Background: The efficacy of widely used povidone-iodine applicators for skin sterilization in abdominal surgery is unclear. The aim of this trial was to evaluate whether sterilization with a povidone-iodine applicator was not inferior to a conventional sterilization method. Methods: Patients undergoing elective abdominal surgery were assigned randomly to receive single sterilization with the applicator or conventional sterilization. The primary endpoint was wound infection rate. Secondary endpoints were rate of organ/space surgical-site infection (SSI), adverse effects of povidone-iodine, amount of povidone-iodine used and total cost of sterilization. Results: Of 498 patients eligible for the study between April 2015 and September 2017, 240 were assigned and analysed in the applicator group and 246 in the conventional group. Wound infection was detected in 16 patients (6·7 per cent) in the applicator group and 16 (6·5 per cent) in the conventional group (absolute difference 0·0016 (90 per cent c.i. -0·037 to 0·040) per cent; P = 0·014 for non-inferiority). There was no difference between the groups in the organ/space SSI rate (11 patients (4·6 per cent) in the applicator group and 16 (6·5 per cent) in the conventional group. Both the amount of povidone-iodine used and the total cost of sterilization were higher in the conventional group than in the applicator group (median 76·7 versus 25 ml respectively, P < 0·001; median €7·0 versus €6·4, P < 0·001). Skin irritation was detected in three patients in the conventional group. Conclusion: In abdominal surgery, this povidone-iodine applicator was not inferior to conventional sterilization in terms of the wound infection rate, and it is cheaper. Registration number: UMIN000018231 (http://www.umin.ac.jp/ctr/).