﬁrms (advertisingagencies, data compilers, and list management ﬁrms), that account for sig-niﬁcant expenditures. The indications are that the database marketing industry is huge and in-creasing. The question is, why? We hypothesize ﬁve major classes of reasons: Information technology : Companies now have the ability to store and ma-nipulate terabytes of data. While the software to do so is expensive, the capabilities are dramatic. Growth of the Internet : The Internet is a data-collection “machine. ” Many companies that previously could not collect and organize data on theircustomers can now do so through the Internet. Lower productivity of mass marketing : While there are no good statis-tics on this, there is the belief that mass advertising and non-customizedmarketing eﬀorts are eliciting poorer response, while costs are increasing and margins are declining. One can write the proﬁtability of a marketing campaign as Π=Npm-Nc,w h e r e Nis the number of customers reached by the campaign, pis the percentage that respond, mis the contribution margin when they respond, and cis the cost of contact per customer. For a campaign to be proﬁtability, we need p > c/m. Unfortunately, all three of these terms are moving in the wrong direction. Response is lower ( p), costs are higher ( c), and margins are lower ( m). Database marketing tar-gets customers for whom response is maximal, helping the proﬁt equationto remain in the black. Marketing accountability : Results-oriented senior managers are requiring all business functions to justify their existence, including marketing. Nolonger is it taken on faith that “marketing works” or “marketing is a cost of doing business. ” The demands of senior managers for proven results feed directly into database marketing's emphasis on analyzing data andmeasuring results. Increasing interest in customer relationships : Companies are more con-cerned than ever about their relationship with the customer. They seetheir products commoditizing and customer loyalty wilting away. Data-base marketing is a systematic way to improve customer relationships. Establishing a competitive advantage : Companies are always trying to de-termine what will be their source of competitive advantage. Perhaps that source lies in the data they have on their owncustomers, which allows them to service those customers better through database marketing. We will discuss the marketing productivity, customer relationship, and com-petitive advantage issues in depth in Chapter 2, because they essentially de-ﬁne the database marketing strategy of the ﬁrm.
8 1 Introduction DBM Strategy Organization Legal Environment Define Problem Situation Analysis Objectives Methodology Implement Analyze Data Compile Data Evaluate Learn Design Campaign Fig. 1. 1 The database marketing process. 1. 3 The Database Marketing Process Database marketing is implemented through a process depicted in Fig. 1. 1. The process originates in an environment characterized by the ﬁrm's over-all database marketing strategy, its organization, and legal issues (especially privacy). These factors determine the nature of problems the ﬁrm faces, and how they will be solved. The ﬁrm then needs to deﬁne the particular prob-lem it wishes to address through database marketing. This entails a situationanalysis, a statement of objectives, and an outline of the methodology that will solve the problem. For example, a ﬁrm whose DBM strategy emphasizes customer relationships may notice that it is losing too many customers. Theobjective may be to reduce the “churn rate” from 20% to 15% per year. The ﬁrm therefore decides to design a proactive churn management program (Chapter 24) with its attendant data requirements and statistical tools. Mostof the work can be done internally because the company has the organiza-tional capability in terms of information technology, marketing analytics, and campaign implementation. The company can then proceed to compile and an-alyze the data. The analysis yields a campaign design that is implemented and evaluated. There are two key feedback loops in this process. First is the learning that takes place over time. After a program is evaluated, it provides guidance onwhat types of issues can be addressed successfully by database marketing, what data are most valuable for providing insights and for predicting cus-tomer behavior, how to analyze the data, and how to translate the analysis into program design and implementation. This learning and the expertiseit breeds is one way in which database marketing can become a competi-tive advantage for the ﬁrm. The second feedback loop is that each database marketing campaign provides datafor use in future analyses to solve future problems. For example, customer response to a catalog mailing is used to update “recency”, “frequency”, and “monetary” (RFM) variables for each customer. These become part of the database and are used to develop futuretargeting strategies.
1. 3 The Database Marketing Process 9 Table 1. 1 Database marketing activities Acquiring customers Retaining and developing customers Cross-and up-selling Customer tier programs Frequency reward programs Churn management programs Coordinating acquisition, retention, and development Multichannel customer management Acquisition and retention planning Managing the marketing mix Designing communications Multiple campaign management Price targeting Communications and product personalization Sales force management Table 1. 1 provides a list of database marketing activities-essentially, a list of the marketing problems addressed by database marketing. These in-clude acquiring customers, retaining and developing customers, coordinatingacquisition, retention, and development, and managing the marketing mix. Several of the sub-issues within each of these merit their own chapter in this book. For example, we will devote full chapters to cross-and up-selling, mul-tichannel customer management, etc. These are all very challenging problems and much work has been done on using database marketing to manage themmore eﬀectively. Because of the focus on analyzing customer data, several data analysis techniques have emerged and been applied by database marketers. Table 1. 2lists these techniques. The two most basic analyses are lifetime value of the customer and predictive modeling. Lifetime value of the customer is the net present value of the incremental revenues and costs generated by an acquiredcustomer. The reason LTV is so important is that it includes the long-term
10 1 Introduction Table 1. 2 Database marketing analysis techniques Lifetime value of the customer (LTV) Predictive modeling Statistical techniques Logistic regression Tobit models Hazard models RFM analysis Market basket analysis Collaborative filtering Cluster analysis Decision trees Neural networks Machine learning algorithms Field tests retention and development aspects of managing the customer. We devote three chapters to calculating and applying LTV. Predictive modeling is themost common form of analysis conducted by database marketers. It pertains to the use of statistical analysis to predict future customer behavior-will the customer churn, will the customer buy from this catalog, will the customerbecome more loyal if routed to the top-tier call center, will the customer be receptive to this recommended product? Predictive modeling is itself a process, and we devote a chapter to studying this process. For the statistically oriented individual, “your ship has come in” when it comes to database marketing. Table1. 2 shows the multitude of methodsused by database marketers. The reason why so many techniques have foundapplication is partly due to the variety of problems to be addressed-e. g., collaborative ﬁltering and market-basket analysis can be readily applied to cross-selling, hazard models are useful for predicting how long the customer will remain a customer; logistic regression, decision trees, and neural networks are all useful for predicting “0-1” behavior such as, will the customer respond,or will the customer churn? However, in addition to the variety of problems stimulating the variety of techniques, the other reason for the plethora of statistical techniques that areapplied by database marketers is the frantic race to achieve higher predictive accuracy. As we will see several times in this book, even a nominal increase in predictive accuracy can mean $100,000s in added proﬁts for a single cam-paign. Each bit of information we can squeeze out of the data can be directly
1. 3 The Database Marketing Process 11 Table 1. 3 Organization of the book Part 1: Strategic Issues Part 2: Customer Lifetime Value (LTV) Chapter 5: Customer Lifetime Value-Fundamentals Chapter 6: Issues in Computing Customer Lifetime Value Chapter 7: Customer Lifetime Value Applications Part 3: Database Marketing Tools: The Basics Chapter 8: Sources of Data Chapter 9: Test Design and Analysis Chapter 10: The Predictive Modeling Process Part 4: Database Marketing Tools: Statistical Techniques Chapter 11: Statistical Issues in Predictive Modeling Chapter 12: RFM Analysis Chapter 13: Market Basket Analysis Chapter 14: Collaborative Filtering Chapter 15: Discrete Dependent Variable and Duration Models Chapter 16: Cluster Analysis Chapter 17: Decision Trees Chapter 18: Artificial Neural Networks Chapter 19: Machine Learning Part 5: Customer Management Chapter 20: Acquiring Customers Chapter 21: Cross-Selling and Up-Selling Chapter 22: Frequency Reward Programs Chapter 23: Customer Tier Programs Chapter 24: Churn Management Chapter 25: Multichannel Customer Management Chapter 26: Acquisition and Retention Management Part 6: Managing the Marketing Mix Chapter 27: Designing Database Marketing Communications Chapter 28: Multiple Campaign Management Chapter 29: Pricing Chapter 1: Introduction Chapter 2: Why Database Marketing? Chapter 3: Organizing for Database Marketing Chapter 4: Customer Privacy and Database Marketing linked to marketing proﬁtability and eﬃciency. For example, if a predictive model can increase response to a direct mail oﬀer from 1% to 2%, this can literally make the diﬀerence between a huge loss and a huge gain. The reasonis that while the percentage change is small, it is multiplied by 100,000s of customers, if not millions. In this way, the beneﬁts of marginal increases in predictive accuracy add up, and we have a cornucopia of statistical techniquesthat compete for the title, “most accurate. ”
12 1 Introduction 1. 4 Organization of the Book We have organized the book according to Table 1. 3. Part I deals with the issues that shape the database marketing process-ﬁrm strategy, ﬁrm orga-nization, and the legal environment. Chapter 2, “Why Database Marketing”,relates to the ﬁrm's database marketing strategy, positing three fundamen-tal reasons why companies might want to engage in database marketing: improving marketing productivity, improving customer relationships, or es-tablishing competitive advantage. As discussed earlier, which of these reasons is the impetus for database marketing at a particular ﬁrm will inﬂuence the rest of the DBM process-which problems the ﬁrm attempts to solve, andhow it tries to solve them. Chapter 3 deals with how to organize the ﬁrm's marketing function in order to implement database marketing. Chapter 4 represents the legal environment, in particular, the issue of customer privacy. This certainly determines the types of database marketing eﬀorts the ﬁrm can undertake. Parts II-IV of the book deal with database marketing tools-how to collect the data and do the analysis. Chapters 5-7 focus on the key concept of life-time value of the customer (LTV). Chapters 8-10 focus on the basic tasks ofcompiling data, ﬁeld testing, and predictive modeling. Chapters 11-19 cover the statistical methods used primarily in predictive modeling. Parts V and VI focus on speciﬁc problems addressed by database mar-keting. They largely draw on the tools described in Parts II-IV. Part V cov-ers customer management activities including Acquiring Customers (Chap-ter 20), Cross-and Up-selling (Chapter 21), Frequency Reward Programs(Chapter 22), Customer Tier Programs (Chapter 23), Churn management (Chapter 24), Multichannel Customer Management (Chapter 25), and Ac-quisition and Retention Management (Chapter 26). Part VI focuses on themarketing mix, particularly communications (Chapters 27 and 28) and Pric-ing (Chapter 29). The result is intended to be a comprehensive treatment of the ﬁeld of database marketing, including strategic issues, tools, and problem-solving.
Chapter 2 Why Database Marketing? Abstract A basic yet crucial question is: why should the ﬁrm engage in database marketing? We discuss three fundamental motivations: enhancing marketing productivity, creating and enhancing customer relationships, and creating sustainable competitive advantage. We review the theoretical and empirical evidence in support of each of these motivations. Marketing pro-ductivity has the best support; there is some evidence for both customerrelationships and competitive advantage as well, but further work is needed. Perhaps the most fundamental question we can ask about any marketing activity is what is its raison d'etre-what purpose does it serve in enhancing ﬁrm performance? In this chapter, we propose and evaluate three reasons fordatabase marketing: Enhancing marketing productivity Enabling the development of a customer/ﬁrm relationship Creating a sustainable competitive advantage 2. 1 Enhancing Marketing Productivity 2. 1. 1 The Basic Argument The pioneering retail entrepreneur, John Wannamaker, is said to have lamented about the ineﬃciency of his marketing eﬀorts, “I know that half ofmy marketing is wasted; my problem is that I just don't know which half. ” The promise of database marketing is to identify which marketing eﬀorts are wasted and which are productive, thereby allowing the ﬁrm to focus on theeﬀorts that are productive. Database marketing does this by identifying cus-tomers for whom the marketing eﬀort will pay oﬀ, and then targeting those customers. In this view, database marketing is fundamentally a segmentationand targeting tool for enhancing marketing productivity. 13
14 2 Why Database Marketing? Table 2. 1 The economics of database marketing: A prospecting example Untargeted Mailing Number of oﬀers mailed: 1,000,000 Proﬁt contribution per response: $80 Cost per mailing: $0. 70 Response rate: 1% Proﬁt = 1,000,000 ×0. 01×$80-1,000,000 ×$0. 70 =$800,000-$700,000 =$100,000 Targeted mailing Decile Number Response Proﬁt ( $) Cumulative of prospects rate (%) Proﬁt ( $) 1 100,000 3. 00% 170,000 170,0002 100,000 2. 00 90,000 260,0003 100,000 1. 40 42,000 302,0004 100,000 1. 15 22,000 324,0005 100,000 1. 00 10,000 334,000 6 100,000 0. 60-22,000 312,000 7 100,000 0. 40-38,000 274,000 8 100,000 0. 30-46,000 228,000 9 100,000 0. 10-62,000 166,000 10 100,000 0. 05-66,000 100,000 Total 1,000,000 1. 00% $100,000 =>Target ﬁrst ﬁve deciles (Proﬁt = $334,000) The power of this argument can be seen in the example shown in Table 2. 1. The example depicts the economics of a direct marketing campaign whose goal is to proﬁtably sell a new product to a list of 1,000,000 potential“prospects. ” Each prospect who “responds” to the oﬀer generates $80 in proﬁt. The cost to extend the oﬀer is $0. 70, including costs of mailing and printing of the mail piece. Assuming a 1% response rate-fairly typically for a large-scale mailing-proﬁt would be: Proﬁt = 1,000,000 ×1% response ×$80/response-1,000,000 ×$0. 70/contact = $100,000 The mailing is proﬁtable. However, the above calculation illustrates Wanna-maker's perspective taken to an extreme-99% of the marketing expenditures were wasted! Only 10,000 will respond to the oﬀer, yet we are mailing to1,000,000 customers to ﬁnd those responders. This unfortunately is a typicaloutcome for many marketing expenditures. The cost is not only lost proﬁts to the ﬁrm, but wasted “junk mail” and advertising clutter as well. If we could eliminate some of that waste, proﬁts could be increased and perhapssociety itself could be better served. The lower portion of Table 2. 1 shows how the results can be improved with database marketing. The prospect list is segmented into deciles, 100,000in each decile, prioritized by their likelihood of responding to the oﬀer.
