Aim: Patient-reported outcome measures (PROMs) are increasingly important as a mean for quality assurance. Feasible estimates of recovery can be achieved through the application of Jacobson plots, which is a simple demonstration of the outcome of each case, recommended for clinical use. We applied this approach with PROMs collected regarding group psychotherapy in a mental health service (MHS) setting. We hypothesized a recovery rate of above 50% of all cases, expecting a lower recovery rate amongst patients with severe depression. Methods: We made a secondary sub-sample analysis of data from patients with unipolar depression (N = 171) within a pragmatic, non-inferiority, randomized controlled clinical trial comparing two cognitive behavior therapy (CBT) group interventions. The treatment consisted of 14 2-hours weekly group CBT sessions. We collected depression PROMs with the Becks Depression Inventory-II and functional levels PROMs with the Work and Social Adjustment Scale at baseline, end-of-treatment, and at a 6-months follow-up. Results: At follow-up, 35% (N = 43/123) of cases with moderate or severe depression (BDI > 19) at baseline reached scores below the cut-off for moderate depression. Recovery rates in severe cases were significantly lower (26.5 vs 52.5%; p = 0.0004). We observed severe functional impairment in 36% of the patients at baseline (52/144) and observed no changes or worsening in scores at a 6-months follow-up in 44% of the patients (64/144). Conclusion: We achieved satisfactory remission rates for patients with moderate depression. Patients with severe depression and patients with functional impairment reached recovery rates below the standard of comparable MHSs. Improved MHSs for these patients are needed.