This paper develops a theoretical argument linking time and the timing of conflict management efforts to dispute duration. We test competing hypotheses on conflict data drawn from disputes in the post-1945 period. Our analysis demonstrates that the effects of mediation vary substantially over the course of a dispute. Specifically, we note that mediation has a curvilinear relationship with time and the ending of disputes. Mediation efforts that occur soon after disputes begin have the best chance of reducing expected future dispute duration. Following this initial period, subsequent mediation efforts lead to longer rather than shorter disputes. After a long period, mediation again leads to shorter rather than longer disputes. We also find that there should be consistency in the mediators used to manage a conflict rather than shifting personnel to interject new ideas.