Pre-notched small punch test (p-SPT) samples (10 × 10 × 0.5 mm) have been fabricated to determine the crack initiation toughness. The machined notch profile is fabricated longitudinally from the centre of one side of the sample to the centre of the opposite side, producing a notch depth to thickness ratio a/t = 0.3.Plastic damage analysis of the p-SPT specimen has been performed by Gurson-Tvergaard-Needlemen (GTN) material damage model which shows the distribution sites of maximum equivalent plastic strain and maximum total void volume fraction (TVVF) is found to approximately similar pattern at different corresponding punch displacements. Numerical results show that the maximum equivalent plastic strain and maximum TVVF location is shifting from centre of the p-SPT specimen to the circumference side with the increment of punch displacement. Due to this numerical observation, the initiation of ductile tearing starts away from the centre of the p-SPT specimen.The crack initiation point has been predicted by numerical and experimental methods. The numerical method involves the GTN material damage model to identify crack initiation point. This is observed when first Gauss point failed completely. The GTN parameters which are used to invoke the crack initiation point in the p-SPT specimen were calibrated by experimental results of 3 mm diameter disk SPT specimens. However, the experimental method involves the analysis of the slope just before the maximum load to predict the crack initiation point. The predicted crack initiation points using both methods are utilized to estimate crack initiation toughness of structural steel. The crack initiation toughness values were obtained for two structural steels viz. T91 and SS304LN using the p-SPT test and compared with the standard test. The differences found were duly justified.