2. 1 Enhancing Marketing Productivity 15 The prioritization is determined by a process called predictive modeling (Chapter 10). Predictive modeling identiﬁes a top decile of customers who have a response rate of 3%. The second decile has a response rate of 2%, etc., down to the 10th decile, which has a response rate of 0. 05%. The proﬁts fromtargeting the ﬁrst decile would be 100,000 ×3% response ×$80/response-100,000 ×$0. 70/contact = $170,000. Targeting this decile alone would yield more proﬁt than targeting the entire list. The key is that we are saving onthe mailing costs-“only” 97%, not 99%, of the mail costs are wasted in this segment. Going through the calculations for each decile, we see that it would be proﬁtable to target the top 5 deciles, yielding a cumulative proﬁt of $334,000, much higher than the $100,000 gained by targeting the full list. Database marketing allows ﬁrms to segment their customers according to “lift tables” such as in Table 2. 1, and then deliver the marketing eﬀort tothe customers whom the analysis predicts will be proﬁtable. The key to the proﬁt improvement is that the top deciles have substantially higher responserates than the lower deciles. The ratio of response rate in a decile to the average response rate is known as “lift. ” Note that a ﬁrst-decile lift of 3 to 1 (3% response for that decile divided by 1% for the entire database) is enough to enhance proﬁts signiﬁcantly. The lift for the top 5 deciles is 1. 71%/1% = 1. 71. Lift levels of this magnitude are quite feasible given current statistical technology. This provides a fundamental reason for ﬁrms to employ databasemarketing-it increases the proﬁts generated by marketing campaigns by targeting customers more eﬀectively. 2. 1. 2 The Marketing Productivity Argument in Depth The marketing productivity argument for database marketing follows fromthe recognition of three major forces: (a) a major problem of mass market-ing (e. g., traditional electronic media such as television) is lack of targeting and database marketing provides the ability to target, (b) marketing needs to be accountable and database marketing provides accountability, and (c)mass marketing eﬀorts are diﬃcult to assess and adjust, whereas database marketing provides a process for learning how to target more eﬀectively. 2. 1. 2. 1 Database Marketing as a Solution to Targeting Ineﬃciencies of Mass Marketing Beginning with Wannamaker's observation that half his advertising was wasted, marketers have long lamented their inability to target eﬀorts ef-fectively. For example, mass media advertising can be targeted only to alimited degree. Market research services identify demographic characteristics
16 2 Why Database Marketing? and product preferences associated with particular television shows, or geo-graphic regions, but this produces nowhere near the desired level of individual targetability. Blattberg and Deighton (1991) pioneered the notion that data technol-ogy can improve targeting in their concept of the “addressable consumer. ”Their main point was that database marketing could create a dialogue be-tween the customer and the company, whereby the company would learn theresponses of individual customers and respond to their needs. This was a radical departure from mass media. Deighton et al. (1994) elaborated on this theme: “At its most sophisticated, then, a transaction database is a recordof the conversation between a ﬁrm and each[italics added] of its customers, in which the ﬁrm's oﬀering evolves as the dialogue unfolds” (p. 60). Coincident with the conceptual argument that data technology could im-prove targeting was the practical observation that the costs of maintaining and storing databases had decreased rapidly. Blattberg and Deighton (1991) maintained that “the cost of holding a consumer's name, address, and pur-chase history on line has fallen by a factor of a thousand since 1970 and is continuing to fall at this rate. ” Sheth and Sisodia (1995b) report that “Computing power that used to cost a million dollars can be had today for less than a dollar. ” Peppers and Rogers (1993, pp. 13-14) echo similar themes. Second was the observation that the tools for extracting the necessary learning from the data (to construct the lift table in Table 2. 1) were availableand getting better. This led to an explosive growth in “data mining” (e. g.,Peacock 1998). Peacock deﬁnes data mining as “the automated discovery of 'interesting,' nonobvious patterns hidden in a database that have a high potential for contributing to the bottom line... 'interesting' relationships are those that could have an impact on strategy or tactics and ultimately on an organization's objectives. ” He cites a few examples: Marriott's Vacation Club used data mining to cut the level of direct mail needed to accomplish a desired response level. This is a prime illustration of Table 2. 1. Prudential Insurance tested the results of data mining for improved re-sponse rates among prospects, and found them to be doubled. American Express used data mining to “score” customers in terms of howlikely they were to purchase various items. It then used these data togenerate oﬀers that match the products of its partners with the needs of its customers. In summary, the recognition that targeting was the problem with mass mar-keting, that database marketing could theoretically improve targeting, that database costs were declining, and that data mining was eﬀective in prac-tice at developing the targeting plans, contributed mightily to the growth indatabase marketing as a tool for improving marketing productivity.
2. 1 Enhancing Marketing Productivity 17 2. 1. 2. 2 Marketing Accountability and the ROI Perspective Emerging from the period of high inﬂation in the 1970s, senior manage-ment became very concerned with costs-production, labor, and materials. Webster (1981) (see also Lodish 1986) reported that by the early 1980s, CEO's had begun to focus on marketing. The fact that it was general man-agers-the CEO's-who were calling attention to marketing meant twothings. First, the issue was broader than costs. It was productivity in the sense of Return on Investment (ROI), i. e., how much proﬁt was being generated per marketing dollar. Second, marketing needed to be accountable, so thatmarketing productivity needed to be measured. Sheth and Sisodia (1995a) report that by the mid-1990s, “CEO's are demanding major cost savings and a higher level of accountability from marketing than ever before. ” As illustrated in Table 2. 1, database marketing fulﬁlls the need to measure ROI. Rather than spending $700,000 to produce a proﬁt of $100,000 (an “ROI” of 15%), database marketing would spend $350,000 to produce a proﬁt of$334,000 (an ROI of 95%). 1Expenditures have decreased and proﬁts have increased. The key however is that the results are measurable. The entiredatabase marketing mentality is based on measuring results. In Table 2. 1, it is relatively simple since response can be measured and tabulated, and the costs can be calculated. Costs, at least direct costs, are almost always easy to measure in a direct marketing context. Incremental revenues are sometimes diﬃcult to measure,however, because it is not clear what response would have been withoutthe marketing campaign. This is where the role of experimentation and learning comes in. For example, assume that in Table 2. 1, it was possible that consumers could buy the product even without a direct mail campaign,e. g., through a diﬀerent sales channel. The database marketer would then design an experiment by creating a control groups. Rather than mailing to all 100,000 prospects in Decile 1, he or she would mail to just 90,000, holding10,000 aside as controls. The incremental gain from the campaign could then be calculated as the response rate for the 90,000 minus the “response” rate for the 10,000. The ease of conducting experiments plays a key rolein measuring the results of database marketing, hence in making database marketing accountable. While marketing ROI is naturally measured as proﬁt generated per incre-mental expenditure divided by the investment, there are many other ways tomeasure it. Sheth and Sisodia (1995a, b) propose that marketing productivity be measured as a weighted average of customer acquisition productivity and customer retention productivity. Customer acquisition productivity would 1Note it is not clear that ﬁrms should maximize ROI rather than the absolute level of proﬁts. ROI may be maximized at a lower level of expenditure than would maximize proﬁts (see Table 2. 1, where targeting just the ﬁrst decile would maximize ROI, whiletargeting the ﬁrst 5 deciles will maximize proﬁts). (The authors thank Preyas Desai forthese insights. )
18 2 Why Database Marketing? consist of revenues generated by new customers divided by expenditures on acquiring new customers, “adjusted by a customer satisfaction index” (p. 11). The adjustment serves to quantify the long-run beneﬁts of this acquisition. Customer retention productivity would consist of revenues from existing cus-tomers divided by expenditures for serving existing customers, adjusted by a “customer loyalty index,” again to bring in the long-term value of the invest-ment. There are several practical issues in constructing these measures, butthe emphasis on acquisition and retention plays to the very deﬁnition of data-base marketing (the use of customer databases to increase the eﬀectiveness of marketing in acquiring and retaining customers). We add that cross-sellingor up-selling customers is also very important. Once a ﬁrm has a customer, the ability to sell additional products through database marketing provides the ﬁrm a signiﬁcant advantage (Blattberg et al. 2001). 2. 1. 2. 3 Database Marketing as a Learning System Mass marketing eﬀorts are diﬃcult to assess and adjust. While marketing mix modeling has become very popular and generates useful results, a key limita-tion is the diﬃculty and cost in setting up controlled experiments. Database marketing is a learning marketing system because ﬁrms use both experimen-tation and data mining techniques to learn about the eﬀectiveness of their marketing mix decisions and about their customers' behavior, and then ad-justs these decisions accordingly. Experimentation is fundamental to databasemarketing. In its extreme database marketers test micro tactical decisions such as the color of the paper used in a direct marketing campaign or the greeting used in telemarketing. While very tactical, experimentation meansthat database marketers can learn from their “mistakes”-unsuccessful copy, pricing or oﬀers-and can improve the eﬀectiveness and eﬃciency of their marketing activities. Traditional mass marketers in theory can set up experiments but they are prone to small sample sizes, diﬃculty creating controls and high costs. Tools such as IRI's Behavior Scan can be used in the consumer packagedgoods industry to test advertising. However, for most products that can not be tracked with consumer panels, this option does not exist. Hence, database marketing has a signiﬁcant advantage to ﬁrms because of the ability of the ﬁrm to experiment, learn, and adjust. Database marketers such as Amazon now use more sophisticated targeting tools to learn about their customers' behavior and then use this to cross-sellother products. One technique used to analyze customer behavior and make product recommendations is called collaborative ﬁltering (Chapter 14). Thisand similar techniques use purchase histories and other information to deter-mine the likelihood a customer will purchase a related product. For example, Amazon uses a customer's book purchase history to make a recommendationof books the customer might be interested in purchasing.
2. 1 Enhancing Marketing Productivity 19 Customer Acquisition Customer Interactions Customer Response (purchase, usage patterns, etc. ) Information Information Strategy Refinement Product Communications Price Promotion Fig. 2. 1 The learning marketing system fostered by database marketing. We call the process of implementing database marketing campaigns, learn-ing, and adjusting a “learning marketing system”. This system is depicted in Fig. 2. 1. The ﬁgure shows that the ﬁrm uses information it collects in the process of acquiring and retaining customers to update its strategyfor interacting with customers. This entails the product oﬀering, commu-nications, price, and promotion. The ﬁrm is able to target these elements more eﬀectively because it has learned about consumer preferences andresponsiveness. A learning marketing system can also provide a competitive advantage to a ﬁrm because, if carefully crafted, it can provide better product recommenda-tions and more targeted communications to the customer than if the customer switches companies and makes his or her ﬁrst purchase from a competitor. That competitor does not have the information available to customize prod-uct recommendations and communications. Amazon should therefore have a signiﬁcant advantage relative to Barnes & Noble and Borders because it hasbeen tracking customer purchase much longer and oﬀering recommendations throughout the customer's purchase experience with Amazon. 2. 1. 3 Evidence for the Marketing Productivity Argument Table 2. 1 suggests two crucial components to the marketing productivity ar-gument for database marketing. First is that predictive modeling generates lift tables that separate customers who will respond from those who willnot. Second is that these tables actually predict what will happen once the marketing campaign is launched. There are several examples to demonstrate the feasibility of lift tables (“charts” when shown graphically). Figure2. 2 is from Ansari and Mela (2003)
20 2 Why Database Marketing? 0%20%40%60%80% 1 23 456789 1 0 Decile% Who Click Through Fig. 2. 2 Lift chart for an e-mail campaign (From Ansari and Mela 2003). on targeted e-mail. The goal was to use e-mail to generate visits to an information-oriented website. As the ﬁgure shows, the average response rate was 20%. However, the authors were able to separate customers into decilessuch that customers in the ﬁrst 3 deciles had a response rate of 40%, a 2 to 1 lift. See Sheppard (1999) and Chapter 10 for a detailed discussion of lift tables. Figure2. 3 shows a lift chart for predicting which credit card customers will close their accounts (i. e., “churn”). Predictions are based on customerbehavior over the previous six months. As the chart shows, those in the top decile have a 7% chance of churning, compared to an average of lessthan 1% over the entire customer base. The top decile customers could be targeted with a customer retention program-perhaps a new oﬀer, or simply a reminder of the favorable features of their credit card. Figure2. 4 shows the predicted “next-product-to-buy” adoption of web banking for a retail bank (Knott et al. 2002). The most important variablefor making these predictions was products currently owned by customers. 0%2%4%6%8% 123 45 6 78 9 1 0 Decile% Churn Fig. 2. 3 Lift chart for predicting credit card customer attrition (Courtesy of ASA, Pitts-burgh, PA, Model Max Demonstration Data).
