Background: Gluteal tendons tears are increasingly being recognised as a cause of recalcitrant trochanteric pain, but there is a paucity of robust studies analysing the type of tears, results and predictors of outcome. Methods: Patients with gluteal tendon tears resistant to conservative treatment who underwent isolated open repair (IR) or total hip arthroplasty and concomitant repair (THA+repair) with a minimum 1-year follow-up were retrospectively assessed separately. Type of tear, surgical approach, and fixation methods were registered. Complications and postoperative outcomes were analysed: visual analogue scale (VAS) of pain, VAS satisfaction, and activity level. Univariate regressions and multivariable models were developed. Results: 90 cases were included: 62 cases underwent IR and 28 cases THA+repair, with an average follow-up of 3.9 years and 2.6 years, respectively. 13 complications were found ( n = 5 IR, n = 8 THA+repair), including 9 re-operations ( n = 4 IR, n = 5 THA+repair). Mean VAS pain score was 3.02 (SD 2.74) for the IR and 2.32 (SD 2.43) for THA+repair group. Mean VAS satisfaction was 7.09 (SD 3.07) and 7.68 (SD 2.71) for the IR and THA+repair group, respectively. In the IR group 61.4% returned to all pre-injury activities, whereas in the THA+repair group 79% did. Full-thickness tears had higher VAS pain scores ( p = 0.0175), and there was trend ( p > 0.05) towards higher complications, re-tears and lower VAS satisfaction in this type of tears in both groups. No statistically significant differences were found in outcomes when comparing THA+repair through direct anterior (DAA) with lateral approach (LA). Conclusions: Isolated gluteal tendon repair or THA+repairs seem to be safe procedures with high levels of satisfaction at short- to mid-term follow-up. The presence of a full-thickness tear is a predictor of inferior outcomes.