Self-criticism increases vulnerability to psychological distress and is linked to various psychopathologies. In psychotherapy, it is typically associated with greater symptom severity, lower adherence, poorer outcomes, and higher dropout. However, evidence from preventive interventions suggests that self-criticism can sometimes act as a facilitator of change, with higher baseline levels predicting greater improvements. This study aims to expand this line of research by examining how self-criticism affects adherence, working alliance, and outcomes within NoiBene, a guided self-help program designed to promote well-being and prevent distress among nonclinical university students. A total of 455 participants (82% female; M = 23.5 years) were included in the study. Before and after the intervention, participants completed measures of self-criticism, psychological well-being, emotional awareness, rumination, worry, perfectionism, psychological inflexibility, and assertiveness. After the intervention, adherence and working alliance were assessed. Following the assessment phase, participants began the NoiBene intervention. Baseline levels of self-criticism were used to categorize participants into three groups: mild, moderate, and severe, based on tertile splits. Most participants concluded the entire NoiBene intervention (90.8%). Contrary to evidence from clinical settings, high self-criticism was not associated with increased dropout or weaker alliance. Furthermore, those with severe self-criticism showed greater improvements in emotion awareness, perfectionism, rumination, and psychological inflexibility. They also started with higher baseline difficulties, indicating both greater need and more room for change. These findings suggest that, in preventive guided self-help contexts, self-criticism does not necessarily hinder engagement or outcomes and may, under certain conditions, function as a catalyst for change. Limitations include the self-selected sample, which may limit generalizability; the lack of qualitative data to understand reasons for dropout; and the assessment of working alliance only at post-intervention, which may not capture its development over time. Implications for tailoring digital preventive interventions and addressing dropout risk are discussed.