Objective: To investigate if a hospital-initiated home-based rebozo intervention performed by the pregnant woman and her partner before external cephalic version (ECV) would increase the rate of cephalic presentations at birth. Design: A multicentre randomised controlled trial. Setting: Three university hospitals in Copenhagen, Denmark. Population: Pregnant women with a breech or transverse presentation at 35 weeks or more of gestation eligible for ECV. Methods: We compared rebozo before ECV with ECV alone. The randomisation was computer-generated in blocks and stratified by parity. The woman and her partner were instructed in the technique by a project midwife and performed the technique at home three times daily for 3–5 days before the scheduled ECV. Analyses were by intention-to-treat. Main outcome measure: The number of cephalic presentations at the time of birth. Odds ratios (OR) and 95% confidence intervals (CI) were calculated. Results: A total of 372 women were randomly assigned (1:1) to either rebozo intervention (n = 187) or control (n = 185). At birth, 95 (51%) in the intervention group versus 112 (62%) in the control group had a fetus in cephalic presentation (OR 0.61; 95% CI 0.40–0.95). No adverse events were observed in relation to the intervention. Conclusions: In breech or transverse presentation, home-based rebozo exercise before ECV lowered the overall rate of cephalic presentation at birth. Tweetable abstract: Home-based rebozo for breech presentation before external version reduces the rate of cephalic presentation at birth.