Abstract Background Shoulder pain is the third most common musculoskeletal disorder, often affecting people’s daily living and work capacity. The most common shoulder disorder is the subacromial impingement syndrome (SIS) which, among other pathophysiological changes, is often characterised by rotator cuff tendinopathy. Exercise is often considered the primary treatment option for rotator cuff tendinopathy, but there is no consensus on which exercise strategy is the most effective. As eccentric and high-load strength training have been shown to have a positive effect on patella and Achilles tendinopathy, the aim of this trial is to compare the efficacy of progressive high-load exercises with traditional low-load exercises in patients with rotator cuff tendinopathy. Methods/Design The current study is a randomised, participant- and assessor-blinded, controlled multicentre trial. A total of 260 patients with rotator cuff tendinopathy will be recruited from three outpatient shoulder departments in Denmark, and randomised to either 12 weeks of progressive high-load strength training or to general low-load exercises. Patients will receive six individually guided exercise sessions with a physiotherapist and perform home-based exercises three times a week. The primary outcome measure will be change from baseline to 12 weeks in the patient-reported outcome Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire. Discussion Previous studies of exercise treatment for SIS have not differentiated between subgroups of SIS and have often had methodological flaws, making it difficult to specifically design target treatment for patients diagnosed with SIS. Therefore, it was considered important to focus on a subgroup such as tendinopathy, with a specific tailored intervention strategy based on evidence from other regions of the body, and to clearly describe the intervention in a methodologically strong study. Trial registration The trial was registered with Clinicaltrials.gov ( NCT01984203 ) on 31 October 2013.