“Product Line Design with Uninformed Consumer Preferences: The Implications of Data Collection” by Mr. Zibin Xu
Mr. Zibin Xu
PhD Candidate in Marketing
USC‐Marshall School of Business
With the explosion of data collection technologies, a firm may learn about consumers’ preferences better than consumers themselves, especially when consumers have uninformed preferences ‐ imperfect prior knowledge about their marginal utility for quality. This research examines the welfare implications of behavioral data collection for product line design: Do consumers receive better fittng products or simply have more surplus extracted? When should consumers opt out? Is data collection ever unprofitable? We develop an analytical model of data collection when rational consumers observe noisy private information (data) about their intrinsic preferences. Collecting data in aggregate may create superior information when only the firm can isolate environmental noises that affect consumers’ prior. Consumers’ rational suspicion, however, may prevent the firm from exploiting consumers with uninformed preferences. We find that data collection may be a Pareto optimal outcome for both the firm and every consumer. In addition, the datacollec ting firm may signal to the uninformed consumers with high value by lowering the price of the highquality product while raising the quality of the low‐quality product. This tends to reverse the classic quality distortion in product line design and dominate other signaling strategies using exogenous costs. In other cases, however, data collection may either (1) increase the firm’s ability of second‐degree price discrimination, leading consumers to opt out, or (2) harm the firm due to consumers’ suspicions about being exploited.