Revision of unstable reverse shoulder arthroplasty (RSA) is significantly challenging, with recurrence rates ranging from 20% to 40%. The purpose of this study was to identify factors associated with recurrent instability. The factors studied included (1) indication for revision RSA (failed primary RSA vs. failed revision RSA), (2) previous attempt at stabilization, (3) mechanism of instability, (4) clinical history of instability, and (5) surgical technique. Outcomes were reported in patients with 2-year follow-up. All patients undergoing RSA for instability at our institution were identified. A total of 43 surgical procedures in 36 patients were included. Arthroplasty indication prior to instability (14 failed primary RSAs vs. 22 failed revision RSAs), instances of prior attempts at stabilization (14 patients treated at outside institution), mechanism-of-instability classification, clinical history of instability (17 recurrent and 26 chronic cases), and surgical technique were collected. Stability at final follow-up (minimum, 12 months) and clinical outcomes at 2-year follow-up were assessed. Overall, 32 of 36 patients (89%) required 38 revisions to achieve stability at final follow-up (mean, 53 ± 47 months; range, 12-210 months). On comparison of stability by indication, stability was achieved in 13 of 14 patients (93%) in the failed primary group (mean, 65 ± 59 months; range, 12-210 months) compared with 19 of 22 (86%) in the failed revision group (mean, 45 ± 36 months; range, 12-148 months; P = .365). The average number of procedures per patient was 3 (range, 2-10) in the failed primary group vs. 4.5 (range, 3-7) in the failed revision group (P = .008). Stability was achieved in 12 of 14 patients (86%) with a history of failed stabilization procedures. The most common mechanism leading to persistent instability was loss of compression. Stability was achieved in 14 of 16 patients treated for recurrent instability compared with 18 of 20 treated for chronically locked dislocation (P = .813). Continued instability occurred in 33% of patients who underwent glenoid side-only management, 33% who underwent humeral side-only management, and 10% who underwent bipolar revision tactics. At 2-year follow-up, stability was achieved in 18 of 21 patients, with improvements in the American Shoulder and Elbow Surgeons (ASES) score, forward flexion, abduction, external rotation, and the Simple Shoulder Test score (P = .016, P < .01, P = .01, P < .01, and P = .247, respectively). Patients who underwent multiple revisions after failed previous arthroplasty will require more surgical attempts to achieve stability compared with patients who underwent a revision after failed primary RSA. Loss of compression was the most common mechanism of persistent instability. Stabilization was more reliably achieved in cases of recurrent instability than in cases of chronically locked dislocation. Continued instability was noted in one-third of patients who underwent humeral side-only or glenoid side-only revisions and in 10% of those who underwent bipolar revisions. Patients in whom stabilization was successful had improved clinical outcomes. Copyright © 2023 Journal of Shoulder and Elbow Surgery Board of Trustees. Published by Elsevier Inc. All rights reserved.