Two studies tested the efficacy of Marlatt and Gordon's relapse-prevention approach in increasing attendance during an exercise program (short-term adherence) and continuation of exercise activities for 12 weeks following termination of the formal program (longer term adherence). Participants in both studies were registrants in 10-week exercise groups (jogging, aerobic dance, and pre-ski training) sponsored by the Université de Montréal Sports Centre. The intervention, designed to increase awareness of obstacles to exercise and to develop appropriate techniques for coping with them, was delivered by group leaders within the context of the regular program. Results of both studies indicate a small but consistent superiority of adherence in the experimental condition compared to the control condition. The low cost of this intervention, however, makes even small gains cost effective. Possible methods for strengthening the treatment effect are discussed.