Designing Behaviourally Informed Market Mechanisms in the Streaming Industry: Efficiency, Engagement, and Fairness in Two-Sided Platforms

Streaming platforms rely heavily on ad auctions and recommendation algorithms to allocate attention between users, advertisers, and creators. Traditional market design approaches assume rational, quasi-linear agents, yet user behaviour in digital environments systematically deviates from these assumptions due to cognitive limits, loss aversion, and engagement fatigue. This paper examines how integrating behavioural economics with market design can improve efficiency, engagement, and fairness in streaming ecosystems. Using examples from ad-supported platforms such as YouTube and subscription-driven services like Netflix and Spotify, the study highlights how Generalized Second-Price auctions and multi-layered recommendation systems may generate unintended distortions, including advertising weariness, algorithmic lock-in, and creator bias. It proposes constrained optimisation frameworks that incorporate user-specific tolerance thresholds, lifetime value considerations, and diversity-aware ranking rules. The findings suggest that allocation mechanisms grounded in both mathematical optimisation and behavioural realism are better positioned to sustain long-term platform welfare.