Aim Emerging evidence from epidemiologic studies and basic science suggests a potential antitumour effect of metformin. However, whether metformin improves survival in cancer patients remains inconclusive. Methods A literature search was performed using the PubMed, EMbase and SciVerse Scopus databases. Pooled effect estimates were derived using a random‐effects meta‐analysis model. Results Of the 28 studies retrieved, the pooled effect estimates showed that metformin was associated with lower risk of all‐cause mortality in cancer patients with concurrent diabetes, particularly for breast pooled relative risk (RR) 0.70, 95% CI 0.55, 0.88; p = 0.003, colorectal (RR 0.70, 95% CI 0.59, 0.84; p < 0.001), ovarian (RR 0.44, 95% CI 0.30, 0.64; p < 0.001) and endometrial cancer (RR 0.49, 95% CI 0.32, 0.73; p = 0.001). In addition, metformin was associated with lower risks of cancer‐specific mortality. Conclusions The findings of this study support the hypothesis that metformin improves the survival for cancer patients with concurrent diabetes, particularly for breast, colorectal, ovarian, and endometrial cancer. Further investigation is warranted.