Background and objectives Cognitive behavioral therapy (CBT) and acceptance and commitment therapy (ACT) have both garnered empirical support for the effective treatment of social anxiety disorder. However, not every patient benefits equally from either treatment. Identifying moderators of treatment outcome can help to better understand which treatment is best suited for a particular patient. Methods Forty-nine individuals who met criteria for social anxiety disorder were assessed as part of a randomized controlled trial comparing 12 weeks of CBT and ACT. Pre-treatment avoidance of social situations (measured via a public speaking task and clinician rating) was investigated as a moderator of post-treatment, 6-month follow-up, and 12-month follow-up social anxiety symptoms, stress reactivity, and quality of life. Results Public speaking avoidance was found to be a robust moderator of outcome measures, with more avoidant individuals generally benefitting more from CBT than ACT by 12-month follow-up. In contrast, clinician-rated social avoidance was not found to be a significant moderator of any outcome measure. Limitations Results were found only at 12-month follow-up. More comprehensive measures of avoidance would be useful for the field moving forward. Conclusions Findings inform personalized medicine, suggesting that social avoidance measured behaviorally via a public speaking task may be a more robust factor in treatment prescription compared to clinician-rated social avoidance.