Previous studies of the association between intake of dairy products and colorectal cancer risk have indicated an inverse association with milk, however, the evidence for cheese or other dairy products is inconsistent. We conducted a systematic review and meta-analysis to clarify the shape of the dose–response relationship between dairy products and colorectal cancer risk. We searched the PubMed database for prospective studies published up to May 2010. Summary relative risks (RRs) were estimated using a random effects model. Nineteen cohort studies were included. The summary RR was 0.83 (95% CI confidence interval: 0.78–0.88, I2 = 25%) per 400 g/day of total dairy products, 0.91 (95% CI: 0.85–0.94, I2 = 0%) per 200 g/day of milk intake and 0.96 (95% CI: 0.83–1.12, I2 = 28%) per 50 g/day of cheese. Inverse associations were observed in both men and women but were restricted to colon cancer. There was evidence of a nonlinear association between milk and total dairy products and colorectal cancer risk, P < 0.001, and the inverse associations appeared to be the strongest at the higher range of intake. This meta-analysis shows that milk and total dairy products, but not cheese or other dairy products, are associated with a reduction in colorectal cancer risk.