Internet-based problem-solving training (IPST) effectively reduces depressive symptoms in employees. Yet, it is unknown which employees benefit most from this particular treatment. The study aimed to identify predictors and moderators of treatment outcome in IPST offered to employees with depressive symptoms. Within a randomized controlled trial (N. =. 150), designed to test the effectiveness of IPST, variables that predict and moderate the effects of IPST when compared with a waitlist control group (WLC) were explored. The outcome was change in depression severity, assessed using the Center for Epidemiological Studies Depression Scale (CES-D). Both depression severity and other psychopathological symptoms and potential predictors/moderators were assessed as self-reports at baseline (t1) and in follow-up assessments after seven weeks (t2), three months (t3) and six months (t4). Higher depression severity at baseline predicted improvement in depressive symptomology in follow-up assessments after seven weeks, and three- and six months. Depression severity moderated the effectiveness of IPST assessed at six-month follow-up. Simple slope analyses revealed that the long-term effectiveness of the intervention was more pronounced among participants with high (CES-D range: 33-44, M. =. 37.0, SD. =. 3.2) and moderate (CES-D range: 14-32, M. =. 23.1, SD. =. 5.6) depression baseline scores, compared to participants displaying low depression baseline scores (CES-D range: 5-13, M. =. 9.0, SD. =. 2.2). No indication was found that participants presenting low depression severity at baseline significantly benefitted from IPST in the long-term. IPST might be appropriate for employees with a wide range of different characteristics. While there appears to be no reason to exclude employees with severe depression from Internet-based occupational mental health interventions, for employees low in depression severity, watchful waiting or potentially no intervention should be considered. These findings may not apply to other low-intensity interventions and/or target groups.