About AMA

Executive Summary
In recent years, a new type of firm—customer data intermediaries (CDIs)—has emerged that specializes in collecting customer behavior and demographic data and offers customer-specific marketing services. For example, in the grocery and drugstore markets, Catalina Marketing obtains purchase history data through cooperating retailers and provides targeted coupons on behalf of grocery manufacturers to households that purchase at that particular retailer. Targeted advertising by companies such as DoubleClick and Tacoda Systems on the Internet and targeted direct marketing services by firms such as Abacus B2C Alliance and i-Behavior in the catalog and specialty retailing industry also use prior transaction data from cooperating Web sites and cooperating catalog firms. With widespread adoption of customer relationship management and one-to-one (1:1) marketing, the demand for CDI services continues to grow. Several companies in this industry have revenues in the hundreds of millions of dollars and valuations of more than $1 billion.

Despite their growing economic importance, there is little empirical research addressing strategic issues of concern to this industry. Extant research on this industry tends to be of an "engineering" nature, focusing on how firms should use individual browsing/purchasing data from their customer databases to personalize advertising or promotions. From the CDI perspective, this research develops the technology to create customer-specific marketing services. In contrast to such engineering research, this article focuses on "marketing" problems facing the CDIs. Specifically, the authors ask the research questions, Conditional on the availability of the 1:1 technology, what is the optimal customer and product strategy for the CDI? and Conditional on the customer and product strategy, what price should the CDI charge for the service?

In practice, there is considerable diversity in CDIs' customer and product strategies. Some sell their services on an exclusive basis, and others sell them on a nonexclusive basis. For example, Catalina sells on an exclusive basis to only one grocery manufacturer in a particular category in any given period. In contrast, Abacus and i-Behavior sell on a nonexclusive basis to any catalog marketer or specialty retailer that requests their services. Customer data intermediaries also differ in their outlook toward increasing the accuracy of their targeting services. Catalina voluntarily restricts the length of transaction history used for couponing to a maximum of 65 weeks, and Abacus pools data from more than 1550 catalog marketers/specialty retailers on more than 90 million households and uses data for up to five years on each household in its database.

Can customer CDIs benefit from changing their current customer and product strategies? Currently, there is little research to guide them on what the optimal strategy should be. This article offers an empirical framework to help a CDI arrive at an optimal customer and product strategy. The authors illustrate the framework for a 1:1 coupon service firm, such as Catalina, using data from the ketchup market. This approach can also be applied in other empirical contexts. For example, the framework can be used to help answer whether DoubleClick should sell its targeted advertising services on an exclusive basis or a nonexclusive basis by calibrating the impact of advertising (as opposed to couponing) on the downstream firms' profitability.

The authors find that selling on a nonexclusive basis using the maximum available purchase history data is the most profitable strategy for the CDI in the particular market. Thus, Catalina can increase revenues by changing its current policy of exclusive sales in a category. Furthermore, because 1:1 marketing can increase the retailer's profits from goods sold, a retailer that chooses to enter the data intermediary business will find it optimal to substantially undercut the pure play CDI's prices. Thus, the major threat to CDIs such as Catalina in the future may be from the large retailers. This threat should be salient given that many retailers (e.g., Tesco in United Kingdom, Kroger in United States) are developing their own technologies for offering 1:1 coupons to customers. Furthermore, although the authors find that Catalina would find it more profitable to sell its services to the retailer rather than to manufacturers, it has traditionally focused on manufacturer services because it has limited bargaining power to charge high prices from retailers since the customer data on which its services are predicated on is owned by the retailer.

Biography
Joseph Pancras is Assistant Professor in the Marketing Department at the School of Business, University of Connecticut. He obtained his PhD in Marketing from the Leonard N. Stern School of Business at New York University in 2006. His research interests are in the areas of customer relationship management, choice models, and empirical industrial organization. His research has been published in Journal of Marketing Research and Journal of Service Research.

K. Sudhir is Professor of Marketing at the Yale School of Management. He was previously on the faculty at Stern School of Business at New York University. He received his PhD from Cornell University. His research spans a wide range of markets: automobiles, film, grocery retailing, movies and DVDs, personalization services, banking, and high technology. His research topics of interest include structural models of competition and channels, slotting allowances, customer relationship management, international diffusion, and marketing-mix responsiveness. He received the 2003 Frank Bass Dissertation Paper Award. He was a finalist for the 2001 John Little Best Paper Award in Marketing Science and received an honorable mention for 2001 Best Paper Award in the International Journal of Research in Marketing. He serves on the editorial boards of Journal of Marketing Research and Marketing Science.

Journal of Marketing Research, Vol. XLIV, No. 4, November 2007
View Table of Contents.