Although the effects of different intraoral scanners, model scanners, and CAM units on the marginal and internal fitting of restorations have been investigated, the effects of CAD software in particular has not been evaluated. The marginal and internal fit of indirect restorations may vary according to the CAD software used, even when using the same intraoral scanner and milling machine. The purpose of this study was to evaluate the marginal and internal fit of milled full ceramic crowns designed with three different CAD systems. Eleven typodont maxillary first premolar teeth were prepared and scanned using a 3Shape TRIOS Intraoral Dental Scanner. The obtained STL scan data were exported and used to design a full crown using three different CAD systems (CEREC, KaVo, and Planmeca). An independent milling unit was used to manufacture the crowns for each group (n = 11). The marginal and internal fit were evaluated for each restoration using 2D and 3D micro-CT analysis. For 2D analysis, 18 measurements for each sample were made, covering the marginal (Marginal Gap Buccal (MG-A), Marginal Gap Palatinal (MG-B), Finish Line Buccal (FL-A), Finish Line Palatinal (FA-B)) and internal fit locations (Axial Wall Buccal (AW-A), Axial Wall Palatinal (AW-B), Lingual Cusp (LC), Buccal Cusp (BC), and Occlusal Central Fossa (OCF)). Statistical analyses were performed using Open Source R Statistical Software (α = 0.05) The results of Duncan's multiple range test showed that the values for the marginal measurement points MG-A, MG-B, FL-A, and FL-B in the Planmeca group were significantly higher than the values obtained in the CEREC and KaVo groups (p < 0.05). In AW1, values of the CEREC group were found to be higher than those of the KaVo and Planmeca groups (p < 0.05). CAD software showed an effect on the marginal fit values of crowns whereas no significant difference was observed in terms of the internal fit, except for a single measurement point made from the buccal direction.