Aims: To investigate the relationship between Zn concentrations in serum and those in milk or faeces, and to assess the ability of the Zn concentrations in milk, serum and faeces to predict intake of ZnO in dairy cattle. Method Seventy cows from one commercial farm in the Waikato region of New Zealand received one of seven dose rates (0, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5 g/100 kg bodyweight (BW)) of ZnO given by oral drench, every morning, for 7 consecutive days. Every afternoon, milk and blood samples were collected from all cows. Free-catch faecal samples were collected during the afternoon milking on 3 days throughout the trial. Linear mixed models were used to assess the relationship between the concentration of Zn in serum and that in milk, and in faeces, respectively, and the relationship between dose rate of ZnO and concentrations of Zn in serum, faeces and milk, respectively. Receiver operating characteristic curve analysis was used to determine the ability of the Zn concentration in serum, milk and faeces to predict that a cow had been treated with a dose of ZnO ≥2.5 g/100 kg, the industry-recommended dose rate needed to protect against facial eczema. Results: A 1-µmol/L increase in Zn concentration in milk was associated with a 0.14 (95% CI=0.11–0.17) µmol/L increase in Zn concentration in serum. Zn concentration in faeces was scaled by its SD; a 1 SD increase was associated with a 1.83 (95% CI=0.54–3.12) µmol/L increase in zinc concentration in serum. Zn concentrations in serum and faeces increased with increasing dose rates of ZnO. No differences in Zn concentrations in milk were noted between animals dosed with 1.5–3.5 g ZnO/100 kg BW, inclusive. At the optimal threshold for Zn concentration in serum, 22 µmol/L, the sensitivity was 0.76 (95% CI=0.69–0.82) and specificity 0.85 (95% CI=0.80–0.89). For the concentration of Zn in faeces, the optimal threshold was defined as 17.36 mmol/kg, with a corresponding sensitivity of 0.84 (95% CI=0.84–0.85) and specificity of 0.85 (95% CI=0.73–0.94). At the optimal threshold for the Zn concentration in milk (76.6 µmol/L), the sensitivity was lower than the other two sample types at 0.59 (95% CI=0.52–0.67), but with a similar specificity of 0.84 (95% CI=0.79–0.88). Conclusions and clinical relevance: The concentration of Zn in milk shows promise as an initial screening test to identify dairy farms that do not provide adequate zinc to provide protection against FE.