The aim of this study was to evaluate the accuracy of navigation-assisted maxillofacial reconstruction and to identify the predictors of the clinical outcomes. A total of 112 patients who underwent navigation-assisted maxillofacial reconstruction with free flaps between 2014 and 2019, performed by a single surgical team, were assessed. Accuracy was evaluated by superimposing the postoperative computed tomography data with the preoperative virtual surgical plan. Predictors of the clinical outcomes affecting the accuracy were identified and analysed. The mean deviation and root mean square (RMS) estimate of the orbital, maxillary, and mandibular reconstructions were 0.88 ± 3.25 mm and 3.38 ± 0.73 mm, 0.77 ± 3.44 mm and 3.69 ± 0.82 mm, and 1.07 ± 4.16 mm and 4.67 ± 3.95 mm, respectively (P < 0.05). There was no significant difference in orbital volume or projection between the preoperative, postoperative, and healthy orbits (P = 0.093 and P = 0.225, respectively). Multivariate linear regression analysis confirmed significant associations between the accuracy of navigation-assisted mandibular reconstruction and preservation of the condyle, type of reconstruction, type of osteosynthesis plate, and number of bony segments. Navigation-assisted midface reconstruction yielded a higher level of accuracy in the final surgical outcome when compared to mandibular reconstruction. Computer-assisted techniques and intraoperative navigation can be an alternative or adjunct to current surgical techniques to improve the final surgical outcome, especially in more complex maxillofacial reconstructions.