Objective: The purpose of the present study was to evaluate the efficacy of emotion-focused therapy (EFT) for adults suffering from social anxiety disorder (SAD). Method: Using a nonconcurrent multiple-baseline design, 12 patients (mean age = 26.75 years, SD = 5.15; 7 males) meeting criteria for SAD were treated with up to 28 sessions of EFT. EFT was based on an empathic relationship, 2-chair work for self-criticism, empty-chair work for unresolved feelings, and focusing. Patients were randomized to wait 4, 8, or 12 weeks between the intake and the first therapy session. Intake assessment included the MINI International Neuropsychiatric Interview (MINI; Sheehan et al. 1998), the clinician-administered Liebowitz Social Anxiety Scale (LSAS; Liebowitz, 1987), and various self-report questionnaires. The LSAS was also administered at the end of the baseline period and at posttreatment. The MINI was administered again at posttreatment. Self-reports were administered throughout the baseline, before each therapy session, and at 6-month and 12-month follow-ups. Results: One patient dropped out prematurely. Of the 11 completers, 7 did not meet criteria for SAD at the end of treatment. Intent-to-treat analysis showed that LSAS scores did not change during baseline, significantly improved during treatment (Cohen’s d = −2.37), and remained improved during follow-up. Mixed regression models showed that SAD symptoms and self-criticism did not change during baseline, significantly improved during treatment, and remained improved during follow-up. Self-reassurance improved significantly during the follow-up phase. Conclusion: This study provides initial evidence supporting the efficacy of EFT for SAD.