Commitment on Volunteer Crowdsourcing Platforms: Implications for Growth and Engagement

Volunteer crowdsourcing platforms, such as food recovery organizations, match volunteers with tasks which are often recurring. To ensure completion of such tasks, platforms frequently use a commitment lever known as "adoption." Despite being effective in reducing match uncertainty, high levels of adoption reduce match availability for volunteers which in turn can suppress future engagement. We study how platforms should balance these two opposing factors. Our research is motivated by a collaboration with Food Rescue U.S. (FRUS), a volunteer-based food recovery organization active in over 33 locations across the U.S. For platforms such as FRUS, success crucially depends on efficient volunteer utilization and engagement. Consequently, effectively utilizing non-monetary levers, such as adoption, is critical. Based on our analysis of fine-grained FRUS data, we develop a model for a repeated two-sided matching market consisting of tasks (prearranged donations) and volunteers. Our model incorporates the uncertainty in match compatibility as well as the negative impact of failing to match on future engagement. We study the platform's optimal policy for setting the adoption level to maximize the total discounted number of matches. Our analysis reveals that the optimal myopic policy is either full or no adoption. For sufficiently thick markets, we show that such a myopic policy is also optimal in the long run. In thinner markets, even though a static policy of full or no adoption can be suboptimal, we show it achieves a constant-factor approximation where the factor improves with market thickness. Using our analytical and empirical results, we revisit the current design of the FRUS platform and make location-specific policy recommendations. More broadly, our work sheds light on how other two-sided platforms can control the double-edged impacts that commitment levers have on growth and engagement.