As interactive systems increasingly involve multiple users with varying attention patterns and cognitive states, the need for precise explanation timing becomes essential. Our research focuses on minimising cognitive workload by strategically identifying when to deliver explanations that are both timely and effective for all users. We employ a Markov Decision Process (MDP) framework and use backward Bellman induction to calculate the optimal timing for explanation delivery. This paper utilises a game-theoretic approach to optimising the timing of explanations in multi-user systems, guided by the psychologically validated SEEV (Salience, Effort, Expectancy, Value) model. The resulting strategy aims to reduce unnecessary cognitive strain while enhancing user comprehension and trust. The model’s performance is evaluated in several simulated scenarios, demonstrating its ability to adapt explanation timing to different user needs and contexts. Our findings suggest that this approach can significantly improve the user experience in multi-user environments by reducing cognitive load and optimizing information delivery. This work contributes to the broader field of human-computer interaction by providing a scalable method for enhancing multi-user systems through intelligent explanation timing. Future directions involve extending the model to incorporate multi-step explanations for multiple users and expanding its scope to accommodate more complex scenarios.