Selling Consumer Data for Profit: Optimal ­Market-Segmentation Design and Its Consequences

Abstract

A data broker sells market segmentations to a producer with private cost who sells a product to a unit mass of consumers. This paper characterizes the revenue-maximizing mechanisms for the data broker. Every optimal mechanism induces quasi-perfect price discrimination. All the consumers with values above a cost-dependent cutoff buy by paying their values while the rest of consumers do not buy. The characterization implies that market outcomes remain unchanged even if the data broker becomes more powerful—either by gaining the ability to sell access to consumers or by becoming a retailer who purchases the product and sells to the consumers exclusively. (JEL D42, D82, D83, L81, M31)

Data, Competition, and Digital Platforms

“…At a broad level, this paper relates to information structures in advertising auctions, e.g., Bergemann et al (2021), and to nonlinear pricing, market segmentation, and competition, e.g., Bergemann et al (2015), Bonatti (2011), Elliott et al (2020), and Yang (2022. Finally, our analysis can be easily extended to discuss self-preferencing by a monopoly platform.…”

Selling Data to a Competitor

“…The literature on the sale of data by a data provider (which includes, among others, the papers of Admati and Pfleiderer, 1986;Bergemann et al, 2018;Montes et al, 2019;Segura-Rodriguez, 2021;Yang, 2022) studies how a third-party data-provider can maximize profits by selling data to a monopolist or to competing firms who use this data to price discriminate. Within this literature, one paper that is closely related to ours is that of Elliott et al (2021).…”

Data Collection by an Informed Seller

“…Our mechanism-design approach to study equilibrium outcomes follows Myerson (1983), and our seller can be viewed as his informed principal. 1 The mechanism-design machinery has recently been used in the literature on information provision by Kolotilin, Mylovanov, Zapechelnyuk, and Li (2017), Bergemann, Bonatti, and Smolin (2018), Smolin (2020), andYang (2022). The flexible way of designing information collection follows the Bayesian persuasion literature (e.g., Rayo and Segal (2010) and Kamenica and Gentzkow (2011)).…”