Personalized Pricing via Strategic Learning of Buyers’ Social Interactions

Abstract

As the sociological theory of homophily suggests, people tend to interact with those of similar preferences. This motivates product sellers to learn buyers’ product preferences from the buyers’ friends’ purchase records. Although such learning allows sellers to enable personalized pricing to improve profits, buyers are also increasingly aware of such practices and may alter their behaviors accordingly. This paper presents the first study regarding how buyers may strategically manipulate their social interaction signals considering their preference correlations, and how an informed seller can take buyers’ strategic social behaviors into consideration when designing the pricing schemes. Our analytical results show that only high-preference buyers tend to manipulate their social interactions to hurdle the seller’s personalized pricing. Surprisingly, these high-preference buyers’ payoff may become worse after their strategic manipulation. Furthermore, we show that the seller can greatly benefit from the learning practice, no matter whether the buyers are aware of such learning or not. In fact, buyers’ learning-aware strategic manipulation only slightly reduces the seller’s revenue. Considering the increasingly stricter policies on data access by authorities, it is thus advisable for sellers to make buyers aware of their access and learning based on social interaction data. This justifies well with current regulatory policies and industry practices regarding informed consent for data sharing.