2. 1 Enhancing Marketing Productivity 21 0%2%4%6%8% 1234 56789 1 0Decile% Buy Fig. 2. 4 Lift chart for predicting adoption of web banking using a next-product-to-buy (NPTB) model (From Knott et al. 2002). The average adoption rate is 2. 3%; the adoption rate in the top 3 deciles is 5%. These customers appear to be good prospects for a web banking direct mail piece. In all three examples, the database marketer uses predictive models to separate customers in segments (deciles) in prioritized order of their partakingin some behavior-be it response to an e-mail, giving up a credit card, or adopting a new product. Diﬀerent actions are called for depending on thedecile in which a given customer falls. These results are impressive and show customers can be segmented using predictive models. Note this is not the traditional form of segmentation usedin marketing text books. It is segmentation based on the likelihood of buying determined from statistical models. A critical question is: does targeting im-plied by lift charts actually result in higher revenues and proﬁts? Figure2. 5 $0$25$50$75$100 NPTB Mail NPTB Control Heuristic Mail Heuristic Control Revenues per Customer Fig. 2. 5 Revenues from ﬁeld-tested cross-selling campaign (From Knott et al. 2002).
22 2 Why Database Marketing? shows one example from Knott et al. (2002). A predictive model was used to prioritize customers according to their likelihood of purchasing a home eq-uity loan. The top prospects were then targeted with a direct mail campaign. Note that this tests the ability of the predictive model predictions to hold upwhen the targeting actually occurs, subsequently to the modeling. Figure2. 5 shows the targeted mailing generated revenues of $93 per mailed-to customer. However, customers could have obtained a loan throughother means, for example, simply by walking up to the bank and applying. Did the mailing generated incremental revenue above what would have been obtained through the usual marketing channel? To answer this, the authorsin advance set up a control group consisting of customers who were predicted by the model to be top prospects, but were randomly selected not to receive the direct mail piece. It turned out that some of these people did obtainloans on their own, but revenues for this group were only $37 per customer. Finally, the question arises as to whether the model-based on a neuralnet-worked better than a simple heuristic. In this case, the heuristic wasto target direct mail pieces for the loan to wealthier customers. As Fig. 2. 5 shows, this heuristic barely produced any additional revenues compared to its control group. Knott et al. (2002) suggest three key ﬁndings. First, targeted campaigns based on predictive models produce higher revenues. Second, the revenuesare incremental over what would have been achieved through existing mar-keting eﬀorts. Third, the model outperforms a reasonable but non-statistical heuristic. Overall, we see measurable improved performance from targeting. 2 That is one of the promises of database marketing. The above examples suggest that statistical methods can create beneﬁcial targeting eﬀorts. One consideration is costs. As we saw earlier, the costsinclude: compilation of a database, the lift chart capabilities generated by a given investment, and average contact expenses with and without database marketing. Industries that naturally maintain customer databases, such asservices and catalogs, obviously will ﬁnd the database costs less expensive. 2. 1. 4 Assessment The argument that database marketing's raison d'etre is to improve market-ing productivity is compelling. It is based on (1) the recognition that eﬀec-tive targeting is crucial and that database marketing can deliver it, (2) that modern marketers are accountable and that database marketing can mea-sure ROI, and (3) that learning and reﬁnement is key to eﬀective marketingand database marketing is indeed a learning process. These forces should 2The illustration in Fig. 2. 5 is in terms of revenues, but Knott et al. (2002) show that proﬁts increase as well.
2. 2 Creating and Enhancing Customer Relationships 23 continue into the future. In addition, the targeting and ROI components of the argument have received direct empirical support. While the marketing productivity argument is indeed powerful and undoubtedly has contributed to the growth of database marketing, the pro-ductivity argument is largely tactical. It focuses on the proﬁtability of indi-vidual marketing campaigns. It leaves out two fundamental issues, developing customer relationships and establishing a competitive advantage. These twoissues will be the focus of the next two sections of this chapter. 2. 2 Creating and Enhancing Customer Relationships 2. 2. 1 The Basic Argument The argument is that (1) strong customer relationships are good because they go hand-in-hand with brand loyalty, and (2) database marketing can be used to create and enhance customer relationships. 2. 2. 2 Customer Relationships and the Role of Database Marketing 2. 2. 2. 1 The Emergence of Customer Relationships as an Area of Marketing Focus Among the ﬁrst researchers to articulate the CRM argument for databasemarketing was Berry (1983). Berry urged marketers to be “thinking of mar-keting in terms of having customers, not merely acquiring customers,” (p. 25),and deﬁned relationship marketing as “attracting, maintaining, and enhanc-ing customer relationships in multi-service organizations. ” The importance of customer relationships was echoed by Webster (1992, p. 1): “Customerrelationships will be seen as the key strategic resource of the business. ” Berry outlined a number of relationship marketing strategies, including “customizing the relationship”, which was an especially attractive strategy when “personal service capabilities are combined with electronic data process-ing capabilities. ” He describes examples at Xerox, American Express, andother companies where service capabilities were enhanced by customer data records. The key notion was that a customer service representative could cul-tivate a stronger relationship with the customer by having instant access tothe customer's data ﬁle. Berry's emphasis on relationships stemmed from the idea of enhancing cus-tomer service. Webster's emphasis on relationships stemmed from a desire tomove the deﬁnition of marketing toward one based on social and economic
24 2 Why Database Marketing? processes rather than functional tasks(the 4 P's). More recently, the motiva-tion for emphasizing customer relationships stems from the simple economics of lifetime value. The lifetime proﬁts or “customer equity” delivered by a set of Ncustomers can be written as (see Chapter 5): Proﬁts =N∞∑ t=0(R-c-m)rt (1 +δ)t-Na (2. 1) where: N= Number of customers acquired. a= Acquisition cost per customer. R= Revenues per period per customer. c= COGS per period per customer. m= Ongoing marketing costs per period per customer. δ= Discount rate. r= Retention rate, i. e., the percentage of customers who are retained year to year. Equation 2. 1 can be re-written as: Proﬁts =N(R-c)((1 +δ) (1 +δ-r))-Nm((1 +δ) (1 +δ-r))-Na (2. 2) where the ﬁrst term is long-term proﬁt contribution, the second term is long-term retention costs of marketing, and the third term is total acquisition costs. The emphasis on customer relationships is consistent with the factthat Equation 2. 2 is a convex function of retention rate as opposed to a linear function of the number of acquired customers. The convexity of long-term proﬁts with respect to retention rate can be seen in Fig. 2. 6. The implication is that an increase in retention rate by 20%increases proﬁts more than increasing the number of customers ( N) by 20%. The beneﬁts of customer retention have been reinforced by several re-searchers. Winer (2001) reports a Mc Kinsey study that investigated howacquisition versus retention aﬀects the market value of Internet ﬁrms. The study concluded that retention was far more powerful than acquisition. Reichheld (1996) found that small increases in retention have dramatic im-pact on total proﬁts. Gupta et al. (2004a) reached similar conclusions. A relationship management strategy is partly predicated on the belief that: (a) retaining customers is less expensive than acquiring new customers and (b) increasing retention is more valuable than increasing acquisition. Theabove discussion suggests a solid foundation for the revenue side. Unfortu-nately there is not a solid foundation for the cost side. It may be far more costly, or impossible, to increase retention rates from 80% to 90% than it isto increase acquisition rates from 1% to 5%. Generalizations about the costs of increasing intention rates have not been well documented. This is an em-pirical question and may be ﬁrm speciﬁc. See Chapter 26 for more discussionof acquisition versus retention strategies.
2. 2 Creating and Enhancing Customer Relationships 25 $0$5,000,000$10,000,000$15,000,000$20,000,000$25,000,000$30,000,000 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 Retention Rate Long-Term Total Profits a= $30 acquisition cost per customer N= 100,000 number of customers m= 10 ongoing marketing cost per customer M= 100 revenues per customer c= 50 COGS per customer d= 0. 15 discount rate Fig. 2. 6 Relationship between customer retention rate and total proﬁts per customer (Equation 2. 2). Another impetus for the importance of relationships was research in the 1990s that showed a linkage from relationship strength to customer satisfac-tion to loyalty to ﬁrm performance. Several studies have investigated all orpart of the satisfaction-loyalty-performance linkage. Anderson et al. (1994) used a three-equation model to describe the evolution of customer expecta-tions, satisfaction, and return on assets. Their analysis was at the companylevel-77 Swedish ﬁrms across a wide variety of industries. The critical ﬁnd-ing was a strong link between satisfaction and return-on-investment (ROA). They did not investigate brand loyalty per se but did hypothesize that onereason for a link between satisfaction and ROA is higher loyalty. Rust and Zahorik (1993) model this more formally. They present a broad model that captures the relationship between satisfaction, retention, and mar-ket share. While they do not estimate the entire model, they provide an ex-ample where they predict retention likelihood for a retail bank as a functionof satisfaction factors. The most important satisfaction factor inﬂuencing re-tention is “Warmth,” which includes elements such as “friendliness,” “how well the manager knows me,” “listens to my needs,” as well as “convenienceto home. ” Most of these elements are basically indicators of the strength of the customer relationship. Barnes (2000) studied 400 customers' relationships with companies from a variety of industries, including ﬁnancial institutions, grocery stores, and
26 2 Why Database Marketing? telecommunications. He measured the closeness, strength, and emotional tone of the relationship, and found that closeness correlated strongly with satisfaction. Bolton (1998) studied the eﬀect of satisfaction on the length of the re-lationship. She found that prior cumulative satisfaction directly aﬀects thelength of the duration of the relationship. She also shows that the eﬀect of transaction or service failures on duration times depends upon prior satis-faction. Her results show a direct relationship between customer satisfaction and the lifetime value of a customer. Together, the above papers trace a relationship from the customer relation-ship to customer satisfaction to loyalty/retention to higher ﬁrm performance. They cement the argument that relationships are important because they in-crease retention, and retention is an attractive way to build ﬁrm performance. 2. 2. 2. 2 The Role of Database Marketing in Establishing Customer Relationships The previous discussion established the importance of customer relationships. What is needed next is to establish that database marketing is a way to establish relationships. Fournier (1998) presented the conceptual foundation for customer-brand relationships, and provided exploratory evidence that relationships are a validbehavioral construct. Her conceptual foundation, based on the work of Hinde(1995), was that a relationship involves four aspects: (1) reciprocal exchange between the partners in the relationship, (2) purpose in that relationships satisfy goals of both participants, (3) multiplex in that they take on manydiﬀerent forms, and (4) a process, in that they evolve over time. All four dimensions map to the capabilities of database marketing. The reciprocal exchange is that customers give ﬁrms data and ﬁrms give customersbetter products and service. The goals to be satisﬁed are proﬁts for the ﬁrm and overall utility for the customer. The multiplex nature of relationships suggests that there must be several “touch points” where customers and ﬁrmsinteract, and all must be managed. Database marketing has the capability to manage these touch points. But the strongest ﬁt between database marketing and relationships involves the notion that relationships are processes that evolve over time. The nature of database marketing is to collect data, take action, evaluate the action, collect more data, take more actions, etc. Thecustomer data ﬁle and the conclusions one can draw from it evolve over time, as relationships should by their very nature. Fournier's work suggests that database marketing and relationships bond at a conceptual level. Peppers and Rogers (1993, 1997) articulated that bondfrom a managerial perspective. Peppers and Rogers (1993) emphasized the importance of building relationships with a one-to-one mentality. They dis-cussed critical relationship concepts such as “share of customer”, “customer
2. 2 Creating and Enhancing Customer Relationships 27 Trial Repeat Purchasing Share of Requirements Affinity Brand Recommendation Loyalty Brand Relationship Service Recommendation Fig. 2. 7 Brand relationship management model. driven,” and “lifetime value. ” Peppers and Rogers (1997) emphasize that the way to manage these relationship concepts is through data. They state (p. 11), “the computer is now changing the actual character of the compet-itive model itself, supplanting it with a customer-driven model. ” The mar-keting mantra is now, “I know you. You tell me what you want. I make it. I remember next time. ” Nebel and Blattberg (1999) developed the concept of Brand Relation-ship Management deﬁned by them as, “An integrated eﬀort to establish,maintain, and enhance relationships between a brand and its consumers,and to continuously strengthen these relationships through interactive, individualized and value-added contacts, and a mutual exchange and fulﬁllment of promises over a long period of time. ” Rather than concentrateon share of requirements (market share among the brand's customers) as the ultimate goal, they suggest that the end-state of brand relationship management is loyalty developed through aﬃnity and the creation of a brandrelationship. An example is Apple Computer who has created numerous customer interactions through their IPOD and Itunes and Apple Stores. These help build a brand relationship rather than simply a brand. The goalof a strong brand relationship is loyalty and recommendation of the product or service. Their framework is shown in Fig. 2. 7. Another example of brand relationship management is P&G's mother helpline. For a brand like Pampers (diaper) P&G provides an interactive helpline and website to answer mother's questions. Even if these questionsare not directly related to diapers, this creates interactive, individualized, value-added contacts and hence a stronger brand relationship. The issues for academics are: (a) do these contacts strengthen brand loyalty and (b) doesthe enhanced loyalty create a brand relationship that leads to greater lifetime value. Winer (2001) further strengthened the link between customer relation-ships and database marketing with his “framework for customer relationship
