The Impact of Online Consumer Reviews on Online Sales: The Case-Based Decision Theory Approach

Abstract

In this paper, we conduct a two-stage study to examine whether and how consumer reviews influence other consumers. In the first stage, we utilize a difference-in-difference specification to identify the causal effect of online reviews on sales and apply the proposed model to electronic appliances on a large Chinese retail website. In the second stage, we design a computational model with agents guided by case-based decision theory and we calibrate the simulated market to the real data. The literature identifies two channels for the effect of reviews on sales: the awareness effect and the persuasive effect. We find evidence for both effects, but we find the persuasive effect is slightly larger than the awareness effect. This suggests that consumers who have had a bad experience should not hesitate to leave bad reviews. We also find evidence that consumers are concerned whether reviews are genuine, but that the method used by this website of tying a “user grade” to volume of previous purchases may be an effective way of communicating whether a reviewer is a genuine customer. We use the computational model to predict unobserved consumer behavior: consumers’ loyalty. We find the loyalty of artificial consumers is relatively high but falls. We also believe the case-based decision theory simulation approach may help estimate other unobserved consumer behaviors.