Optimal paid promotion strategy for online creators in recommender systems
摘要
This study examines the strategic impact of paid promotion services on user-generated content (UGC) platforms that utilize recommender systems (RS). We develop a game-theoretic model to analyze the interaction between algorithmic recommendations and paid promotion, considering both monopolistic and duopoly market structures. Our analysis reveals a fundamental asymmetry driven by systematic RS bias: the tendency to overestimate high-quality video and underestimate low-quality cvideo. This bias makes paid promotion services a strictly dominated strategy for high-quality creators, who benefit sufficiently from organic recommendations. In contrast, paid promotion services become strategically essential for low-quality creators to counteract the RS underestimation effect. A counterintuitive finding is that improvements in RS accuracy can, within a specific range, incentivize low-quality creators to increase their promotional intensity to better compete for users mismatched by the system. We demonstrate that paid promotion services are a double-edged sword. A win–win outcome for the platform and creator is achievable only if the unit paid promotion cost is maintained below a critical threshold. Otherwise, it risks harming the ecosystem. The key managerial insight is that platforms must carefully design pricing for these services, while creators should adopt asymmetric strategies. That is, high-quality creators should focus on organic discovery, whereas low-quality creators should use paid promotion services, adjusting their intensity based on the quality gap and RS accuracy.