28 2 Why Database Marketing? management. ” This is a framework for implementing customer relationship management. The framework inter-twines data, data analysis, and relation-ship building, and consists of the following steps: 1. Creating a customer database 2. Analyzing the data 3. Selecting customers to target 4. Targeting customers with the appropriate vehicle5. Developing relationship programs-reward programs, customized product, customer service, community building 6. Privacy considerations7. Developing metrics to evaluate the process Steps 5-6 involve the harnessing of database marketing speciﬁcally to developrelationships. 2. 2. 3 Evidence for the Argument that Database Marketing Enhances Customer Relationships The evidence that database marketing exists to build customer relationshipsis not very strong. The issue is clouded by the faddish nature of “CRM” asthe latest answer to company problems. CRM has indeed received less than favorable reviews from the business press. Most of this comes from company surveys of manager satisfaction with CRM initiatives. Insight Technology reported that 31% of companies believed they obtained no return on CRM, 38% got minor gains (Anonymous 2001). Gartner Group reported that 45% of CRM projects fail to improve customer interactions and 51% generate no positive returns in 3 years (Anonymous 2001). Meta Group reports that 75% of CRM initiatives fail to meet objectives (Anonymous 2001). “It is estimated that 60-80% of CRM projects do not achieve their goals, and 30-50% fail outright” (Sheth and Sisodia 2001). Mercer Management Consulting found that only 38% of companies arerealizing expected returns from CRM tools, 26% are realizing expectedreturns from customer proﬁtability tools (Jusko 2001). These surveys do not pinpoint the source of the disappointment with CRMinitiatives. It is not clear whether CRM initiatives have failed, or whether theyare serving as a scapegoat for poor technological investments. In addition, the examples are from a particular time period-the “dot-com boom”-when companies had been over-investing in information technology. In any case, there are several possibilities as to why these initial eﬀorts to integrate database marketing and customer relationship management may not havebeen successful:
2. 2 Creating and Enhancing Customer Relationships 29 Organizational Barriers : Database marketing-based CRM requires orga-nizational coordination. Companies have not been able to achieve this. Marketing quarrels with sales over who owns the customer (Boehm 2001). Marketing and ﬁnance quarrel about how deeply to go down the priori-tized customer ﬁle in investing in the relationship. Marketing and IT do not share insights from the data mining (Gillet 1999). Management re-ward structures are too short term to encourage cultivating the customer. Peppers and Rogers (1997) argue that organization structures and com-pensation schemes must adapt to the customer-centric revolution. Srini-vasan and Moorman (2002) show that a customer-focused reward systemand strong interactions between marketing and IT encourage appropriate investment decisions that in turn breed customer satisfaction and better corporate performance. Reinartz et al. (2004) show that rewarding employ-ees for cultivating relationships enhances the eﬀectiveness of CRM eﬀorts. Acquisition versus Retention Costs : While part of the attraction of CRM is the view that it is cheaper to increase retention than to increaseacquisition, it this assumption may be incorrect. For example, part of the CRM strategy is to develop a “single view of the customer”. This enables the ﬁrm to manage the customer as an entity, rather than focus on individual products. However, this may be very expensive to achieve. Gormley (1999) reports that 92% of companies think the single view ofthe customer is important, but 88% either “not really” or “not at all” have it today. So it may be that the IT costs associated with compiling the data needed to manage the customer are formidable. Cultivating the Customer Side of Customer Relationship Management : Fournier et al. (1998) argue that companies simply have not deliv-ered reciprocal beneﬁts that are a cornerstone of customer relationships(Fournier 1998). Companies make unrealistic demands on customers. They charge loyal customers higher prices rather than lower prices. They appear pre-occupied with their very best customers and under-occupied withtheir average customer. One of the supposed beneﬁts of CRM is being able to identify best customers and focus on them (Peppers and Rogers 1993; Zeithaml et al. 2001). While this may be appropriate, it does not meanthat average customers should be abandoned or relegated to automatic call-handling systems. Malthouse and Blattberg (2005) show that many of the future best customers come from customers who are currently average. Relying on Technology to Solve What Essentially is An Issue of Corporate Culture : The view of CRM as a database marketing activity is that databases are a tool for economies of scale. They allow large ﬁrms to knowcustomers in the way that the corner drugstore used to know its customers (Swift 2001). However, this is only half the equation. The other part isthat the proprietor of the corner drugstore truly cared about his/her cus-tomers. CRM is more than having the memory and database knowledge of consumer needs and wants. It requires a corporate culture oriented towardcaring for individuals than the task-oriented corporate cultures that are
30 2 Why Database Marketing? amenable to information technology (see Deshpand´ e et al. 1993). These points have been raised by Sheth and Sisodia (2001) as well as Day (2000). Companies Have Not Been Able to Balance Customer-Centricity and Product-Centricity : CRM exhorts ﬁrms to become customer centric, that is, view their business as customer management rather than product management. Companies have perhaps come up against the ﬁnancial problems in creating a single view of the customer, the organizationalconﬂicts between CRM and product management and other groups, and the realization that their culture does not focus on the customer, and declared CRM to be a failure. Perhaps the answer is to view the solutionalong a continuum, from fully customer-centric to fully product-centric, and management's task is to ﬁnd the right balance. While the above paints a dim picture of database marketing as the basis for CRM, the Conference Board (Bodenberg 2001) sampled 96 marketing and sales executives, representing a cross-section of companies in terms of manufacturing versus service, revenues, B2B versus B2C, and size of cus-tomer base. Eighty percent of respondents reported their CRM eﬀorts either somewhat or very successful. Companies who report very successful eﬀorts are more likely to warehouse their own data. This suggests a strong commitment to CRM. The report ﬁnds that the factors that often lead to CRM success are: corporate culture and leadership, process and technology improvement,direct communications with the customer, and budgetary and cost savings. There are also anecdotal testimonies to the success of CRM programs. These include companies such as Harrah's Entertainment (Maselli 2002; Swift 2001)and several others. Two important empirical studies connect database marketing, customer relationships, and ﬁrm performance. Zahay and Griﬃn (2004) surveyed 209software and insurance managers. They measured: (1) personalization and customization, i. e., using data to create individual-level products and com-munications, (2) customer information system (CIS) development, i. e., thedegree to which the ﬁrm can generate, remember, disseminate, and interpret customer data, (3) customer performance, i. e., retention, LTV, and share-of-wallet, and (4) business performance, i. e., self-reported growth and income. The authors found that personalization and customization (i. e., the practice of database marketing), related positively to the development of the CIS, which in turn related positively to customer performance, which in turn re-lated positively to ﬁrm performance (p. 186, ﬁg. 2. 5). In summary, database marketing (developing a CIS and using it for personalization and customiza-tion), relationship development (customer performance), and business per-formance go together. Reinartz et al. (2004) surveyed 211 executives to study the relationship between CRM activities and ﬁrm performance. CRM activities consistedof eﬀorts to initiate, maintain, and terminate customer relationships. They used several self-report scale items to measure these constructs. Items in-cluded “We use data from external sources for identifying potential high value
2. 2 Creating and Enhancing Customer Relationships 31 customers” (initiation), “We continuously track customer information in or-der to assess customer value” (maintenance), “We have formalized procedures for cross-selling” (maintenance), and “We have a formal system for identify-ing nonproﬁtable or lower-value customers” (termination). Performance wasmeasured using both self-report, and for a subset of their sample, an objec-tive measure (return on assets). The authors found that CRM eﬀorts were positively associated with self-report and objective measures of performance. They also found organizational factors could enhance this association. Specif-ically, the degree of “organizational alignment,” which entailed reward sys-tems for employees who enhance customer relationships, and organizationalcapabilities to treat customers diﬀerently according to their proﬁtability, in-teracted positively with the impact of CRM eﬀorts on performance. Interestingly, the authors found that investment in CRM technology, which included enhancements to the ﬁrm's ability to target 1-to-1 and to manage“real-time” customer information, was negatively related to the perceptual measures of performance. One interpretation of these results is that while having a good customer database enhances performance, it is all too easy to over-invest in sophisticated technology that does not pay out. These two studies provide an initial set of evidence relating the compilation and utilization of customer data to customer relationships and to ﬁrm perfor-mance. The evidence is not deﬁnitive, and there are several avenues that needinvestigation. For example, Reinartz et al. (2004) do not isolate the role of customer data, treating it as a part of CRM eﬀorts. In addition, the negative results for investment in CRM technology, which is often data-oriented, givepause to the “collect-all-possible-data” dictum, and need further research. The Zahay and Griﬃn (2004) study views CRM as the antecedent of customer data, whereas the causality may be the reverse, i. e., CIS enables CRM, whichin turn enhances performance. In summary, future work should analyze diﬀer-ent models with diﬀerent measures in diﬀerent industries before we can fully understand whether and under what conditions the collection and utilizationof customer data enhances customer relationships and ﬁrm performance. 2. 2. 4 Assessment Overall, the logic for database marketing as a tool for developing customerrelationships is compelling. That retention has a bigger impact than acquisi-tion is a mathematical truism. There is empirical work that says that strong relationships lead to better customer satisfaction, better retention, and hence better ﬁrm performance. One major question is: “Do retention investmentshave a higher payout than acquisition investments?” The literature on this question is almost non-existent. An exception is Reinartz et al. (2005) who ﬁnd: (a) under-spending is more detrimental than over-spending; and (b)suboptimal allocation on retention has a greater eﬀect than under-spending
32 2 Why Database Marketing? on acquisition. However, more research is needed to understand the alloca-tion of resources between acquisition versus retention eﬃciencies and costs. Another question is whether database marketing can be used to create or improve customer relationships. There is evidence on both sides, includingtwo empirical studies supporting a positive association among database mar-keting, CRM initiatives, and ﬁrm performance. But there is a critical need for more systematic research. 2. 3 Creating Sustainable Competitive Advantage 2. 3. 1 The Basic Argument Database marketing utilizes a customer information ﬁle, which by deﬁnition is owned by one company and not the other. The company can use its in-formation to serve its customer better by identifying the correct services to oﬀer, make product recommendations, or tailor promotions more eﬀectively than its competition can do with this set of customers. This asymmetric in-formation gives a company a potential sustainable competitive advantage. It is sustainable because it would cost the competition too much to obtain the same information-they would have to buy the company. In fact, increasinglythe value of a company is determined by the value of its customer ﬁle (Gupta et al. 2004a). This vision is compelling. Customer databases are proprietary and their advantage grows as the company learns from them and improves its customer oﬀerings even more. However, this does not consider competition. In particu-lar, will each competitor assemble its own database and allow a “live and let live” customer information environment, or will they compete more intensely to acquire the competitor's customers and retain their own customers? We in-vestigate these issues as we trace the evolution of the sustainable competitive advantage argument. 2. 3. 2 Evolution of the Sustainable Competitive Advantage Argument The argument that database marketing provides a sustainable competitiveadvantage has evolved in three steps. First was the emergence of “marketing orientation” as a source of competitive advantage. Marketing orientation in-volved the collection and utilization of customer information. However, cus-tomer information was deﬁned broadly and not speciﬁcally as the customer information ﬁleused by database marketers. In the second step, Glazer (1991, 1999) and others sharpened the role of customer information ﬁles, and how
2. 3 Creating Sustainable Competitive Advantage 33 they could provide companies with a competitive edge. In the third step, economists have developed formal models explaining how the customer in-formation ﬁle could provide a sustainable increase in proﬁts. 2. 3. 2. 1 Marketing Orientation Kohli and Jaworski (1990) deﬁned marketing orientation as the “generation” of customer data, its “dissemination,” within the organization, and the “re-sponsiveness” of the organization to the information. A series of studies mea-sured marketing orientation and related it to performance. Jaworski and Kohli (1993) conducted executive surveys using two samples, of 145 and 136 strategic business units (SBU's) units respectively. (Also see Kohli et al. 1993). They deﬁned market orientation similar to their 1990 paper, and measured it on a 32-item scale. The scale included items related to actions such as meeting with customers on a frequent basis, doing in-housemarket research, collecting industry information, etc. There was no explicit measurement of the use of customer information ﬁle. The authors found that market orientation had a signiﬁcant positive rela-tionship with a judgmental business performance measure. However, market orientation had no relationship with an objective business performancemeasure-dollar market share. The antecedents of marketing orientation included top management emphasis, high interdepartmental connectedness and low conﬂict, decentralized organization, and a reward system orientationto executive compensation. This paper established that organizational factors create an environment for developing a marketing orientation. It did not however show that marketing orientation improves ﬁrm performance interms of an objective business performance measure. Moorman (1995) surveyed 92 marketing vice presidents and found that the mere collection and transmission of information had no eﬀect on the ﬁrm'snew product performance, but that “conceptual” and “instrumental” utiliza-tion were positively related. Conceptual utilization is the indirect use of infor-mation such as summarizing results, giving them meaning, etc. Instrumentalutilization is the direct application of the information to evaluating projects and giving clear direction for implementation. Moorman's ﬁndings imply it takes more than the simple collection and dissemination of the information to create an advantage and the key is in making sense of the information and actually using it to guide policy. Moorman and Rust (1999) surveyed two samples of managers, of sizes n = 330 and n = 128. They found that market orientation related toproﬁtability and market performance but interestingly, not to customer rela-tionship performance. Moorman and Rust's results imply that customer infor-mation can improve performance but not necessarily create loyal customers. It is as if the high market orientation ﬁrms use data to improve marketingproductivity, but not necessarily to nurture customer relationships.
