Designing Demand Information Disclosure in the Presence of Capacity Constraints - A Large-Scale Randomized Field Experiment on a Matching Platform.

Digital platforms provide value-added services to individuals in the realm of romance and dating. A common issue in online dating platforms is that users over-pursue popular prospective dating partners, which leads to low matching rates, particularly for users who are not popular on the platform. Our study seeks design solutions for online dating platforms to effectively address this problem. Specifically, we examine whether and how the disclosure of user demand information increases the matching rate by flattening the highly-skewed long-tail demand distribution. The theoretical tension is that on one hand, demand information is traditionally considered to be a signal for quality, as it enables observational learning, which in turn would exacerbate the demand distribution imbalance; on the other hand, demand information may signal available capacity, which helps balance the distribution. We report a large-scale randomized field experiment on Summer, a large mobile dating platform in China, to evaluate these two theoretical effects by comparing the efficacies of alternative demand information disclosure designs. Our results show that the design of demand information disclosure significantly affects the users’ pursuit of dating partners and their subsequent matching outcomes. In particular, compared to plainly disclosing demand information, the design of demand information in tandem with a capacity cue substantially increases the users’ matching requests for low-demand partners, decrease the users’ pursuit of highdemand partners, and results in a higher number of successful matches. Our study provides theoretical implications for the literature at the intersection of dating platform design and observational learning; it also provides actionable insights for matching platform operators regarding optimally disclosing demand information to users.