Objective: The aim of this study was to assess the diagnostic value of colposcopy for the diagnosis of cervical intraepithelial neoplasia 2–3/carcinoma in situ and microinvasive cervical cancer. Methods: Sensitivity, positive predictive value, and rate of false negative results of colposcopy were calculated in 718 patients with verified cervical intraepithelial neoplasia 2–3/carcinoma in situ and microinvasive cervical cancer. Assessment was made after final histological verification referring to the estimated diagnosis at colposcopic examination based on International Federation for Cervical Pathology and Colposcopy criteria. Results: A full agreement of colposcopic and morphological diagnosis was observed in 329 of 718 cases, resulting in a colposcopy sensitivity of 45.8% for the diagnosis of cervical intraepithelial neoplasia 2–3/carcinoma in situ and microinvasive cervical cancer. A type 3 transformation zone, dominant in patients with cervical intraepithelial neoplasia 2–3/carcinoma in situ and microinvasive cervical cancer, regardless of age and neoplasia grade (observed in 81.3% of patients included in the study), and a high rate of acetowhite lesions that were not visible (36.6% of patients) limited the sensitivity of colposcopy and colposcopy-guided biopsy, resulting in underdiagnosis, even in young patients. The risk of underdiagnosis grew significantly in women older than 30 years because of the growing incidence of non-visible acetowhite lesions (p = 0.01). This study suggests that large loop excision of the transformation zone may be recommended as an optimal diagnostic procedure in women with high grade squamous intraepithelial lesion (HSIL)+ cytology, even in the absence of lesions at colposcopy. Conclusion: Colposcopy and colposcopy-guided biopsies are not always sensitive enough to assess maximal degree and even the presence of cervical neoplasia. This study suggests that large loop excision of the transformation zone may be recommended as an optimal diagnostic procedure in women with HSIL+ cytology, even in the absence of lesions at colposcopy.