34 2 Why Database Marketing? As described in Sect. 2. 2. 3, more recent work (Zahay and Griﬃn 2004; Reinartz et al. 2004) has speciﬁcally linked database marketing activities to ﬁrm performance. The information utilization constructs in these studies relate more directly to database marketing activities, and therefore extendthe work relating marketing orientation to performance to the more speciﬁc realm of database marketing and ﬁrm performance. Overall, the line of work linking database marketing to ﬁrm performance is growing although not yet deﬁnitive. Early work on marketing orientationﬁnds some linkages, especially Moorman's (1995) study that it is the utiliza-tion, not the mere collection of data, which builds competitive advantage. This is reinforced by Zahay and Griﬃn (2004) as well as Reinartz et al. (2004). More work is needed, especially relating database marketing to ob-jective performance measures. 2. 3. 2. 2 The Customer Information File as a Firm Asset Glazer (1991, 1999) presented the conceptual link between the general no-tion of customer information and the value of the customer information ﬁle. Glazer (1991) speaks of three types of information-based value creation: the information from upstream transactions with suppliers (V s), the information from internal operations (Vf), and the information from downstream trans-actions with customers (Vc). Customer information is of interest to database marketers, and contributes in three ways: increased revenues from future transactions (e. g., through better targeting of the right products at the right price), reduced costs (e. g., through not having to mail every oﬀer to everycustomer), and the sale of information itself (through say renting the cus-tomer list). These facets combine to determine the extent to which value generated by a product or service is due to customer information (V f). Glazer (1991) discusses that where the ﬁrm stands in terms of supplier, ﬁrm and customer information has important implications for the overallstrategy of the ﬁrm. For example, it can determine whether the ﬁrm pursuesa market share or market niche/targeting strategy. Market share strategies are based on economies of scale, high volume, and low cost, and require high supplier (V s)a n dﬁ r m( Vf) information. High customer information tilts the ﬁrm toward targeting strategies, where the key is product diﬀerentiation andfocus on a particular niche or target group. Glazer argues that if a ﬁrmcan achieve high values on all three components, it can pursue a ﬂexible manufacturing, mass customization strategy. Rust et al. (2002) take a related but somewhat diﬀerent perspective. They conceptualize the choice as between revenue expansion (focus on thecustomer), and cost reduction (focus on decreasing operations and organi-zational costs). Customer information supports the revenue expansion ap-proach, whereas supplier and ﬁrm information supports the cost reduction
2. 3 Creating Sustainable Competitive Advantage 35 Customer Responses to Purchase Proﬁt Characteristics Firm Marketing History Potential Customer 1 Demographics Oﬀers and responses Purchases Lifetime value Customer 2 Demographics Oﬀers and responses Purchases Lifetime value Customer 3 Demographics Oﬀers and responses Purchases Lifetime value Customer 4 Demographics Oﬀers and responses Purchases Lifetime value Customer 5 Demographics Oﬀers and responses Purchases Lifetime value Etc............................... Fig. 2. 8 The customer information ﬁle (CIF) and marketing strategy (From Glazer 1999). approach. They ﬁnd that ﬁrms perform3better when they focus on rev-enue expansion, illustrating the importance of customer information, than when they focus on both revenue expansion and cost reduction. So it ap-pears that companies in practice may have trouble achieving all three types of information-based value creation. Glazer's 1991 paper set the stage for his 1999 paper, where he explicitly discusses the role of the customer information ﬁle (CIF), which is the sourceof V c. He deﬁnes “smart markets” as markets where the stock of customer information changes frequently, and maintains that these markets are on theincrease. He uses the customer information ﬁle as a framework to generate strategies for succeeding in smart markets. The CIF is organized as in Fig. 2. 8 and suggests three “generic” strate-gies: row management, column management, and row and column manage-ment (“whole ﬁle”). We will just cover “row” and “column” strategies. Acolumn management strategy focuses on maximizing responses to a particu-lar marketing program or product. This may involve tailoring the product to the customer (mass customization) or targeting appropriate prices to variousbuyers (yield management). Note that column management strategies are “product-centric”. They start with a product, e. g., a credit card, and ﬁgure out how to tailor features, interest rates, and prices or fees to individualcustomers so as to maximize ﬁrm proﬁts. In contrast, row management strategies focus on each customer and ask what can the ﬁrm do to maximize proﬁts from each or a particular set ofcustomers. The focus is on interactive marketing communications designed to maximize the lifetime value of the customer. An example Glazer provides (p. 64) is American Express using a relationship-billing program with its commercial customers in which it ﬁrst provides a given establishment de-mographic analysis of its customers and then uses this information to sellestablishment advertising space in publications. Glazer echoes Moorman's (1995) point that in smart markets (markets that are driven by customer information ﬁles), the ability to process information, not the information itself, is the scarce resource. Thus, the 3Performance is in terms of return on assets (ROA) and stock market returns.
36 2 Why Database Marketing? source of competitive advantage to a ﬁrm is a combination of creating customer information ﬁles, processing of the information and then utilizing the information to drive superior marketing strategies. 2. 3. 2. 3 Economic Theories of Customer Information as a Strategic Asset The marketing orientation literature provided a conceptual and empirical basis for marketing information as a ﬁrm asset, and Glazer and others moved that literature toward a focus on the customer information ﬁle as the source ofmarketing advantage, and recent work suggests a link between customer data and ﬁrm performance. The economic modeling literature then analyzed the strategic implications of company's pursuit of competitive advantage throughmanagement of the customer information ﬁle. There are several important phenomena we will discuss that have emerged from these eﬀorts, but the central theme is that they focus on the goal of pricediscrimination, whereby the ﬁrms use customer information to identify and oﬀer higher prices to their loyal customers and lower prices to switchers. The central question is: does an environment in which ﬁrms use customerdata to target prices increase proﬁts? Economists have investigated how this is inﬂuenced by competition, by the accuracy of the targeting, and by the strategic behavior of ﬁrms as well as customers. Can Customer Information Be the Source of a Prisoner's Dilemma? Shaﬀer and Zhang (1995) investigated whether company proﬁts increase when customer preferences can be identiﬁed. Their customer behavior model arrayed customers along a continuum of preference for either Firm 1 or Firm 2(the well-known Hotelling framework). Customers trade oﬀ their preference for the ﬁrm's product versus the price of that product to decide which ﬁrm to choose. The authors assumed that both ﬁrms have perfect information oncustomer preferences and on the relative weights customers place on prefer-ences versus price. Shaﬀer and Zhang's set-up is somewhat based on Catalina Marketing, a ﬁrm that targets coupons to customers based on their previous buying habits. The buying habits can be determined based on a full customer history, or sim-ply on the product most recently purchased at the cash register. For example, if the customer buys Coke in a particular week, this suggests they prefer Coke. At that exact purchase occasion, Pepsi could target a coupon to the customerto induce a brand switch on the customer's next purchase occasion. The initially surprising, and from the perspective of database marketing, dispiriting result was that in this scenario, ﬁrms engage in a “targeting war” inwhich proﬁts are lower with customer information than without. The problem
2. 3 Creating Sustainable Competitive Advantage 37 is that ﬁrms cannot practice price discrimination. They want to charge high prices to their loyal customers but cannot do so because the competing ﬁrm can attract these “loyals” with a steeply discounted coupon. As a result, prices for loyal customers are not high enough to eﬀect price discrimination,and prices for switchers (customers in the middle of the Hotelling line) are very low as these customers are relatively indiﬀerent between ﬁrms. Shaﬀer and Zhang (1995) present a rather dismal view of database mar-keting simply as a vehicle for competing more intensively. Obviously, thisdoes not match the real-world since more and more ﬁrms are using database marketing. This goes to the issue of model assumptions. One of the key as-sumptions in their model is that ﬁrms have perfect information on customer preferences. This is rarely the case. Firms typically know only a given set of customers (their own). Imperfect Targetability Chen et al. (2001), and Chen and Iyer (2002) both make a case that in a more realistic world of imperfect targetability, ﬁrm proﬁts actually increase when they utilize customer databases. The reason is that ﬁrms are aware that their targeting is not perfect, and this cushions price competition comparedto the targeting wars in the Shaﬀer and Zhang scenario. We will review these papers in detail because they are crucial for providing the case that database marketing can be a source of sustainable competitive advantage. Chen et al. (2001) use Narasimhan's (1988) consumer model, assuming there are three types of consumers: loyal to Firm 1, loyal to Firm 2, andswitchers, which occur with probabilities γ 1,γ2,a n dχ respectively. Loyal customers will always buy from their preferred ﬁrm as long as its price is lower than their reservation price, anchored at $1 in the model. Switchers will buy the brand that is available at the lower cost to them, or will buy each brand with probability 0. 50 if prices are equal. Note that this model is diﬀerent than the Hotelling model used by Shaﬀer and Zhang, where customers were positioned along a continuum in terms of preference, and all were potentiallyvulnerable to low price discounts. The Chen et al. model is still realistic-there are customers loyal to Coke, Mc Donalds, Fleet Bank or Fidelity, who will continue to purchase these brands as long as their price does not become too high. We will examine how proﬁts change in this scenario as targetability increases. Chen et al. conceptualize “targetability” as the ﬁrm's ability to identify loyals and switchers. Chen et al. assume that a ﬁrm has information on itsown loyal customers and switchers, but not on its competitors' loyals. It cantarget its own loyals, but not its competitors' loyals. Consider Fidelity In-vestments. The assumption Chen et al. make is that Fidelity has information on a given set of customers that they can classify as loyal to them (i. e., onlybuy ﬁnancial services from them), or switchers (buy sometimes from Fidelity,
38 2 Why Database Marketing? but sometimes from Merrill Lynch), but they do not have information on cus-tomers who are loyal to Merrill Lynch. Chen et al. create a targetability index equal to 0 if the ﬁrm's ability to classify is no better than random and 1 if targeting is perfect. In their ﬁrst set of analyses, the targetability index for each ﬁrm is consid-ered exogenous. The question is how proﬁts change depending on this index. To answer this question, they identify three forces that depend on targetabil-ity. First is the segmentation eﬀect which results when ﬁrms can correctly identify their loyals, leading to gains in proﬁts because they can charge themappropriately high prices. Second is the mistargeting eﬀect,w h e r e b yﬁ r m s mis-identify switchers as loyals and hence charge them inappropriately high prices. Third is the price competiton/share eﬀect, where ﬁrms correctly iden-tify switchers and charge them low prices to gain share. Chen et al. 's ﬁrst result is that a ﬁrm that has targeting ability always attains higher proﬁts if it competes with a ﬁrm that cannot target, and the proﬁt advantage increases as targetability increases. The segmentation andprice competition eﬀects allow it to practice price discrimination without the concern of being undercut by the mass marketer, who cannot do so eﬀec-tively because it does not know to whom to target. The mistargeting eﬀect holds down the database marketers proﬁts, but as this eﬀect decreases due to better targetability, the database marketer's proﬁts increase all the more. Interestingly, while the database marketer's proﬁts are always higher than the mass marketer's proﬁts, the mass marketer actually gains over its base proﬁts when mistargeting is high. The reason is that when mistargeting ishigh, the database marketer charges overly high prices to mistargeted switch-ers, and the mass marketer gains some of these switchers without having to charge an excessively low price. In this way, the mass marketer can actuallybe better oﬀ than they would be if they were competing with a database marketer whose mistargeting costs are high. This result says that database marketing provides a sustainable proﬁt advantage if one ﬁrm practices it and another does not. However, a morelikely scenario is that both ﬁrms have the ability to target. Chen et al. show that in this case, proﬁts for both ﬁrms are always at least as high with im-perfect targeting than without it, but the relationship between targetability and proﬁts is an inverse U-shape, as in Fig. 2. 9. Firm proﬁts are maximized at intermediate values of targetability. At low levels of targetability, ﬁrmscannot practice price discrimination and hence proﬁts are low. At interme-diate levels of targetability, the segmentation eﬀect enables ﬁrms to price discriminate and the mistargeting eﬀect softens price undercutting. At high targetability, the price competition eﬀect becomes important because both ﬁrms are identifying switchers, and the mistargeting beneﬁt no longer cush-ions prices. The situation is similar to Shaﬀer and Zhang (1995). This yields lower proﬁts. Chen et al. develop a number of additional results. First, they ﬁnd that Firm 1 has a proﬁt advantage over Firm 2 if it has a larger number of
