A systematic review and a meta-analysis of observational studies were performed to assess the dose–response relationship between specific types of dairy foods and the risk of the metabolic syndrome (MetS) and its components. Studies of dairy foods and the risk of the MetS and its components published up to June 2016 were searched using PubMed, EMBASE and a reference search. Random-effects models were used to estimate the pooled relative risks (RR) with 95 % CI. Finally, ten cross-sectional studies, two nested case–control studies and twenty-nine cohort studies were included for the analysis. In a dose–response analysis of cohort studies and cross-sectional studies, the pooled RR of the MetS for a one-serving/d increment of total dairy food (nine studies) and milk (six studies) consumption (200 g/d) were 0·91 (95 % CI 0·85, 0·96) and 0·87 (95 % CI 0·79, 0·95), respectively. The pooled RR of the MetS for yogurt (three studies) consumption (100 g/d) was 0·82 (95 % CI 0·73, 0·91). Total dairy food consumption was associated with lower risk of MetS components, such as hyperglycaemia, elevated blood pressure, hypertriacylglycerolaemia and low HDL- cholesterol. A one-serving/d increment of milk was related to a 12 % lower risk of abdominal obesity, and a one-serving/d increment of yogurt was associated with a 16 % lower risk of hyperglycaemia. These associations were not significantly different by study design, study location or adjustment factors. This meta-analysis showed that specific types of dairy food consumption such as milk and yogurt as well as total dairy food consumption were inversely related to risk of the MetS and its components.