This assessor-, therapist-, and participant-blinded randomised controlled trial evaluated the effects of an automated internet-based pain coping skills training (PCST) program before home exercise for people with clinically diagnosed hip osteoarthritis. One hundred forty-four people were randomised to either the PCST group or the comparator group. In the first 8 weeks, the PCST group received online education and PCST, whereas the comparison group received online education only. From weeks 8 to 24, both groups visited a physiotherapist 5 times for home exercise prescription. Assessments were performed at baseline, 8, 24, and 52 weeks. Primary outcomes were hip pain on walking (11-point numerical rating scale) and physical function (Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC]) at 24 weeks. Secondary outcomes were other measures of pain, quality-of-life, global change, self-efficacy, pain coping, pain catastrophizing, depression, anxiety, stress, physical activity, and adverse events. Primary outcomes were completed by 137 (95%), 131 (91%), and 127 (88%) participants at 8, 24, and 52 weeks, respectively. There were no significant between-group differences in primary outcomes at week 24 (change in: walking pain [mean difference 0.5 units; 95% confidence interval, -0.3 to 1.3] and function [-0.9 units; 95% confidence interval, -4.8 to 2.9]), with both groups showing clinically relevant improvements. At week 8, the PCST group had greater improvements in function, pain coping, and global improvement than comparison. Greater pain coping improvements persisted at 24 and 52 weeks. In summary, online PCST immediately improved pain coping and function but did not confer additional benefits to a subsequent exercise program, despite sustained pain coping improvements.