What you don't know about customer-perceived quality: The role of customerexpectation distributions

We show that some of the most common beliefs about customer-perceived quali ty are wrong. For example, 1) it is not necessary to exceed customer expect ations to increase preference, 2) receiving an expected level of bad servic e does not reduce preference, 3) rational customers may rationally choose a n option with lower expected quality, even if all non-quality attributes ar e equal, and 4) paying more attention to loyal, experienced customers can s ometimes be counterproductive. These surprising findings make sense in retr ospect, once customer expectations are viewed as distributions, rather than simple point expectations. That is, each customer has a probability densit y function that describes the relative likelihood that a particular quality outcome will be experienced. Customers form these expectation distribution s based on their cumulative experience with the good or service. A customer 's cumulative expectation distribution may be conceptualized as being a pre dictive density for the next transaction. When combined with a diminishing returns (i.e., concave) utility function, this Bayesian theoretical framework results in predictions of: (a) how cons umers will behave over time, and (b) how their perceptions and evaluations will change. In managerial terms, we conclude that customers consider not o nly expected quality, but also risk. This may help explain why current meas ures of customer satisfaction (which is highly related to expected quality) only partially predict future behavior. We find that most of the predictio ns of our theoretical model are borne out by empirical evidence from two ex periments. Thus, we conclude that our approach provides a useful simplifica tion of reality that successfully predicts many aspects of the dynamics of consumer response to quality. These findings are relevant to both academics and managers. Academics in th e area of customer satisfaction and service quality need to be aware that i t may be insufficient to measure only the point expectation, as has always been the standard practice. Instead it may be necessary to measure the unce rtainty that the customer has with respect to the level of service that wil l be received. Due to questionnaire length constraints, it may not be pract ical for managers to include uncertainty questions on customer satisfaction surveys. Nevertheless it is possible to build a proxy for uncertainty by m easuring the extent of experience with the service/good, and this proxy can be used to partially control for uncertainty effects. The findings of the study were obtained using 1) an analytical model of cus tomer expectation updating, based on a set of assumptions that are well-sup ported in the academic literature, and 2) two behavioral experiments using human subjects: a cross-sectional experiment, and a longitudinal experiment . Both the analytical model and the behavioral experiments were designed to investigate the effects that distributions of expectations might have, and especially the effects that might deviate from the predictions that would arise from a traditional point expectation model. The behavioral experiment s largely confirmed the predictions of the analytical model. As it turned o ut, the analytical model correctly (in most cases) predicted behavioral eff ects that contradict some of the best-accepted "truisms" of customer satisf action. It is now clear that a more sophisticated view of customer expectations is required-one that considers not only the point expectation but also the lik elihood across the entire distribution of possible outcomes. This distincti on is not "just academic," because it results in predictable behavior that deviates significantly from that which was traditionally expected based on simpler models.