Whilst any estimation of crime costs is a challenge even at a national level and in respect to crimes not particularly problematic from the definitional viewpoint, like volume crimes, the task is much harder when one has to deal with the harm caused by organised crime, especially from a comparative perspective. First, notwithstanding many international acts and studies, the term 'organised crime' is still one of those most debated and blurred in criminology. To complicate matters further, any cross-country assessment encounters such a wide variety of national differences (cultural, in the definition of offences, and in crime data collection systems) that the results are hardly comparable. Though extremely difficult, discussing the topic makes a great deal of sense today, and especially within the European Union, because there is strong demand for sound knowledge on the most harmful activities perpetrated by organised criminal groups and where these are localised. Considering the increasing importance attached to the issue, this article critically discusses existing attempts to measure organised crime harm from a comparative perspective, highlighting their strengths and weaknesses. It first reviews harm assessment models developed to date at the international level, and mainly consisting of surveys. It then presents a different approach to the subject, one more centered on official statistics and which has recently resulted in the development of a methodology in the context of a EU-funded study entitled IKOC (Improving Knowledge on Organised Crime to develop a common European approach). Tables and references (Published Abstract)