Background: Recent scientific studies have suggested that climbing/bouldering is effective in alleviating depression when the comparison group was a waitlist control group, even when physical activity and other therapeutic approaches were controlled for. In the present study, we aimed to investigate the effectiveness of a manualised psychotherapeutic bouldering intervention for depressed individuals, compared with an active control group performing physical exercise alone. Methods: In a multicentre randomised controlled intervention trial, 133 outpatients with depression were assigned to either a bouldering psychotherapy (BPT) group or a home-based supervised exercise programme (EP). Severity of depression as the primary outcome was assessed at baseline and directly after a ten-week intervention period using the Montgomery-Åsberg Depression Rating Scale (MADRS). Secondary outcomes included anxiety, coping skills, self-esteem, body image, and interpersonal sensitivity. We applied t-tests to test for differences within the groups (t0 vs. t1) and between the BPT and the EP and a multiple regression analysis with the post-intervention MADRS score as the dependent variable. The robustness of estimates was investigated with a sensitivity analyses. Results: Patients in the BPT group showed a significantly larger decrease in depression scores compared with the EP on the MADRS (drop of 8.4 vs. 3.0 points, p = .002, Cohen's d = 0.55). In the confounder-adjusted regression analyses, group allocation was found to be the only significant predictor of the post-intervention MADRS score (β = − 5.60, p = .001) besides the baseline MADRS score. Further significant differences in change scores between the BPT and the EP were found for anxiety (p = .046, d = 0.35), body image (p = .018, d = 0.42), and global self-esteem (p = .011, d = 0.45).