The verbal content of interactions (what was said and who said what) can be important as evidence and intelligence information. Across 3 empirical studies, we examined memory for details of an overheard (Experiment 1) or witnessed (Experiments 2 and 3) conversation using a timeline technique adapted for the reporting of conversations between multiple speakers. Although participants in all conditions received the same general instructions, participants assigned to timeline reporting format reported more verbatim information and made fewer sequencing errors than those using a free recall format. In Experiments 2 and 3, using an extended version of the technique, participants using the timeline reporting format also reported more correct speaker attributions and provided more information about the individuals involved, without compromising overall accuracy rates. With a large effect size across experiments (total N = 134), these findings suggest that timeline reporting formats facilitate the reporting of episodic memories and benefit the reporting of conversations.