Special Session Summary Will You Still Love Me Tomorrow: Dynamic Developments in Service Quality and Customer Retention

Katherine N. Lemon, Duke University
[ to cite ]:
Katherine N. Lemon (1996) ,"Special Session Summary Will You Still Love Me Tomorrow: Dynamic Developments in Service Quality and Customer Retention", in NA - Advances in Consumer Research Volume 23, eds. Kim P. Corfman and John G. Lynch Jr., Provo, UT : Association for Consumer Research, Pages: 11.

Advances in Consumer Research Volume 23, 1996      Page 11



Katherine N. Lemon, Duke University

This special session addressed two questions: (1) How do customers update their perceptions of service quality over time? And, (2), given that customers do change their perceptions of a service over time, how do on-going service relationships (between customers and providers) develop, deteriorate, and, ultimately, end over time? In this session, we sought to bring together divergent research approaches to discuss dynamic models of service relationships, and to work toward a research agenda for service quality and customer retention issues. The three papers address both of these areas. Each incorporates the dynamic nature of the decision processCwhether the customer is assumed to be updating his or her perceptions of the service, or making the decision to continue or disadopt the service. In addition, each of the three papers finds that the traditional, utility maximization model does not fully capture the complexities of the process at hand: past experiences "cloud" perceptions of current experience; consumers take "sunk costs" into effect when evaluating the merits of a service; consumers do not act in a truly "Bayesian-like" fashion when updating service quality perceptions. The three papers direct us to a much richer formulation of models for understanding the formation and updating of service quality perceptions and the customer's decision to continue or disadopt.

In the first paper, "The Quality Double Whammy: The Rich Stay Rich and the Poor Stay Poor," Boulding, Kalra and Staelin investigate the effects of prior expectations on customers' cumulative perceptions of quality. Consumers have been found to exhibit a confirmatory bias in evaluating new data (e.g., Hoch and Ha, 1986). This paradigm suggests that individuals are more likely to select and attend to information that confirms their prior beliefs than information which disconfirms them. Therefore, when evaluating new information about a service (e.g., a new service encounter) the authors find that prior expectations get "double-counted" as customers update perceptions of quality. Unlike previous Bayesian-like models, it appears that individuals form "biased" perceptions of the new experience based upon their prior expectations. The results suggest that customers' prior perceptions influence not only their overall assessments of quality, but also their perceptions of each specific transaction. This double-whammy effect provides insight into why many durable good and service firms find it hard to change customers' perceptions of their quality even after making, by objective standards, significant quality improvements. Thus the model provides a formal explanation for why customers' perceptions are slow to change over time.

The second paper, "A Bayesian Model of Quality and Customer Retention," by Rust, Inman and Zahorik, also examines the process by which customers update their perceptions of service quality. Using a Bayesian framework, the authors derived several propositions regarding the way that consumers form their perceptions of quality, satisfaction and probability of choice and then dynamically update them over time. In contrast to Boulding, Kalra and Staelin, the empirical results suggest that consumers appear to behave in a manner that is largely consistent with the Bayesian framework. On average, they update their perceptions of quality over time and are not overly sensitive to any one particular outcome occasion. Specifically, the results suggest that consumers are sensitive not only to the average performance of a product or service, but also to its variability around this mean. In addition, the authors found that consumers are more likely to rechoose a brand or a service if the brand or service performs as expected, i.e., consumers appear to interpret "meeting expectations" as a favorable outcome. Finally, under certain conditions, the probability of choice is not adversely affected by an outcome that is slightly less than expected. This research adds to our knowledge in the important area of how consumers dynamically update their quality perceptions and their preferences.

Although these first two papers appear to suggest opposing findings, they are actually quite complementary. Boulding, Kalra and Staelin's findings suggest that customer perceptions are very difficult to change when customer attitudes are well formed; therefore, customers do not appear "bayesian." Rust, Inman and Zahorik's results suggest that customers act most like bayesians when they have little or no experience with the product or service category. Results from both papers suggest that the most important time to establish quality perceptions in the minds of customers is at the time when customers have little prior experience with the category.

In the third paper, "The Effect of Cumulative Investment and usage Disconfirmation on Customer Disadoption Decisions," Barnett and Lemon examine specific factors that may impact the inexperienced customer's perceptions and behavioral intentions. The authors hypothesize that customer usage disconfirmation, i.e., whether actual usage of the service is more or less than expected usage of the service, should have a significant impact on customer satisfaction, commitment and likelihood to drop the service. In addition, the customer's perceived total costs or cumulative investment in the service should also impact these measures. The results suggest that providing a cumulative investment "reminder" can moderate the impact of usage disconfirmation on customer perceptions of the service and future behavioral intentions. Specifically, this interaction suggests that if customers are reminded of their overall investment, and actual use of the service is greater than expected use (positive usage disconfirmation), they will be more satisfied, more committed to the service, and less likely to drop the service (compared to the no reminder condition). Alternatively, if customer actual use is less than expected (negative usage disconfirmation), and the cumulative investment reminder is present, customers will be less satisfied, less committed to the service and more likely to drop.

The discussion leader, Don Lehman, tied the three papers together and offered some insightful directions for future research. The discussion focused upon understanding the factors which come into play when consumers update perceptions of service quality, and/or when consumers decide to end relationships with service providers. Additional reserach is warranted to explore the variables that account for heterogeneity in consumers' updating processes. The key managerial insight provided by this session is an understanding of how marketing mix variables can be utilized to (a) improve customer perceptions of service quality, and (b) retain the customer for the long term.