Despite the fact that renography is practised by most nuclear medicine departments, some doubt remains about its clinical utility in certain settings. Unfortunately, lack of standardization has made it difficult to judge whether some reports of poor diagnostic accuracy are due to a fundamental limitation of the test, or simply to sub-optimal implementation. Some important work may therefore have been overlooked. Technical factors, to do with data acquisition and processing, are the most obvious reasons for variability of results between centres, but patient selection can also have a profound effect on diagnostic accuracy in situations where disease prevalence is low (e.g. renovascular hypertension). Belatedly, these issues are being addressed by the nuclear medicine community and this review summarizes the efforts that have been made to date. Published guidelines need to be reinforced by an audit process if real improvements in quality and consistency are to be realized.