2. 3 Creating Sustainable Competitive Advantage 39 0 0. 2 0. 4 0. 6 0. 8 1 Targetability Index Firm Profits Fig. 2. 9 Relationship between targetability and ﬁrm proﬁts (From Chen et al. 2001). accurately identiﬁed loyal users. Database marketing can become a sustain-able competitive advantage especially for the ﬁrm with a strong customerbase. The Chen et al. model is static in the sense that it does not consider the impact of targeting on future loyalty, if ﬁrms can use their targeting ability to nurture their loyal customers, which in turn can increase their ability to tar-get (through more and better information revealed by these increasingly loyal customers), one can see how the ﬁrm can develop a sustainable advantage. Chen et al. also consider the optimal levels that ﬁrms should invest in database marketing. They ﬁnd that ﬁrms will decrease investment in data-base marketing if costs are high, although the ﬁrm with the larger loyalsegment will invest more. When database marketing costs are low, both will invest in database marketing to the fullest extent possible. Keeping in mind Fig. 2. 9, this implies that ﬁrms can “over invest” and end up on the right sideof Fig. 2. 9, where price competition becomes more intense, and the mistarget-ing eﬀect is not strong enough to soften price competition. Chen et al. also ﬁnd that even taking into account investment costs, if the ﬁrms have roughly thesame number of loyal customers, proﬁts for both ﬁrms are higher with target-ing than without. So database marketing is a “win-win” for the industry. If the loyal segments are unbalanced, presumably it is the stronger ﬁrm that wins. Chen and Iyer (2002) provide a diﬀerent perspective on the role of imperfect targeting by changing the analysis in two ways. First, customersare located on a Hotelling line, similar to Shaﬀer and Zhang, and no ﬁrmcommands absolute loyalty. Secondly, they provide a diﬀerent deﬁnition of targetability. Their deﬁnition of targetability is the percentage of customers at each point on the line (preference level) who can be addressed by Firm I(a i), where the assumption is that if the customer can be addressed, its preferences are known. Chen et al. assume all the ﬁrm's customers can be addressed, but that the ﬁrm is not sure whether the customers are loyals or
40 2 Why Database Marketing? switchers. Chen and Iyer assume that ﬁrms can perfectly identify the prefer-ences of all customers it can reach, but that it cannot target all customers. In Chen and Iyer's model, there are three main groups of consumers: Group 1 consists of customers that can be reached by Firm 1 but not by Firm 2(a 1(1-a2)). Group 2 consists of customers that can be reached by Firm 2 but not by Firm 1 ( a2(1-a1)). Group 3 consists of customers that can be reached by both ﬁrms ( a1a2). Firm 1 has monopoly power over Group 1, Firm 2 has monopoly power over Group 2, and both ﬁrms will compete for Group 3. Chen and Iyer call Group 1 and Group 2 the surplus extraction eﬀect, because each ﬁrm can charge a high price and still acquire its customers. This is analogousto the segmentation eﬀect in Chen et al. The Group 3 situation is called the “competitive” eﬀect, since ﬁrms will compete strongly for this segment. This is analogous to the price competition eﬀect in Chen et al. Chen and Iyer capture the mistargeting eﬀect in Chen et al. by assuming that Firm 1 knows what customers it can address but does not know whatcustomers its competition can address. This assumption appears to makesense. Capital One knows what customers are in its database, but does not know which are also in its competitors' databases. More broadly, the assump-tion means that a ﬁrm knows its own marketing eﬀorts for the customers on its list, but does not know the marketing eﬀorts of other ﬁrms with these cus-tomers. As a result of this information asymmetry, each ﬁrm faces a trade-oﬀin determining its prices. The ﬁrms want to price high for Groups 1 and 2, but need to price low in order to attract Group 3. Thus there is a trade-oﬀ between surplus extraction and price competition eﬀects. Chen and Iyer calculate equilibrium proﬁts assuming given levels of a 1and a2. The proﬁt for Firm 1 if these levels are roughly equal is: Proﬁt1=a1(1-a2)(r-t/2) +a1(a1+a2)t 2[(a2-a1)r+a1t (a1+a2)t]2 (2. 3) where tis the per-unit distance disutility incurred by customers on the Hotelling line, and ris the reservation price for one unit of the good. The ﬁrst term represents the proﬁts from Group 1 (of size a1(1-a2)) and represents the surplus extraction eﬀect. The second term represents the prof-its from competing in the switching segment and represents the competitiveeﬀect. In the case where addressability is roughly equal, both these terms are important because Firm 1 realizes both Group 1 and Group 3 are sizable. So it tries to compete in both. If a 1is much greater than a2, Firm 1 realizes that Group 1 is the largest group, and does not bother to compete for Group 3 and proﬁts just equal a1(1-a2)(r-t/2). Firm 2 faces a similar situation if its addressability is much higher than Firm 1's. One of Chen and Iyer's key results is that the equilibrium ratio of proﬁts between Firms 1 and 2 will be proportional to their investments in database marketing. Therefore, the ﬁrm that invests more in database marketing has a competitive advantage. The advantage of having an addressability advantage is the ability to price high without having to worry about losing customers
2. 3 Creating Sustainable Competitive Advantage 41 to the competitive ﬁrm (Group 3 is small). The database marketing leader is able to practice price discrimination along the Hotelling line, unfettered by worries about what its competitor might be doing. Chen and Iyer also show that if addressability is high for both ﬁrms, the result is ruinous price competition for the switching segment. Both ﬁrmsrealize they have no monopoly power and must compete for switchers. This is analogous to the Chen et al. result that proﬁts are lowest at very low or veryhigh levels of targetability. Chen and Iyer show that if the costs of obtaining addressability are low, both ﬁrms will not invest in full addressability. One will choose a i= 1 and the other will choose aj=0. 5. The reason is that if Firm ihas full addressability, Firm jrealizes that to also achieve full addressability will precipitate targeting wars for the switching segment. Firmjis better oﬀ not investing fully in addressability. This creates a monopoly segment for Firm i, which in turn cushions Firm i's prices, since it now must trade oﬀ the surplus extraction and competition eﬀects. Firm jmakes less money than Firm i, but is better oﬀ than if invested fully in addressability. An important implication of this is that in the real world, where ﬁrms can invest sequentially, there is a ﬁrst-mover advantage, and the smart company that is behind on database marketing should hold back investment to avoid the targeting war scenario of Shaﬀer and Zhang (1995). Chen and Iyer explore two important assumptions regarding their analy-sis. First, concerns the segment (1-a 1)(1-a2) that is not addressable by either ﬁrm. Their model assumes these consumers are lost to the market, but they argue that if these customers can pay “posted prices” for the prod-uct, there still is the asymmetric equilibrium when addressability is low cost. Second, concerns the assumption that addressability is the same for all con-sumers, regardless of preferences. The authors ﬁnd that if ﬁrms can chooseaddressability as a function of preference, they ﬁrst will invest in being able to address customers who have high preference for their product. This makes sense, because then they can charge higher prices. In summary, both Chen et al. (2001) and Chen and Iyer (2002) ﬁnd that companies can obtain sustainable competitive advantages through investmentin database marketing. Firms make more money when they have databasemarketing capabilities compared to when they do not. The key to this result is that there must be some mechanism that keeps ﬁrms from targeting wars as in Shaﬀer and Zhang. For Chen et al. that mechanism is that ﬁrms arenot sure if a customer in their database is loyal to their ﬁrm or a switcher. Certainly this ﬁts most situations. For Chen and Iyer, the mechanism is that ﬁrms do not know if their competitors can target their customers. This keeps ﬁrms from charging low prices because they realize that they may be “leaving money on the table” by charging low prices to customers who arenot addressable by the competition. So the interesting conclusion is that an intermediate level of database marketing capability is best because it creates enough information to obtain the gains from targeting, but not too much asto spark targeting wars.
42 2 Why Database Marketing? The Strategic Consumer The Chen et al. and Chen and Iyer papers assume that ﬁrms are pre-scient. They do not have perfect targeting information, but they are awareof what they know and do not know, and consider the short and long-term implications of their information set. The consumer, on the other hand, is considered to be passive. However, what happens if the consumer realizesthat the underlying goal of the ﬁrms is to practice price discrimination, and that by revealing their preferences, they may be the subject of price dis-crimination? In two important papers, Villas-Boas (1999, 2004) shows thatif consumers behave strategically, ﬁrms can be worse oﬀ if they can iden-tify their customers. Villas-Boas (2004) is particularly important because this is the monopolist case, and demonstrates that the disadvantage is notdue to a competitive targeting war as in Shaﬀer and Zhang. The problem is that consumers hold out for lower prices because they realize that if they do not reveal their preferences, ﬁrms will not be able to distinguish themfrom brand switchers or customers new to the market and they will get low prices. Chen and Zhang (2002) acknowledge this possibility but argue that the eﬀect will be more than counter-balanced by the “price-for-information” ef-fect. The eﬀect arises as follows. Firms want to price discriminate but needto identify customer preferences in order to do so. In a two-period model, they are tempted to price low in the ﬁrst period because they realize some customers are holding out for cheaper two-period prices. However, they alsorealize that by pricing high, they do not attract as many customers but the customers they attract are clearly loyal to them, and they can use this in-formation to charge appropriately high prices in the second period. In otherwords, ﬁrms charge higher prices for the information they gain about cus-tomers that can be utilized in the long-term. This is the price-for-information eﬀect. Chen and Zhang show that even taking into account strategic cus-tomers, ﬁrms can be better oﬀ with database marketing than without. They do have to lower their ﬁrst-period prices to discourage their loyals from wait-ing, but they do not need to lower them completely because they realizethey will gain in the long run from learning about the customers they do attract. The area of strategic consumers is a crucial one for the success of database marketing. A very diﬀerent venue where the eﬀect shows up is a static ratherthan a dynamic one. Feinberg et al. (2002) argue that customers can become jealous of other customers who get better deals than them. They then may refrain from purchasing from the ﬁrm according to their preferences. Essen-tially, the customer is taking into account prices available to other consumersto assess its likelihood of buying from the ﬁrm. This may not be seen as strictly rational (why should what someone else gets aﬀect your utility for a product), but Feinberg et al. show in experiments that the jealousy eﬀect isreal. To the extent that this jealousy eﬀect is large, it decreases the ability
2. 3 Creating Sustainable Competitive Advantage 43 of ﬁrms to price discriminate, which is the driving force behind the economic arguments to date for database marketing. The economic models described above make a set of key assumptions which drive their results: (1) the only strategic variable is price, (2) the purposeof database marketing is to allow the ﬁrm to price discriminate, (3) ﬁrms can target their loyal customers, and 4) only two ﬁrms compete. Each of these assumptions is suspect in the real-world. These models assume thepurpose of database marketing is price discrimination. There is no empiri-cal evidence that this is the goal of database marketing. Database market-ing goals are far broader than simply price discrimination, as Glazer (1999)discusses. Glazer (1999) shows that ﬁrms can compete using diﬀerent (row and col-umn) strategies, some of which are diﬀerent than price. Under column strate-gies, he provides examples, one of which is yield management (similar to price in the economic models), but discusses mass customization as another example. He also discusses row strategies in which ﬁrms use addressabilityto develop customer interaction strategies to increase their loyalty. Economic models (to date) do not consider customer interaction strategies to increase loyalty as a goal of database marketing. Many ﬁrms do not have any information about their customer's loyalty. All they can observe is purchase behavior (and maybe demographic informa-tion). The assumption of all of these models is that the ﬁrm somehow knows the loyalty level of its customers and then targets based on it. For exam-ple, Fidelity Investments does not know if its customers have accounts with Merrill Lynch, T. Rowe Price or Vanguard. One of the few industries which might know its customer loyalty is credit card issuers in the USA because they have information about the number and usage of cards through creditbureaus. However, it is diﬃcult to identify many other industries that know the loyalty level of their customers. Some ﬁrms use customer behavioral data to price their bestcustomers lower than the competition. Vanguard oﬀers lower fees to its Admiral cus-tomers, determined by the size of balances they have within a given mutual fund. The higher balance customers receive lower fees as a percentage ofmoney invested. This may be a form of competitive pricing but is not price discrimination as in the models reviewed above. Firms may try to price discriminate (airlines) but can succeed because they use another strategic variable (level of service) as the basis for customers'willingness to stay loyal even though they may be paying a higher price. The database allows the ﬁrm to identify those customers to oﬀer a better service. The assumption that only two ﬁrms compete may also pose problems. If a new entrant cannot enter the industry because of the use of customerdatabases by incumbent ﬁrms, then there is a return to database marketing. Clearly in some industries, new entrants face an uphill battle because they cannot target. An important research area is to identify industries in whichdatabase marketing is an entry deterrent.
