Abstract Purpose To improve the long-term physical and psychological health of prostate and colorectal cancer survivors, physical activity (PA) interventions aimed at maintaining PA are necessary. OncoActive is a print- and web-based intervention that was was able to increase PA of prostate and colorectal cancer patients and survirors shortly after the intervention ended. This study evaluated the long-term efficacy (12-month follow-up) comparing OncoActive and a control group. Furthermore, we explored whether the increased PA was maintained after an 8 month non-intervention follow-up period. Methods Prostate or colorectal cancer patients were randomly assigned to OncoActive (N = 249) or a usual care control group (N = 229). OncoActive participants received 3 times PA advice and a pedometer. PA outcomes (i.e., ActiGraph and self-report MVPA min/week and days with ≥ 30 min PA) and health-related outcomes (i.e., fatigue, depression, physical functioning) that were significantly changed by the intervention at 6-month follow-up, were included. Differences between groups and changes over time were assessed with multilevel linear regressions. Results At 12 months, OncoActive participants did not perform better than control group participants at ActiGraph MVPA min/week, self-report MVPA min/week or ActiGraph days with PA. Only self-report days with PA were significantly higher in OncoActive compared to the control group. Fatigue was the only health-related outcome that was significantly lower in OncoActive compared to the control group at 12-month follow-up. When exploratively looking within the OncoActive group, the previously found PA increase shortly after the intervention (6 months follow-up) was maintained at 12 months and PA improved significantly from baseline to 12 months. The control group showed small but non-significant improvements from 6 to 12 resulting in a decline of differences between groups. Conclusion At long-term follow-up (i.e., 12-months), most significant differences between OncoActive and the control group at short-term follow-up (i.e., 6 months) were no longer significant. Natural improvements in the control group may have contributed to this. Fatigue was significantly lower in OncoActive compared to control group participants at 12 months. Computer-tailored PA advice may give participants an early start to recovery and potentially contributes to improving long-term health. Funding KWF Kankerbestrijding (Dutch cancer society) NOU2012-5585