Background Depression often starts in adolescence making it an ideal time to intervene. We developed a universal cognitive behavioural therapy‐based programme (MEMO CBT) to be delivered via multimedia mobile phone messages for teens. Methods We conducted a prospective multicentre, randomised, placebo‐controlled superiority trial in 15 high schools in Auckland, New Zealand, comparing MEMO CBT with a control programme [MEMO control] matched for intensity and type of message but with alternative content not targeting depression. The primary outcome was the change in score on the Children's Depression Rating Scale‐Revised from baseline to 12 months. Secondary outcomes included the change in scores in the self‐reported Reynold's Adolescent Depression Rating Scale‐Second Edition, the Moods and Feelings Questionnaire, suicidal ideation using selected items from the Youth Risk Behaviour Survey, the Pediatric Quality of Life questionnaire, 12‐month period prevalence of the diagnosis of depressive disorder using the Kiddie‐Schedule for Affective Disorders and Schizophrenia, and students' ratings of their satisfaction with the programme. Results Eight hundred and fifty‐five students (13–17 years old, mean 14.3 years) were randomly assigned to MEMO CBT (426) or to MEMO Control (429). Participants (68% female) had a mean CDRS‐R at baseline of 21.5 (SD: 5). Overall 394 (93%) from the intervention group and 392 (91%) from the control group were followed up at 12 months. At the end of the intervention (approximately 9 weeks) the mean CDRS‐R scores were 20.8 in the intervention group versus 20.4 in the control group, and at 12 months they were 22.4 versus 22.4 (p value for difference in change from baseline = 0.3). There was no obvious association between the amount of the intervention viewed by participants and outcomes. Conclusions There was no evidence of benefit from the mobile phone CBT intervention compared with a control programme. Universal depression prevention remains a challenge.