44 2 Why Database Marketing? 2. 3. 3 Assessment The evidence to date regarding database marketing as a route to sustainable competitive advantage is built on the following arguments: Empirical studies ﬁnd some, although not overwhelming, evidence that marketing orientation-the ability of ﬁrms to collect, process, and imple-ment customer information-as well as undertaking database marketingactivities, is positively related to ﬁrm performance. The customer information ﬁle-the ﬁrm's database of its customers-isthe modern source of customer information. The ﬁle suggests two principlestrategies-customer centric (row strategies), and product centric (column strategies). Strategic advantage is based on maintaining customer infor-mation and developing these strategies. Economic models develop theories under which ﬁrms using pricing-oriented column strategies can practice eﬀective price discrimination. The main requirement is that targeting abilities need to be “moderately eﬀective”. Too little and there are not enough beneﬁts of targeting; too much and ﬁrms engage in targeting wars. The arguments are interesting but more is needed to make establish thatdatabase marketing is a long-term source of competitive advantage. The mar-keting orientation studies provide some empirical evidence, but they refer to customer information in general and not to database marketing per se. Z a h a y and Griﬃn (2004) and Reinartz et al. (2004) provide important evidence that database marketing itself-using the customer information ﬁle-can be asso-ciated with better performance. However, the performance measures in sev-eral of these studies are self-report. More studies with objective performancemeasures are needed. The conceptual arguments regarding the customer in-formation ﬁle and row (customer-centric) versus column (product-centric) strategies are well-taken, but have not undergone empirical testing. Do rowstrategies really increase loyalty? Can they be implemented inexpensively enough to increase proﬁts? The economic models provide logic and some insights, but they have not been tested empirically. Empirical research along the lines of the marketingorientation literature is needed, with the focus on targetability through cus-tomer information, not customer information in general. In terms of columnstrategies, more work is needed to understand whether prices for loyal cus-tomers should be higher (price discrimination) or lower (pay the customerfor their loyalty) (see Shaﬀer and Zhang 2000), or to keep the loyal customer from getting jealous as in Feinberg et al. (2002). The theory also needs to be extended to non-price column strategies, e. g., cross-selling, and to row strategies, i. e., long-term management of customervalue. The extension to non-price column strategies would be particularly interesting. Managers would certainly like to think that customer databasesenable them to serve customers better by targeting appropriate services from
2. 4 Summary 45 their product line, or by tailoring their product line to the customer. It might be that this type of targeting is more sustainable because it is more diﬃcult for a competitor to understand the details of a customer's preferences for vari-ous product attributes than it is to understand price response. Row strategiesmight also be a source of more sustainable advantage, because long-term re-lationships may create switching costs that bind the customer to the ﬁrm. This leads to the existence of database marketing as a tool for enhancingcustomer relationships. 2. 4 Summary In this chapter, we have proposed and reviewed three fundamental reasons for companies to practice database marketing: enhancing marketing produc-tivity, enabling the enhancement of customer relationships, and establishinga sustainable competitive advantage. The marketing productivity argument is based on the use of data and data mining tools to prioritize and target customers with appropriate products, services, and prices. There is good evidence that this can work. Data mining indeed can produce “lift charts” for predicting customer behavior that aremuch better than random, and therefore can identify the customers for whom marketing eﬀorts would be wasted. The enhancing relationship argument is based on the notion that enhanced customer relationships improve ﬁrm performance, and database marketingcan enhance relationships. The ﬁrst part of the equation is well-supported by the importance of customer retention in lifetime customer value, and em-pirical studies that link customer relationships, customer satisfaction, cus-tomer retention/loyalty, and ﬁrm performance. Regarding the second part of the equation, there is a host of articles in the managerial literature thatraise questions about whether CRM investments lead to improved ﬁnancial performance. However, systematic empirical studies are beginning to ﬁndthat indeed these investments can pay oﬀ. The competitive advantage argument is based on the notion that the cus-tomer data ﬁle is a company resource that is impossible for companies toduplicate, that the data enable ﬁrms to service customers better than com-petitors, and that the better-than-random yet imperfect nature of predictions that come from the model cushions price competition. This area has receivedthe least empirical study although the concept is compelling. There is signiﬁcant academic research pertaining to the fundamental rea-sons ﬁrms should use database marketing but there is much more to do. Regarding the productivity argument, we need more ﬁeld tests that show predictive models work, that they generate incremental proﬁts beyond chan-nel cannibalization and beyond what could be generated by simple manage-ment heuristics. Regarding the sustainable competitive advantage argument,
46 2 Why Database Marketing? we need survey-based research similar to the marketing orientation literature that links database marketing, as opposed to customer information in general, to ﬁrm proﬁts. We need more economic theory on non-price targeting, high versus low prices for loyals, and the strategic consumer. We need empiricaltests of the economic models, particularly the role of imperfect targeting. Regarding the enhancing CRM argument, we need to establish the link from database marketing to enhanced relationships to satisfaction to reten-tion to performance. The last four links have been investigated; the crucial link is that database marketing enhances relationships. We also need to in-vestigate the cost side of database marketing, and in particular, whether ac-quisition costs are truly higher than marginal retention costs. More generally, we need to investigate if and under what conditions retention management is more cost-eﬀective than customer acquisition strategies.
Chapter 3 Organizing for Database Marketing Abstract Quantitative analysis is endemic to database marketing, but these analyses and their implementation are not conducted in an organizational vacuum. In this chapter, we discuss how companies organize to implement database marketing. The key concept is the “customer-centric” organization, whereby the organization is structured “around” the customer. We discuss key ingredients of a customer-centric organizational structure: customer man-agement and knowledge management. We also discuss types of database mar-keting strategies that precede organizational structure, as well as employee compensation and incentive issues. 3. 1 The Customer-Centric Organization Successful implementation of database marketing certainly requires mas-tery of data management and modeling methodology. However, these tools are not applied in an organizational vacuum. In this chapter we discuss how to design organizations for implementing database marketing success-fully. A key concept to emerge in this context is that of the “customer-centric” organization. This means that the organization is structured “around” thecustomer-from the customer in, rather than from the product out. Inthe words of industry expert David Siegel as quoted by Stauﬀer (2001), “If you really care about customers... then you have to reorganize your entire company around customers. ” Stauﬀer then says, “It's not organizing the com-pany to serve customers. It's letting customers determine how you organize. ” Galbraith (2005, p. 6), states customer-centricity as an imperative: “The need for customer-centricity is not going away, and it is up to each company tod e t e r m i n et h el e v e l o f a p p l i c a t i o n... re quired for success. ” 47
48 3 Organizing for Database Marketing Strategy Customer Intimacy * Operational Efficiency * Marketing Efficiency * The Strategy Locator Structure Customer Managers Customer Portfolios Acquisition vs. Retention People Training Coordination Culture Rewards Customer Metrics Short vs. Long Term Processes Knowledge Management Fig. 3. 1 Star model of the customer-centric organization (From Galbraith 2005). ∗These concepts are used by Langerak and Verhoef (2003). We will frame our discussion using the “Star” model developed by Galbraith (2002, 2005). The Star model emphasizes ﬁve ingredients for suc-cessful organizational design: strategy, structure, processes, rewards, and people (Galbraith 2005, p. 15). Strategy refers to the goals of the organizationand the means by which it intends to achieve them. Structure refers to the organizational chart-what departments and positions need to be cre-ated, and how will they interact. Processes refer to the means by whichinformation ﬂows within the organization. Rewards refer to the compen-sation and incentives that ensure the employees of the organization per-form eﬀectively. People refers to the policies that ensure that employ-ees have the right skills and “mind-set” to implement the organizational design. Figure 3. 1 shows the Star model applied to designing the customer-centric organization. Listed under each of the ﬁve components of the framework arethe key issues that will be discussed in the following sections. 3. 2 Database Marketing Strategy The organization design for implementing database marketing emerges fromthe ﬁrm's database marketing strategy. The key issues are: (1) What is that
3. 2 Database Marketing Strategy 49 strategy, and (2) How will the organizational design establish a competitive advantage? 3. 2. 1 Strategies for Implementing DBM 3. 2. 1. 1 The Langerak/Verhoef Taxonomy Langerak and Verhoef (2003) distinguish three types of CRM strate-gies: Customer Intimacy, Operational Eﬃciency, and Marketing Eﬃciency. Customer Intimacy means that the company's strategy truly is to de-liver personal service to its customers, to know them on an intimate baseand customize its products, services, and communications to them. Opera-tional Eﬃciency employs CRM to reduce costs and utilize non-marketing resources eﬃciently. Marketing Eﬃciency uses customer data to improvemarketing productivity, i. e., making marketing more eﬀective at achieving less churn, more successful cross-selling, and in general, greater customer proﬁtability. Langerak and Verhoef argue that organization design should follow from which of the three strategies the company pursues. For example, they studya private investment banking ﬁrm whose strategy was Customer Intimacy, but the company approach to customer service was actually quite imper-sonal. The ﬁrm realized it needed to develop personal, intimate relationshipswith its customers. They grouped their customers into three need segments (“self-made man,” “strategy maker,” and “security seeker”) and assigned a customer management team to each group. They created an organizationalstructure that best implemented their strategy. Langerak and Verhoef also studied an insurance company that competed on operational excellence, i. e., “price, convenience, and speed. ” This meantthat the company needed to keep operations costs as low as possible, and develop ways of interacting with customers that were as fast and eﬃcient as possible. This strategy required a highly transactional relationship withcustomers. The company adopted an organizational structure based funda-mentally on data management. The data management group fed information to the rest of the organization to help it be more eﬃcient. It especially sup-ported the ﬁrm's eﬀorts on the Internet channel, where products could be personalized at low cost. Finally, Langerak and Verhoef studied a holiday resort company whose marketing eﬀorts were highly ineﬃcient. They provided mass-mailing oﬀerswith very low response rates. They needed CRM to improve marketing ef-ﬁciency. Accordingly, they set up a CRM department that focused on data mining, database management, and integrating database marketing and cus-tomer contact eﬀorts. The system was in place only to increase the produc-tivity of their marketing eﬀorts.
50 3 Organizing for Database Marketing The main point is that the three generic CRM strategies identiﬁed by Langerak and Verhoef each require diﬀerent organizational designs and dif-ferent levels of customer-centricity. 3. 2. 1. 2 Galbraith's “Strategy Locator” Galbraith (2005, pp. 32-33) also proposes that the desired degree of customer-centricity depends on the strategy of the company. He develops a “Strategy Locator”, a measurement scale consisting of two dimensions: Scale and Scope, and Integration. Scale and Scope refers to the number and variety of products marketed by the company. Integration refers to the degree that the company'sproducts must be packaged or bundled together to deliver satisfaction to the customer. According to Galbraith, the higher the company scores on this scale, i. e., the degree to which the company oﬀers many varied productsthat must be integrated, determines the degree to which the ﬁrm must be customer-centric. Galbraith describes a chemical company that only required “light-level” customer-centricity. The company had relatively few products that did notneed to be integrated. It therefore rated low on the strategy locator. The organizational design did include some elements of customer-centricity-e. g., customer management teams-although the formal organizational structure centered on functions and geographic areas. Galbraith then describes an investment bank that required a “medium-level” degree of customer-centricity. This company had a moderate numberof banking products that required integration. It therefore rated medium onthe strategy locator. The organizational design included not only customer managers, but formal processes to ensure that customer contacts were co-ordinated within the customer management team. Formal reward structuresbased on customer performance were implemented, and formal CRM training programs were put in place. Galbraith uses IBM as an example of requiring a “complete-level” de-gree of customer-centricity. IBM has several diﬀerent products, requiringa high degree of integration. IBM therefore rates high on the strategy lo-cator. IBM's strategy focused on delivering customer “solutions”, a highlycustomer-centric idea. The notion was to solve the customer's problem, whatever products and services were required. Given the complexity of problems, this required very high coordination among IBM management. IBM now has a solutions-oriented structure where Product managers work with the customer to deliver the right combination of IBM products andservices to solve the customer's problem. Its processes help ensure that customer plans and priorities are shared easily among the relevant man-agers involved with the customer. The company still uses quotas to re-ward salespeople, a product-centric approach, but also formally assesses
3. 2 Database Marketing Strategy 51 the “competencies” of its employees to make sure they match customer needs. 3. 2. 2 Generating a Competitive Advantage Firms are constantly trying to establish a competitive advantage-a corecompetence that gives them a sustainable edge over its competition. One possibility is that the organizational design through which the company im-plements database marketing might be a source of competitive advantage. Peteraf (1993), articulating the “resource-based view of the ﬁrm,” deﬁnes four factors that determine whether a company's competences will translateinto competitive advantage: heterogeneity, ex-post limits to competition, im-perfect mobility, and ex-ante limits to competition. 1Heterogeneity means that ﬁrms within the industry have diﬀerent competencies. For example, one ﬁrm may develop a marketing analytics group that is diﬀerent, and better, than the groups at other companies. Ex-post limits mean that the company's capabilities are diﬃcult to replicate. For example, competitors may knowwhich software package the ﬁrm uses for cross-selling, but because the ﬁrm has an organizational structure that emphasizes customer management, it knows its customers so well that no other ﬁrm can duplicate its success. Imperfect mobility means that the resources that give the ﬁrm its compet-itive advantage cannot be obtained by another ﬁrm. Competitors often try to hire away a ﬁrm's best managers. However, a customer manager might beeﬀective because the scale of the ﬁrm permits frequent interaction with the marketing analytics group. So a ﬁrm cannot simply hire this manager away and expect the same success. Ex-ante limits refer to ﬁrst-mover advantage. For example, a company that ﬁrst uses CRM for operational eﬃciency may be “ahead of the curve” in terms of the organizational structure that bestsupports this strategy. 3. 2. 3 Summary Strategy plays a pivotal role in determining the organizational structure for implementing database marketing. While “customer-centricity” has come into fashion, Langerak and Verhoef (2003) as well as Galbraith (2005) ar-gue that not all organizations need to adopt the same degree of customer-centricity. Another major theme is that the goal is to wed the ﬁrm's databasemarketing strategy with an organizational design that creates a competitive advantage for the ﬁrm. 1The authors thank Professor Margaret Peteraf and Justin Engelland, Tuck MBA 2005, for helpful discussions on this topic.
