The origin of this paper lies in a perceived inconsistency between the laboratory CBR’s poor reproducibility and the CBR’s special place within pavement technology. The paper summarises the reproducibility and repeatability of the laboratory CBR and discusses the implications of the poor reproducibility on design and product quality decisions. It shows that the load-deformation properties assumed for the standard crushed rock (CBR=100) differ significantly from the measured properties of crushed rocks and that this inconsistency results in an undesirable bias in the reported CBR. It explores the laboratory test’s ability to replicate the in situ CBR and investigates the practicality of replacing the CBR of cohesive subgrades with the undrained shear strength. It contends that continuing reliance on the CBR hinders the development of pavement technology.