Improved short-term outcomes have been demonstrated with higher surgical volume in shoulder arthroplasty. There is however, little data regarding long-term outcomes.Revision data from the Australian Orthopaedic Association National Joint Replacement Registry from 2004-2017 was analyzed according to 3 selected surgeon volume thresholds: <10, 10-20, and >20 shoulder arthroplasty cases per surgeon, per year.There was a significantly higher rate of revision for stemmed total shoulder arthroplasty (TSA) for osteoarthritis (OA) for the <10/yr compared with the >20/yr group for the first 1.5 years only (hazard ratio [HR] 1.36, 95% confidence interval [CI] 1.08-1.71, P = .009). For reverse total shoulder arthroplasty (rTSA) performed for OA, there was a higher revision rate for the <10/yr compared with the >20/yr group for the first 3 months only (HR 2.58, 95% CI 1.67-3.97, P < .001). In rTSA for cuff arthropathy, there was a significantly higher rate of revision for the <10/yr compared with the >20/yr group throughout the follow-up period (HR 1.66, 95% CI 1.21-2.28, P = .001). There was no significant difference for the primary diagnosis of fracture.Lower surgical volume was associated with higher all-cause revision rates in the early postoperative period in TSA and rTSA for OA and throughout the follow-up period in rTSA for cuff arthropathy. Despite increases in the volume of shoulder arthroplasties performed in recent years, more than 78% of surgeons undertake fewer than 10 procedures per year.