The aim of this study was to determine the additional values of multiple cervical biopsies when any colposcopy was performed. We developed a cross-sectional study of 92 women, who had been referred for a colposcopy because of their abnormal cervical cytology. Colposcopy-directed biopsies were taken from lesions and random non-directed biopsies were added, if their directed biopsies were fewer than four in number. The biopsy sites were ranked according to the impression of the clinicians. Among the 92 women, the first biopsy was normal in 29.4%, revealing CIN1 in 28.3% and CIN2+ in 42.3%. In the second and third biopsies, the CIN2+ was found to have increased to 35.8% and 36.8%, respectively. The accumulative sensitivity for detecting CIN2+ was 84.8%, for a single biopsy. This increased to 97.0%, after two biopsies and then to 100%, after three and four biopsies. To conclude, although the taking of the additional biopsies increased the CIN2+ detection, collecting three cervical biopsies might be sufficient.Impact StatementWhat is already known on this subject? The colposcopy is considered to be the standard procedure in the detection of precancerous lesions of the cervix. However, nowadays, colposcopic biopsy practices do not have any single, acceptable guideline for the number of biopsies performed, and whilst a single biopsy is the most commonly adopted practice, some centres have used a biopsy protocol with multiple biopsies.What the results of this study add? This study determined the rate of the detection of CIN2+ by using multiple biopsy protocols during colposcopy for women who were referred with their abnormal cervical cytology. We explored the benefit of collecting additional lesion-directed biopsies and additional biopsies of a normal-appearing cervix in addition to a single biopsy. We found that two or three biopsies from a colposcopy should be enough for increasing the detection of CIN2+. Also, multiple biopsies increased the sensitivity of CIN2+ detection, especially in colposcopic impression for the low grade lesions.What the implications are of these findings for clinical practice and/or further research? We suggest that colposcopy-directed biopsies should be supplied by one or two random biopsies from other quadrants of the cervix.