ObjectiveThe optimal repair method for bursal‐side partial‐thickness rotator cuff tears (PTRCTs) involving >50% of the thickness remains a controversial topic. The study was aimed to compare the functional and magnetic resonance imaging (MRI) outcomes after in situ repair or tear completion before repair of bursal‐side PTRCTs.MethodsA retrospective clinical study was conducted involving 58 patients who underwent in situ repair or tear completion before repair of bursal‐side PTRCTs between January 2019 and December 2020. These patients were divided into two groups: the in situ repair group and the tear completion before repair group. Functional assessment consisted of active range of motion (ROM), visual analog scale (VAS), American Shoulder and Elbow Surgeons (ASES) score, and Constant–Murley score. The percentages of patients in each group achieving the minimal clinical important difference (MCID) of the functional scores were determined. The healing status of the rotator cuff was assessed by postoperative MRI.ResultsThere were no statistically significant differences between the two groups in terms of demographic data. The mean follow‐up period was 14.53 ± 2.64 months in the in situ repair group and 15.40 ± 2.66 months in the tear completion before repair group. At the final follow‐up, the forward elevation, external rotation, and internal rotation improved significantly in both groups. The VAS, ASES score, and Constant–Murley score improved significantly in the in situ repair group (5.17 ± 2.00 points to 0.11 ± 0.41 points, p = 0.001; 44.04 ± 17.40 points to 95.47 ± 4.32 points, p = 0.001; 49.50 ± 14.38 points to 93.50 ± 3.49 points, p = 0.001) and in the tear completion before repair group (5.43 ± 3.32 points to 0.03 ± 0.18 points, p = 0.001; 41.50 ± 19.59 points to 95.94 ± 2.68 points, p = 0.001; 47.54 ± 17.13 points to 93.97 ± 2.61 points, p = 0.001). Postoperative MRI revealed that the re‐tear rate was 7.1% (2/28) in the in situ repair group and 3.3% (1/30) in the tear completion before repair group. No significant differences were observed in terms of the functional scores, the percentages of patients achieving the MCID of the functional scores, and the re‐tear rate between the two groups (p > 0.05).ConclusionsBoth in situ repair and tear completion before repair yielded satisfactory clinical outcomes for patients with bursal‐side PTRCTs. No significant differences were observed in the functional and MRI outcomes between the two groups.