Algorithmic Attention and Content Creation on Social Media Platforms*

Abstract

We study the revenue-maximizing allocation of attention on an ad-funded social media platform governed by content recommendation algorithms. Attention is costly and serves a dual role: it can be directly monetized through advertising, or it can be allocated to creators to increase exposure, strengthen effort incentives, and thereby generate more attention thanks to better content. This tension creates a trade-off between monetization and production incentives. In a two-sided model with heterogeneous viewers and creators under private information, the optimal recommendation mix includes content that is ex post suboptimal for some viewers in order to leverage cross-side network externalities. Two-sided complementarities amplify informational distortions and overturn the standard no-distortion-at-the-top intuition. When advertising markets are strong, monetary transfers substitute for attention-based incentives.

Keywords

JEL codes