52 3 Organizing for Database Marketing Vice President Marketing Product Management Marketing Services Advertising Promotion Product A Manager Product B Manager Product C Manager Fig. 3. 2a Product management (Adapted from Peppers and Rogers 1993). Vice President Marketing Marketing Services Customer Management Advertising Promotion Capabilities Managers Customer Manager Portfolio 1 Customer Manager Portfolio 2 Customer Manager Portfolio 3 Product A Product B Product C Fig. 3. 2b Customer management (Adapted from Peppers and Rogers 1993). 3. 3 Customer Management: The Structural Foundation of the Customer-Centric Organization 3. 3. 1 What Is Customer Management? The customer management organization structure has been articulated by Peppers and Rogers (1993, pp. 175-206). Their idea is that the marketing eﬀorts of the ﬁrm should be organized by customer groups or “portfolios”, each portfolio managed by a customer manager. This is in stark contrast to the product management structure. Figure3. 2 illustrates. In the product management structure (Fig. 3. 2a), product managers run their products asproﬁt centers. They are responsible for generating sales and proﬁts. They rely on the traditional “Four P's” (product, price, distribution, and promo-tion), draw on services provided by advertising and promotion departments,and work closely with production managers on product improvements and quality. The customer management framework (Fig. 3. 2b) clusters the ﬁrm's cus-tomers into portfolios. One possible clustering is by sales level-heavy,
3. 3 Customer Management 53 medium, and light user customer portfolios. Each customer would be as-signed to one and only one portfolio. Each portfolio would be managed by a customer manager. The customer manager would draw support from ad-vertising and promotion departments, and from “capabilities managers,” theformer product managers who would now be responsible for making sure the products performed up to the standards needed to serve customers. Customer managers would work with product managers on quality issues as well as newproduct features and other product development tasks. The customer manager's goal is to increase the lifetime value of the cus-tomers in his or her portfolio. This emphasizes the long-term orientation ofthe customer manager. Peppers and Rogers deﬁne the customer manager's job as follows (1997, pp. 356-357): “... someone must be assigned the re-sponsibility for managing customers individually....T h ec u s t o m e rm a n a g e r ' s responsibility is to manage each customer relationship, supervising the ﬁrm'sdialogue with each, ﬁnding products and services for each, and determining how best to customize to meet each customer's individual speciﬁcations. Inshort, the customer manager's job is to delve more and more deeply into each individual customer's needs in order to lock the customer in, make the ﬁrm more valuable to the customer, and increase the company's margin-with each customer. ” 3. 3. 2 The Motivation for Customer Management The motivation for customer management rests on three assumptions: (1)Stronger customer relationships yield higher sales and proﬁts. (2) The product management system is not eﬀective at developing customer rela-tionships. (3) The customer management system is eﬀective at developing customer relationships. The premise for the ﬁrst assumption is that the customer is more powerful today than ever before. In a B2C context, customers in industries rangingfrom ﬁnancial services to telecom to travel to retail face an ever-expanding array of choices and they make choices with more information (due to the Internet). In B2B industries, companies ranging from IBM to Xerox face the same sophisticated customer. Companies like P&G are becoming more like B2B companies-their customers are Wal-Mart and the newly consolidated supermarket companies. The assumption that better customer relationships feed ﬁrm performance has received some empirical support (Reinartz et al. 2004; Zahay and Griﬃn 2004; Day and Van den Bulte 2002; Chaston et al. 2003), although more work is needed. The second assumption has not received empirical testing. The logic is that product management maximizes sales, not customer satisfaction. Each of theﬁrm's product managers acts individually, with the result that customers are bombarded with oﬀers and selling pitches. The customer is “turned oﬀ”by this marketing blitz, and perhaps most importantly, ﬁnds him or herself
54 3 Organizing for Database Marketing owning the wrong products. A good example would be ﬁnancial services, where the customer becomes over-invested in retirement products like IRAs when he or she should be investing in college-funding instruments. In short, the ﬁrm spends too much money on marketing, many of its eﬀorts cannibalizeeach other, and they don't yield better customer relationships. The third assumption, that customer management is eﬀective for devel-oping customer relationships, has also not been tested directly. In Sect. 3. 5,we discuss evidence that customer-oriented incentive systems produce more satisﬁed customers and better marketing performance. But this does not vali-date customer management per se. These incentives could be used for product managers as well as customer managers. In summary, the motivation for customer management is that customer relationships are vital, product management is antithetical to this goal, andcustomer management will be successful at achieving this goal. This motiva-tion has received some empirical support but much more evidence is needed. 3. 3. 3 Forming Customer Portfolios A major challenge is how to deﬁne customer portfolios. Peppers and Rogers advocate that ﬁrms deﬁne portfolios based on customer needs. This allowsthe customer manager to specialize in serving the needs relevant to these customers. There are many ways to actualize this idea. One method is to group customers by volume. This is consistent with customer tier manage-ment (Chapter 23). An airline for example may have customer managers for its premium tier customers. While customer volume is a natural grouping scheme, there are many others. A ﬁnancial services company may groupcustomers by life-stage, e. g., young professionals, families, and retirees. A software company may group customers by line-of-business, e. g., education versus business, or by industry. In fact, a major challenge in customer man-agement is to decide exactly how to form the customer portfolios, and how many portfolios should be deﬁned. This is very much the perennial marketing issue of how a market should be segmented. One challenge in deﬁning customer portfolios is customer movement be-tween portfolios. For example, the ﬁnancial services customer manager foryoung professionals should be concerned with passing along good customers to the customer manager for young families. The customer manager for low-volume customers should be concerned with turning them into high valuecustomers. The compensation system becomes key-it should be based not only the current proﬁtability of the customer portfolio, but also on how many customers the customer manager converts to high volume customers, or thenumber and quality of young professionals the customer manager passes along to the young family customer manager. Referring back to the Star model (Fig. 3. 1), this is an example where structure (the customer managementsystem) interacts with compensation.
3. 3 Customer Management 55 3. 3. 4 Is Customer Management the Wave of the Future? To ﬂesh out the key issues for ﬁrms deciding whether to pursue customer management organizational structures, in this section we discuss the pros and cons from an advocacy viewpoint. 3. 3. 4. 1 Why Customer Management Is Inevitable Customer management is inevitable and the ﬁrms that move ﬁrst toward this system will achieve the highest rewards. The reasons for this are: Customer satisfaction is the key to success and customer management will produce higher customer satisfaction than product management. Customer management is truly focused on serving customer needs, whereas the prod-uct manager's goal is to sell product. Customer management creates sustainable advantage. Customer manage-ment encourages each company to know the needs of itscustomers better, and it is diﬃcult for other ﬁrms to replicate this knowledge. Product management is inherently short-term. This is because it empha-sizes current proﬁts for one product. Customer managers are concernedwith lifetime value of the customer, which is inherently long-term. Modern Information technology enables customer management. Until re-cently, ﬁrms did not have the data management systems nor the statis-tical tools required to pursue customer management activities such ascross-selling, lifetime value management, churn management, etc. These systems and tools are now in place. Customer management may be revolutionary but it can be implemented in an evolutionary fashion. For example, Mac Donald (2001) reports that Nike Canada assigns “consumer champions” to speciﬁc customer groups. Customer champions do not have line responsibility as prescribed by a complete customer management structure, but they can change the con-versation from “let's sell more basketball shoes” to “let's increases salesto teenage boys”. 3. 3. 4. 2 Why Customer Management Would Not Work There are just too many practical, cultural, and structural reasons why cus-tomer management will be very diﬃcult to implement. These include: Product management is deeply ingrained in corporate culture. Companies are product/sales/short-term oriented. Wall Street demands this, and it produces the most easily measured results. Customer management requirestoo much of a change in organizational culture.
56 3 Organizing for Database Marketing Customer management will steer companies away from their distinctive competencies. Most companies have distinctive competencies and cannot deliver the best product in each category. Customer managers may urge aﬁnancial services ﬁrm sell a mutual fund, but if this is not a high qualityfund, this will produce dissatisﬁed clients in the long run. Customer management will create even worse conﬂicts than those foundamong product managers. Each customer manager will want more funds and will make competing demands on capabilities managers. For example, managers of the teen-age customers will demand certain features for the company credit card, while mangers of the 50+ customers will demandother features. Who has the authority to referee the demands of the cus-tomer managers for new product features versus the capabilities manager's view that these features are too expensive? It is diﬃcult to measure the key performance indexes for customer man-agers. Performance indices for customer managers include share-of-wallet (SOW) and lifetime value (LTV). But SOW is diﬃcult to measure because Firm A does not have data for how much business each customer does with Firms B and C. LTV calculations require many assumptions about reten-tion rates, etc. It's impossible to design a reward system based on such fuzzy measurements. It is not practical for many companies. How can General Motors organize around Teens, Young Families, Young Professionals, Elderly, etc? How can General Mills organize around Families with Children, Singles, Elderly, etc? They just don't have direct access to customers on that basis. Do customer managers have the expertise ? A customer manager must have the expertise to diagnose customer needs and prescribe the right productsfor each customer. In many industries, the product is so technical that noone manager can possibly understand all the products. IBM may try to address this through a team approach, but that requires a lot of coordi-nation. Customer management is expensive. It adds a new layer of managers-the customer manager. It does not eliminate the product manager-it justchanges his or her responsibilities. The result is higher personnel costs insalary and support. Product management takes into account customer needs anyway. Product managers are marketers-they develop products to ﬁt the needs of a targetgroup. 3. 3. 5 Acquisition and Retention Departmentalization Until now, our focus has been on managing current customers, but whatabout the management of customer acquisition? An important aspect of thecustomer-centric organization is the division of eﬀorts into acquisition and
3. 4 Processes for Managing Information: Knowledge Management 57 retention. These are two very diﬀerent functions. For example, Del Rio (2002) describes a wireless phone company with separate departments for acquisi-tion and retention. Publishers have traditionally employed acquisition edi-tors, who sign up authors and books, and managing editors, who manage theediting, production, and marketing of the books. The advantage of this departmentalization comes from the fact that acqui-sition and retention require two diﬀerent mind-sets, involve diﬀerent tools,and have diﬀerent success measures. Acquisition is entrepreneurial. It is more straightforward to measure, reward, and motivate. It is short-term. Reten-tion management is quite diﬀerent. It is diﬃcult to measure (i. e., it relies onlifetime value and share-of-wallet), and therefore diﬃcult to reward. It is long term. The disadvantage of acquisition/retention departmentalization is that the acquisition department may not acquire the right customers. For example,an acquisition manager might use price discounts to attract customers who are inherently deal prone churners; impossible to retain. The key challenge therefore lies in coordination. Incentives could be used to make sure the acquirers attract the right customers. These might entailmeasures such as lifetime value. Ainslie and Pitt (1998) provide interesting evidence that it is possible to develop models that guide acquisition eﬀorts according to long-term customer management goals. They model prospectsin terms of their ultimate proﬁtability, risk, and responsiveness to future modeling eﬀorts. Then they prioritize customers in terms of an overall index of these three criteria. Thus it may be possible to use predictive models tofacilitate coordination between acquisition and retention. 3. 4 Processes for Managing Information: Knowledge Management 3. 4. 1 The Concept Knowledge management is the systematic process of creating, codifying, transferring, and using knowledge to improve business performance. Knowl-edge management pertains to any type of knowledge generated by the organization, but in the context of database marketing, we are concerned with knowledge about the customer. Davenport and Prusak (1998, pp. 1-6) distinguish among data, informa-tion, and knowledge. Knowledge management systems entail all three. Data isthe raw, stored input, unanalyzed. Information is compiled data that “makesa diﬀerence” (Davenport and Prusak p. 3) in a decision. Knowledge is one step up from information. It is a mix of “experience, values, contextual infor-mation, and expert insight that provides a framework for evaluating and in-corporating new experiences and information” (Davenport and Prusak p. 5).
58 3 Organizing for Database Marketing For example, consider a cross-selling campaign for audio speakers that can be used with a computer. Each customer's response can be recorded. This is data. The data can be compiled to yield a response rate. This is information. It can be used to calculate proﬁtability of the campaign. The data could beanalyzed to determine that those who responded had bought a computer in the last 3 months. The insight, or knowledge, generated is that customers who have recently invested in computer hardware are “ripe” for peripherals. This suggests a particular target group as well as copy (“no new computer s y s t e mi s c o m p l e t e w i t h o u t t h ebe s ts pe a k e r s... ” ). Knowledge management draws on information technology, economics, or-ganization behavior, human resource management, and marketing. Informa-tion technology underlies the data warehousing issues that are crucial for knowledge management. While we are not aware of formal economic analysesof knowledge management, Davenport and Prusak (1998) argue that the ﬁrm faces both an internal and external market for knowledge. There are buyers, sellers, and prices. Organizational behavior scholars have studied knowledgemanagement under the label “organizational learning” (e. g., Argote 1999), fo-cusing on how organizations learn, how they forget, how they remember, and how information is shared. Human resource management views knowledge management as a human capital issue, and is concerned with how to provide the skills for employees to learn and share their learning (Tapp 2002, p. 110). Marketers have touched upon knowledge management in their study of “marketing orientation. ” In fact, Kohli and Jaworski (1990) deﬁne marketingorientation as the generation, dissemination, and utilization of informationrelated to customer needs. This is very close to our deﬁnition of knowledge management. 3. 4. 2 Does Eﬀective Knowledge Management Enhance Performance? As just mentioned, the concept of marketing orientation is similar to knowl-edge management. Therefore, the evidence of a positive relationship between marketing orientation and ﬁrm performance (Jaworski and Kohli 1993; Moor-man 1995; Moorman and Rust 1999) suggests knowledge management can pay oﬀ. The caveat, however, is that these studies focused on the generalcollection and utilization of customer information, and not on knowledge gained through database marketing. Some research connects knowledge management with successful CRM. Chaston et al. (2003) surveyed 223 UK accounting ﬁrms. They measuredknowledge management in terms of orientation toward acquiring knowledge from external sources, exploiting new knowledge, documenting carefully, mak-ing information available to all employees, and improving employee skills. They thus covered the create-codify-transfer-use dimensions of knowledge management (Di Bella et al. 1996). They measured CRM orientation in termsof maintaining close contact with clients, regularly meeting with clients,

End of preview. Expand in Data Studio

README.md exists but content is empty.

Downloads last month: 8

Size of downloaded dataset files:

1.1 MB

Size of the auto-converted Parquet files:

1.1 MB

Number of rows:

869