Depression is common but undertreated. Web-based self-help provides a widely accessible treatment alternative for mild to moderate depression. However, the lack of therapist guidance may limit its efficacy. The authors assess the efficacy of therapist-guided web-based cognitive behavioural treatment (web-CBT) of mild to moderate depression. Fifty-four individuals with chronic, moderate depression participated in a randomized wait-list controlled trial, with an 18-month follow-up (immediate treatment: n = 36, wait-list control: n = 18). Primary outcome measures were the Beck Depression Inventory (BDI-IA) and the Depression scale of the Symptom Checklist-90-Revised (SCL-90-R. DEP). Secondary outcome measures were the Depression Anxiety Stress Scales and the Well-Being Questionnaire. Five participants (9%) dropped out. Intention-to-treat analyses of covariance revealed that participants in the treatment condition improved significantly more than those in the wait-list control condition (.011 < p < .015). With regard to the primary measures, between-group effects (d) were 0.7 for the BDI-IA and 1.1 for the SCL-90-R DEP. Posttest SCL-90- R DEP scores indicated recovery of 49% of the participants in the treatment group compared with 6% in the control group (odds ratio = 14.5; p < .004). On average, the effects were stable up to 18 months (n = 39), although medication was a strong predictor of relapse. The results demonstrate the efficacy of web-CBT for mild to moderate depression and the importance of therapist guidance in psychological interventions.