Time to Leave Your Comfort Zone? Optimal Variation-Seeking Strategies for Social Media Influencers on Streaming Media Platforms

Abstract

Social media influencers (SMIs) face intense competition in the era of streaming media. To win and retain large audiences in the long run, SMIs need to step out of their comfort zone to seek variation in their media content style and repertoire. Our study is centered around developing effective variation-seeking strategies for SMIs on streaming media professionally generated content (PGC) platforms. In particular, we propose a finite-horizon Markov Decision Process (MDP) model that captures the inter-temporal dynamics between the SMIs and the audiences. This computational model helps individual SMI determine the optimal variation-seeking strategy, including when and how to change her performance style and repertoire during the streaming season. To calibrate the parameters of the MDP model, we leverage real-time data on a major streaming media PGC platform in China and perform state-of-the-art matrix completion techniques. Our findings reveal that the optimal variation-seeking policy is state-dependent, where the state includes the SMI intrinsic type and the remaining time horizon. In particular, we identify a subset of SMI types that should take variation-seeking actions during the streaming season and characterize their degrees of variation-seeking actions. Moreover, we find that the variation-seeking action (should it be taken) is encouraged earlier than later. Our study represents the first effort in the literature to capture the inter-temporal dynamics between the SMIs and the audiences on streaming media PGC platforms and make recommendations to SMIs on the individual level.