Background Psychosocial interventions for bipolar disorder are frequently unavailable and resource intensive. Mobile technology may improve access to evidence-based interventions and may increase their efficacy. We evaluated the feasibility, acceptability and efficacy of an augmentative mobile ecological momentary intervention targeting self-management of mood symptoms. Methods This was a randomized single-blind controlled trial with 82 consumers diagnosed with bipolar disorder who completed a four-session psychoeducational intervention and were assigned to 10 weeks of either: 1) mobile device delivered interactive intervention linking patient-reported mood states with personalized self-management strategies, or 2) paper-and-pencil mood monitoring. Participants were assessed at baseline, 6 weeks (mid-point), 12 weeks (post-treatment), and 24 weeks (follow up) with clinician-rated depression and mania scales and self-reported functioning. Results Retention at 12 weeks was 93% and both conditions were associated with high satisfaction. Compared to the paper-and-pencil condition, participants in the augmented mobile intervention condition showed significantly greater reductions in depressive symptoms at 6 and 12 weeks (Cohen's d for both were d=0.48). However, these effects were not maintained at 24-week follow up. Conditions did not differ significantly in the impact on manic symptoms or functional impairment. Limitations This was not a definitive trial and was not powered to detect moderators and mediators. Conclusions Automated mobile-phone intervention is feasible, acceptable, and may enhance the impact of brief psychoeducation on depressive symptoms in bipolar disorder. However, sustainment of gains from symptom self-management mobile interventions, once stopped, may be limited.