Ocular oncologists require a strong indication for intraocular biopsy before the procedure can be performed because it carries a risk for serious eye complications and the dissemination of malignant cells. The purpose of this review is to evaluate the extent to which this restricted practice is supported by evidence from previous reports and to outline our main indications and contraindications. The different intraocular biopsy techniques in the anterior and posterior segment are discussed with a focus on our preferred method, fine-needle aspiration biopsy (FNAB). In the literature, complications are typically under-reported, which reduces the possibilities of evaluating the risks correctly and of making fair comparisons with other biopsy methods. In FNAB, the exact placement of the needle is critical, as is an accurate assessment of the size of the lesion. Fine-needle aspiration biopsy is usually not a reliable diagnostic tool in lesions < 2 mm in thickness. It is very advantageous to have a cytopathologist present in the operating theatre or close by. This ensures adequate sampling and encourages repeated biopsy attempts if necessary. This approach reduces false negative results to < 3%. Adjunct immunocytochemistry is documented to increase specificity and is essential for diagnosis and management in about 10% of cases. In some rare pathological processes the diagnosis depends ultimately on the identification of specific cell markers. An accurate diagnosis may have a decisive influence on prognosis. The cytogenetic prognostications made possible after FNAB are reliable. Biopsy by FNA has a low complication rate. The calculated risk for retinal detachment is < 4%. Intraocular haemorrhage is frequently observed, but clears spontaneously in nearly all cases. Only a single case of epibulbar seeding of malignant cells at the scleral pars plana puncture site of transvitreal FNAB has been documented. Endophthalmitis has been reported and adequate standard preoperative preparation is obligatory. An open biopsy is still an option in the anterior segment, but has been abandoned in the posterior segment. Although vitrectomy-based procedures are becoming increasingly popular, we recommend using FNAB as part of a stepwise approach. A vitrectomy-assisted biopsy should be considered in cases where FNAB fails. In any adult patient with suspected intraocular malignancy in which enucleation is not the obvious treatment, the clinician should strive for a diagnosis based on biopsy. When the lesion is too small for biopsy or the risks related to the procedure are too great, it is reasonable to be reluctant to biopsy. The standards applied in the treatment of intraocular malignant diseases should be equivalent to those in other fields of oncology. Our view is controversial and contrary to opinion that supports current standards of care for this group of patients.