The article analyzes the approaches of international courts (the UN International Court of Justice, International Criminal Court, International Criminal Tribunal for the Former Yugoslavia, International Criminal Tribunal for Rwanda and hybrid Extraordinary Chambers in the Courts of Cambodia) to the criteria for defining genocide groups. The article emphasizes that the definition of belonging to a group is a contextual circumstance (contextual element) of the crime of genocide. In particular, the paper studies how the international courts applied positive/negative and objective/subjective identification strategies to conclude that certain groups constitute those protected by the Genocide Convention or the statutes of the international criminal courts. In addition, the article deals with the problem of the stability and mobility of the groups and the ways these characteristics help the international courts to apply the Convention.The article focuses on a search for algorithms that allow international courts to identify genocide groups. It stresses that the international criminal courts have not demonstrated consistency in their assessment of the definition of the groups. Neither have they showed the synchronized understanding of the approaches (objective/subjective, positive/negative, stable/mobile) to be used for the identification of these groups. Therefore, it is further argued that, due to the variability of approaches and strategies used by international courts to identify genocide groups, belonging to the group is a window of opportunity for a contextual reading of international criminal law.