Abstract Background A school-based physical activity intervention designed to encourage adolescent girls to be more active was more effective for some participants than for others. We examined whether baseline enjoyment of exercise moderated response to the intervention. Methods Adolescent girls with a low level of baseline activity who participated in a controlled trial of an intervention to promote increased physical activity participation (n = 122) self-reported their enjoyment of exercise and physical activity participation at baseline, mid-way through the intervention, and at the end of the 9-month intervention period. At all three time points, participants also underwent assessments of cardiovascular fitness (VO2peak) and body composition (percent body fat). Repeated measures analysis of variance examined the relationship of baseline enjoyment to change in physical activity, cardiovascular fitness, body composition and enjoyment of exercise. Results A significant three-way interaction between time, baseline enjoyment, and group assignment (p < .01) showed that baseline enjoyment moderated the effect of the intervention on vigorous activity. Within the intervention group, girls with low enjoyment of exercise at baseline increased vigorous activity from pre-to post-intervention, and girls with high baseline enjoyment of exercise showed no pre-post change in vigorous activity. No differences emerged in the comparison group between low-and high-enjoyment girls. Conclusion Adolescent girls responded differently to a physical activity promotion intervention depending on their baseline levels of exercise enjoyment. Girls with low enjoyment of exercise may benefit most from a physical-education based intervention to increase physical activity that targets identified barriers to physical activity among low-active adolescent girls.