Using a scrambled sentence priming protocol, the first aim of this study was to investigate whether differences in novel exercise-related goals existed between participants primed with motivational or non-motivational material (i.e. autonomous, controlled or neutral primes). The second aim was to explore whether an indirect effect was present between priming condition and the goal-related exercise sessions that individuals performed over the week following administration of the prime. No effects were observed across priming conditions with respect to subjective vitality, goal concordance and the frequency with which participants planned to exercise. However, autonomy-primed individuals set goals for their exercise sessions that were significantly longer in intended duration (M = 57.72 min, SD = 37.42) than those set by their counterparts in both the controlled-prime condition (M = 44.12 min, SD = 27.78) and the neutral-prime condition (M = 37.10 min, SD = 20.47). Bootstrapped analyses also revealed a significant indirect relationship between prime and exercise behaviour, with the autonomy prime predicting longer goal-based exercise sessions, via the effect on the duration of participants' intended exercise sessions. These findings highlight the potential influence that priming autonomous motivation may have on individuals' exercise aspirations, as well as the way in which primes may indirectly shape exercise engagement.