This study assessed the effectiveness of a single intervention targeting work style and a combined intervention targeting work style and physical activity on the recovery from neck and upper limb symptoms. Computer workers with frequent or long-term neck and upper limb symptoms were randomised into the work style group (WS, n = 152), work style and physical activity group (WSPA, n = 156), or usual care group (n = 158). The WS and WSPA group attended six group meetings. All meetings focused on behavioural change with regard to body posture, workplace adjustment, breaks and coping with high work demands (WS and WSPA group) and physical activity (WSPA group). Pain, disability at work, days with symptoms and months without symptoms were measured at baseline and after 6 (T1) and 12 months (T2). Self-reported recovery was assessed at T1/T2. Both interventions were ineffective in improving recovery. The work style intervention but not the combined intervention was effective in reducing all pain measures. These effects were present in the neck/shoulder, not in the arm/wrist/hand. For the neck/shoulder, the work style intervention group also showed an increased recovery-rate. Total physical activity increased in all study groups but no differences between groups were observed. To conclude, a group-based work style intervention focused on behavioural change was effective in improving recovery from neck/shoulder symptoms and reducing pain on the long-term. The combined intervention was ineffective in increasing total physical activity. Therefore we cannot draw conclusions on the effect of increasing physical activity on the recovery from neck and upper limb symptoms. © 2007 International Association for the Study of Pain.