#99-11
"Customer Learning and Loyalty When Quality is Uncertain"
Abstract: A consumer has repeated contacts with a set of product or service providers. Each visit to a supplier yields the consumer some randomly distributed utility. The suppliers' utility distributions are unknown to the consumer, and to decide which supplier to visit, she uses a myopic variant of the decision rule used by a classical, utility-maximizing Bayesian. This rule is designed to be roughly consistent with empirical findings regarding individual choice under uncertainty. For this model, we develop closed-form expressions that characterize both short-term and long-term measures of customer loyalty to a supplier. These results offer a rich picture of how consumer discrimination and prior beliefs interact with the level of quality actually offered by suppliers to determine customer loyalty.
Download the paper.