Purpose: To increase exercise adherence among insufficiently active adult employees. Design: A quasi-experimental separate samples pre-test-post-test group design was used to compare treatment and comparison group. Setting: The worksite. Subjects: Employees (n = 127) who did not meet current American College of Sports Medicine (ACSM) recommendations for exercise. Intervention: An eight-week educational programme targeting the social cognitive theory constructs. Measures: Free-living exercise, self-regulation, self-efficacy, social support, and outcome expectations and expectancies. Measurement was at pre-test, post-test, one month and three months post-intervention. Analysis: One-way analysis of variance. Results: The treatment group reported important increases in exercise and mediators of exercise. There was a significant difference between groups for moderate intensity exercise and vigorous intensity exercise at post-test and follow-up (p = 0.001). There was a significant difference between groups for self-regulation at post-test and follow-up (p = 0.001). There was not a significant difference between groups for self-efficacy or outcome expectancies. Family and friend social support group differences were non-significant at post-test and one-month follow-up, but was significant difference at three-month follow-up (p = 0.001). Outcome expectations and expectancies were non-significant at all time periods. Conclusion: The educational intervention was effective in increasing the exercise rates of employees at the worksite. (Contains 3 tables.)