Research on training therapists to deliver discrete trial teaching (DTT) has tended to focus on a limited range of therapist competencies and may have neglected important variables such as the rate and variation of trial types delivered across complete teaching sessions. Stimulus prompting procedures may facilitate the delivery of DTT for the broad range of competencies needed during therapy. In the present study, three experienced therapists were taught to deliver DTT with and without a stimulus prompt. A multiple baseline across participants with embedded reversal design was used to analyze performance during simulated teaching sessions. Results showed that all participants delivered both a higher rate of accurate trials and a greater variety of trials in sessions that incorporated stimulus prompting.