Social Learning with Bounded Rationality: Negative Reviews Persist under Newest First

2024-06-11

ACM Conference on Economics and Computation

AI Summary

This paper investigates how platform “chronological ordering” distorts social learning when customers have limited attention—reading only the top few reviews—leading to persistent negative reviews dominating the first page (“chronological-ordering cost”) and inducing non-monotonic revenue dynamics, with theoretically unbounded long-term revenue loss. Method: We develop a game-theoretic model integrating Bayesian belief updating, strategic review timing, and dynamic pricing optimization under attention constraints. Contribution/Results: We formally define and characterize the chronological-ordering cost; prove that the optimal pricing policy sustains a constant purchase probability across time, thereby bounding revenue loss within a factor of two relative to the first-best benchmark. Empirical validation using TripAdvisor data from 24 hotels shows that, for 22 hotels, the mean rating of the top 10 reviews is significantly lower than the global mean (average deviation: −0.15 on a 5-point scale), confirming the theoretical prediction of systematic downward bias in visible reviews.

Technology Category

• Search and Optimization: Learning to Search
• Machine Learning: Learning Preferences or Rankings
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: The sharing economy
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience

Abstract

The use of product reviews in online platforms is ubiquitous and it is well established that reviews play a significant role on customer purchase decisions. The process in which reviews impact product purchases can be seen as a problem of social learning, which generically studies how agents update their beliefs for an unknown quantity of interest (e.g., product quality) based on observing actions of past agents (e.g., reading reviews by past customers). The typical assumption in the literature of social learning with reviews is that, when deciding whether to purchase a product, customers consider either all reviews provided by previous customers or a summary statistic such as their average rating. However, in practice, a common scenario may be somewhere "in between" the above two assumptions: customers read a small number of reviews in detail. So motivated, we study a model of social learning from reviews where customers are computationally limited and make purchases based on reading only the first few reviews displayed by the platform. Under this bounded rationality, we establish that the review ordering policy can have a significant impact. In particular, the popular Newest First ordering induces a negative review to persist as the most recent review longer than a positive review. This phenomenon, which we term the Cost of Newest First, can make the long-term revenue unboundedly lower than a counterpart where reviews are exogenously drawn for each customer. We show that the impact of the Cost of Newest First can be mitigated under dynamic pricing, which allows the price to depend on the set of displayed reviews. Under the optimal dynamic pricing policy, the revenue loss is at most a factor of 2. On the way, we identify a structural property for this optimal dynamic pricing: the prices should ensure that the probability of a purchase is always the same, regardless of the state of reviews. We also study an extension of the model where customers tend to trust more recent reviews (and discount older reviews based on their time of posting), which may contribute to why Newest First is used in practice. We show that Newest First is still not the optimal ordering policy if customers discount slowly, and we uncover an unexpected non-monotonicity in revenue with respect to the customers' discount factor. Finally, we corroborate our theoretical findings using real-world review data from Tripadvisor, an online platform where the default review ordering policy is newest first. We evaluate 24 hotel pages, and we find that for 22 of them, the average review rating of the first 10 reviews is lower than the average review rating of all reviews. The magnitude of this discrepancy is 4% on average, or 0.15 points out of 5 in absolute terms. A full version of this paper can be found at: https://arxiv.org/abs/2406.06929.

Problem

• Impact of review ordering on social learning with limited attention
• Negative review persistence under Newest First ordering reduces revenue
• Dynamic pricing mitigates revenue loss from review ordering effects

Innovation

• Limited attention model for social learning
• Dynamic pricing mitigates negative review persistence
• Newest First reduces revenue despite tracking accuracy