Objectives: Goal-setting is one of the most common strategies used to increase physical activity. Current practice is often based on specific, measurable, achievable, realistic, and time-bound (SMART) goals. However, theory and research suggests that this approach may be problematic. Open goals (e.g., “see how well you can do”) have emerged as a possible alternative, but are yet to be tested experimentally in physically active tasks. In a walking-based session, this study aimed to experimentally compare the effects of open, SMART and do-your-best goals with a control condition on distance walked and psychological variables related to engagement. Design: Repeated measures design (mixed model). Method: Participants (N = 78; Mage = 55.88) were randomly assigned to one of four goal conditions: an open, SMART, or do-your-best goal, or a control condition (“walk at your normal pace”), before completing a baseline and two manipulated attempts of a 6-min walking test. Results: Open, SMART, and do-your-best goals achieved greater distance walked, and higher ratings of perceived exertion, than the control across both experimental attempts. Open and SMART goals led to greater enjoyment of the session. However, SMART goals led to higher pressure/tension, while open goals led to higher perceptions of performance and higher interest in repeating the session. Conclusions: These findings provide preliminary evidence for the efficacy of setting open goals in physical activity, and suggest that they may be more psychologically adaptive to pursue than SMART or do-your-best goals. Implications are discussed, and recommendations are made for future goal-setting research in physical activity.