Research
Working
- Clicks or Comments? The quality-quantity trade-off of review systemsMatthew J. H. Murphy2025
I study the optimal design of review systems. A platform, seeking to learn an unknown state, faces a trade-off between the informativeness and frequency of user-generated reviews. Detailed reviews provide more information per review, but reviewers submit them less frequently than simple reviews. I characterize the informational content of review systems, which depends on the precision of each review and the rate at which they are submitted by reviewers. I use this characterization to derive the platform’s optimal review system when reviewer’s signals are imprecise. I apply these results to study how the dispersion of reviewers’ idiosyncratic preferences impacts the optimal binary review system. When reviewers’ preferences are sufficiently dispersed, the optimal binary review is symmetric. When reviewers’ preferences are sufficiently concentrated, the optimal review is asymmetric. Regardless of the level of dispersion, the optimal binary review is preferred to a fully detailed review system if it is submitted at least 3.25 times as frequently.