Incentivizing High Quality Entrants When Creators Are Strategic (2509.14102v1)

Abstract: We study how a platform should design early exposure and rewards when creators strategically choose quality before release. A short testing window with a pass/fail bar induces a pass probability, the slope of which is the key sufficient statistic for incentives. We derive three main results. First, a closed-form ``implementability bounty'' can perfectly align creator and platform objectives, correcting for incomplete revenue sharing. Second, front-loading guaranteed impressions is the most effective way to strengthen incentives for a given attention budget. Third, when impression and cash budgets are constrained, the optimal policy follows an equal-marginal-value rule based on the prize spread and certain exposure. We map realistic ranking engines (e.g., Thompson sampling) into the model's parameters and provide telemetry-based estimators. The framework is simple to operationalize and offers a direct, managerially interpretable solution for platforms to solve the creator cold-start problem and cultivate high-quality supply.