High-immune quality colostrum (IgG1 concentration ⩾50 g/l) is crucial for the health and development of the young calf. Studies on colostrum quality tend to focus on external factors such as breed, parity or dry period length, but few have focused on within-cow variations. Here we ran experiments to gain a deeper insight into within-cow variation in IgG1 concentrations in dairy cow colostrum. Trials were performed in an experimental farm, located in the Western part of France. Colostrum from each quarter and a composite sample (mix of four quarters) were concomitantly collected on 77 Holstein dairy cows just after calving to assess the influence of sample type on IgG1 concentrations. Variation in IgG1 concentrations during the first milking was studied on samples from nine cows collected every minute from the start of milking. Repeatability of colostral IgG1 concentration was estimated from 2009 and 2010 data on 16 healthy cows. IgG1 concentrations were tested using a radial immunodiffusion method. Sensitivity and specificity were similar regardless of sample type tested (individual quarter or composite milk). Mean average IgG1 concentration was 54.1 g/l in composite colostrum, and was significantly higher in hind quarter teats (56.2 g/l) than front quarter teats (53.1 g/l). Average IgG1 concentration did not change significantly during colostrum milking, and the variations observed (15% or less) were likely due to the laboratory method (CV 15%). IgG1 concentrations in dam colostrum increased slightly from 2009 to 2010 due to BW and parity effects. In 56% of cases, colostrum quality could have been assessed on either individual or composite colostrum samples collected at any time during the first milking without affecting the reliability of the measurement. However, in other cases, differences were significant enough to mean that estimates of average IgG1 concentration in colostrum from any one quarter would not be reliable. It is concluded that colostrum quality, from an IgG1 concentration point of view, could be assessed with a composite sample taken at any time during the first milking.