Many analytical tools for the prediction of the shear capacity of reinforced concrete beams do not explicitly predict the crack kinematics that reflect the actual cracking behaviour. As a result, most models require experimentally measured crack patterns and kinematics as input parameters. Hence the crack analysis is a post-assessment technique for shear failures exhibited in laboratory specimens. A significant challenge in the development of a priori critical shear crack analysis is that the critical shear crack shape is known to be unpredictable. There is thus an important need for approaches that reliably and accurately predict the shape of a critical shear crack in slender reinforced concrete beams. Experiments on shear-critical concrete beams without transverse reinforcement were undertaken to investigate the influence of the span to depth ratio and longitudinal reinforcement ratio on the crack shape at failure. Digital image correlation was used to map the dominant crack which was then modelled mathematically using curve fitting tools. The crack shapes were normalised with respect to the beam depth to compare the results for different span to depth ratios. It was found that the beams with longitudinal reinforcement ratios between 0.84% and 1.50% had similar critical crack shapes but the crack shape in the beam with 2.35% reinforcement differed. In the latter case, the dominant crack intersected with the longitudinal steel closer to the beam support leading to a shorter delamination crack and higher ultimate capacity. The initial findings of the study provide a basis for the development of crack-based predictive tools for the shear behaviour of slender reinforced concrete beams without transverse reinforcement.