OBJECTIVE In previous studies, skin retraction of dermato-pathological specimens after the surgical excision of tumours was calculated at 30% for the surface, with approximately 20% for the length and 15% for the width. The aim of this study was to analyse the retraction of the specimens and the retraction of the lesion and the margins. METHODS Patients who underwent excision of a skin tumour between January 2013 and July 2014 were randomly included. RESULTS A total of 104 patients was included. There were 52% male with a mean age of 68.3 years. Seventy-eight per cent of the lesions were malignant (51% were basal cell carcinoma, 10% squamous cell carcinoma). The retraction of the area of the specimen (29%) was significantly greater than the retraction of the tumour (21%). On multivariate analysis, the localisation and the duration of fixation were independent predictors of the specimen area retraction. The retraction of the specimen was 17% in length and 15% in width. The retraction of the margins was calculated at 19% in length and 12% in width. The surgeon correctly evaluated the localisation of the smallest margin in 55% of cases. CONCLUSIONS Our study provided additional data regarding the retraction of the tumours and margins. The guidelines for surgical excision of skin cancers recommend a clinical margin before excision, but the evaluation of the sufficiency of the margins is based on histological measurement. Our data are useful for the interpretation of the sufficiency of the margins.