In this study, we compare the primary efficacy of computed tomography-navigated stereotactic guidance to that of manual guidance for percutaneous microwave ablation of liver malignancies. In total, 221 patients (140, 17, and 64 with hepatocellular carcinoma, cholangiocellular carcinoma, and liver metastases, respectively) with 423 treated liver lesions underwent microwave ablation (MWA). Manual guidance (M) and stereotactic guidance (S) were used for 136 and 287 lesions, respectively. The primary endpoint was the primary efficacy and the secondary endpoint was the radiation dose. A generalised estimating equation was applied to analyse the correlation between the primary efficacy (lesion basis) and the type of guidance, size and location of lesion. The primary efficacy rate was significantly higher in the S-group (84.3%) than in the M-group (75.0%, p=0.03). Lesion size>30 mm was negatively correlated with the efficacy rate (odds ratio 0.38; 95% confidence interval 0.20-0.74). Stereotactic guidance was associated with a significantly lower dose length product (p<0.01). In this retrospective study, percutaneous microwave ablation under stereotactic guidance exhibited significantly greater primary efficacy than conventional manual guidance.