About AMA

Executive Summary
In many services (e.g., the wireless service industry), consumers choose a service plan on the basis of their expected consumption. In such situations, consumers experience two forms of uncertainty. First, consumers may be uncertain about the quality of their service provider and can learn about it after repeated use of the service. Second, consumers can be uncertain about their own usage of minutes and learn about it after observing their actual consumption.

Service contexts have another interesting aspect—namely, the presence of nonlinear pricing schemes. Within the wireless industry, pricing schemes are typically characterized by an access fee, included free minutes, and a per-minute marginal price for any consumption in excess of the free minutes. Such pricing schemes are termed "increasing block" because the applicable marginal price increases with consumption. Although the authors focus on the wireless industry, increasing-block schemes are also used for other services. For example, in the electricity and water supply industry, the total charges payable by consumers are based on their consumption in the billing period, and the applied per-unit rates typically increase with increasing consumption. These kinds of pricing schemes create a simultaneity between price and consumption; that is, the applicable marginal price depends on consumption, and vice versa.

In summary, the authors consider three aspects—consumer uncertainty and learning of a service provider's quality, consumer uncertainty and learning of their own consumption patterns, and nonlinear pricing that creates a simultaneity between price and quantity. Although previous research has considered some of these issues, the authors address all three aspects together. They propose a model within a Bayesian learning framework. They apply it to customer-level monthly billing data from a single wireless service provider and use hierarchial Bayesian methods to estimate the model. In this data set, the authors observe consumers' choice of plans, their consumption of minutes, and whether they decide to leave the service provider.

The authors find that both quality learning and quantity learning are important aspects of the model. In addition, in the application, consumers learn about service quality rapidly. More than 90% of quality learning occurs within the first five service encounters. This suggests that firms need to manage the first few service encounters strategically. The authors then use policy experiments to investigate the effects of consumer learning and find that consumer learning can result in a win–win situation for both consumers and the firm; specifically, consumers leave fewer minutes on the table, and the firm achieves an increase in overall customer lifetime value (CLV). In particular, the authors estimate that there is approximately a 35% increase in CLV (approximately $75) in the presence of consumer learning. The key driver of this difference is the change in the retention rate with and without consumer learning. The authors also perform simulations that relate service quality to CLV and determine that, on average, a 1% increase in mean service quality leads to approximately a $2.00 increase in CLV. Because the service provider has 21 million customers, this increase in mean quality results in an overall long-term increase in profit of approximately $42 million dollars. Finally, policy experiments related to pricing show that changes in access fee have a significantly greater influence on the CLV of "light users."

Biography
Raghuram Iyengar is Assistant Professor of Marketing in the Wharton School of the University of Pennsylvania. He has an undergraduate degree from I.I.T. Kanpur, India, and a PhD in Marketing from Columbia University. Professor Iyengar's research interests are in pricing, structural models, and Bayesian methods.

Asim Ansari is Professor of Marketing in the Columbia Business School at Columbia University. He has an undergraduate degree from Osmania University, India; an MBA from I.I.M., Bangalore, India; and a PhD in Marketing from New York University. Professor Ansari's research interests are in the area of customer relationship management, mass customization, and Bayesian modeling of marketing decisions.

Sunil Gupta is Edward W. Carter Professor of Business Administration in the Harvard Business School at Harvard University. He has an undergraduate degree from I.I.T. Delhi; an MBA from I.I.M. Ahmedabad, India; and a PhD in Marketing from Columbia University. Professor Gupta's research interests are in the area of marketing strategy, pricing, and customer relationship management.

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