Rex Yuxing Du, Wagner A. Kamakura, & Carl F. Mela
Executive Summary
Firms often collect substantial data on their customers. However, information on customers’ transactions with competing firms is often sparse or nonexistent. As a result, firms are compelled to assume an inward-focused view of their customers. This is problematic for two reasons. First, there is often little correlation between customers’ transactions with the firm and their transactions with competitors, suggesting that internal data alone do not reflect customers’ potential value to the firm. Second, a small percentage of customers account for a large portion of all external transactions, suggesting considerable potential to increase sales if these customers can be correctly identified and incentivized to switch. Accordingly, the authors develop a list augmentation approach to augment firms’ internal records with information on the size and share of their customers’ wallets. This approach infers individual customers’ total category purchases and a focal firm’s share by augmenting the firm’s internal database with survey information on a sample of customers’ purchases from the firm’s competitors.
The authors apply their approach to a proprietary data set provided by a major U.S. bank covering ten categories of financial products. The approach correctly predicts 72% of cases in which customers use competing firms’ products and offers accurate estimates of total demand and share-of-category requirements based on what the firm knows from its customer database. The model yields additional insights into customers’ share-of-wallet decisions that can be used to manage customer relationships better. For example, the authors find that (1) longer relationships are not necessarily associated with larger share of wallet; (2) customers with a high share in one category also tend to have a high share in another; (3) for some categories, customers with small shares within the focal firm tend to transact large volume outside of it; and (4) customers with higher income tend to balance share of requirements across firms.
Furthermore, the authors explore various targeting strategies predicated on the model and find that substantial lifts in targeting efficiency can be obtained by using the estimated total expenditures and share of wallet obtained with the model. For example, 13% of customers in the validation sample are identified as high-potential customers because their estimated total wallet is in the top quintile but their estimated share of wallet at the focal firm is below average. These customers account for 53% of the validation sample’s financial requirements that are fulfilled outside the focal bank, suggesting considerable potential for increasing revenue, to the extent that the focal firm can induce them to switch.
Biography
Rex Yuxing Du is Assistant Professor of Marketing at the University of Georgia. He received his BE from Shanghai Jiaotong University and his PhD from Duke University. His research work has been published in Journal of Marketing, Journal of Marketing Research, Marketing Science, Quantitative Marketing and Economics, and Long Range Planning, among other journals. Dr. Du has received the Alden G. Clayton Award from the Marketing Science Institute and the Jim Kobs Best Papers Award from the Direct Marketing Educational Foundation. His current research interests are in customer relationship management, database marketing, and household budget allocation and portfolio management (for more information, see his home page at http://www.arches.uga.edu/~rexdu).
Wagner A. Kamakura is Ford Motor Company Professor of Global Marketing in the Fuqua School of Business at Duke University. Professor Kamakura has a PhD in Marketing from the University of Texas at Austin, an MS in Industrial Engineering from the University of Sao Paulo (Brazil), and a BS in Mechanical Engineering from the Technological Aeronautics Institute (Brazil). He is a coauthor of Market Segmentation: Conceptual and Methodological Foundations, as well as more than 70 articles in academic journals. His current research interests are in voter choice modeling, customer relationship management, market segmentation and structure, and database marketing.
Carl F. Mela is Professor of Marketing at Duke University. He received his BSc from Brown University, his MBA from University of California, Los Angeles, and his MPhil and PhD from Columbia University. Before Duke, he held management positions at Hewlett-Packard, Hughes, and Proxima. Professor Mela’s research focuses on the long-term effects of marketing activity, customer management, and the Internet. Articles along these lines appear in Journal of Marketing Research, Marketing Science, Journal of Marketing, and Journal of Consumer Research, among other journals. He has received nine best-paper awards from the Marketing Science Institute, INFORMS, the American Marketing Association, and other professional organizations. He serves on the editorial boards of Journal of Marketing, Journal of Marketing Research, Marketing Science, Marketing Letters, Quantitative Marketing and Economics, and Journal of Public Policy & Marketing (for more information, see his home page at http://faculty.fuqua.duke.edu/~mela/bio).
Journal of Marketing, Vol. 71, No. 2, April 2007
View Table of Contents