Our aim was to test a novel method for estimating the age of young calves, using serum γ-glutamyl transferase (GGT) activity, total protein concentration and immunoglobulin G (IgG) concentration. Blood samples were taken from 59 dairy calves at 2, 4, 6, 8, 10, 12, and 14 days of age, and serum GGT activity, total protein concentration, and IgG concentration were measured. Bayesian regression models were used to estimate the association of GGT, total protein, and IgG, with calf age. A multivariate hierarchical regression model was then created, and the model's reliability in estimating the age of simulated subjects was assessed. The univariate models showed a strong within-calf relationship between age and GGT (conditional R2 (cR2) = 0.93), and age and total protein (cR2 = 0.75), while the relationship between calf age and IgG was less consistent (cR2 = 0.63). There was a high degree of variation between calves for these parameters, particularly for the relationship between age and total protein (marginal R2 (mR2) 0.02), and age and IgG (mR2 0.01); somewhat less between-calf variation was seen for GGT (mR2 0.30). For the final multivariate model, we tested the model reliability by simulating new subjects. The credible intervals for estimates of calf age generated from the model were wide, indicating poor reliability. We concluded that single measurements of serum GGT activity, total protein concentrations and IgG concentrations cannot be used to reliably estimate the age of young calves to within 1–2 days.