Background: Successful total shoulder arthroplasty (TSA) relies on accurate placement of implants, which is difficult in the setting of bone loss or deformity. Technologies are becoming available to provide intraoperative assistance to better execute the preoperative plan. The purpose of this study was to compare patient outcomes following TSA utilizing either computer navigation or conventional techniques. Methods: This retrospective review included 180 Primary Exactech TSA cases with a minimum 2-year follow-up. There were 40 anatomic (12 non-navigated, 28 navigated) and 140 reverse (80 non-navigated; 60 navigated) TSA procedures. Patient groups were similar in age, sex, side involved, and prior surgery. Patient-reported outcome measures, complications, revisions, and reoperations were assessed and compared between non-navigated and conventional groups. Results: Of available patients, the navigated anatomic cohort had statistically significant improvements compared to the non-navigated cohort in American Shoulder and Elbow Surgeons (ASES), Oxford, and pain scores. For the navigated reverse cohort, significant improvements were seen in Oxford score. In multivariate analysis, all outcomes favored the navigated cohorts, with ASES, Oxford, and patient function scores reaching statistical significance. Complications occurred more frequently in the non-navigated reverse cohort. Revisions and/or reoperations were more frequent in non-navigated shoulders. Conclusions The use of computer navigation in TSA may be associated with decreased complication rates and improved patient outcomes, a benefit to surgeons and their patients. However, the lack of radiographic assessment is a limitation, and as with all new technology, further research with longer follow-up is needed to fully define the role of navigation in TSA. Level of evidence: III.