Since the 1970s, the ability of some invertebrate species to recognize individual conspecifics has attracted increased scientific interest. However, there is still confusion in the literature, possibly due to the lack of unambiguous criteria for classifying social recognition in its different forms. Here, we synthesize the results of studies on invertebrates and provide a framework with the purpose of identifying research needs and directions for future investigations. Following in part Sherman et al.'s (Behavioural ecology: an evolutionary approach. Blackwell Science, Oxford, pp 69-96, 1997) definition of 'recognition systems' and Tibbetts and Dale's (Trends Ecol Evol 22:529-537, 2007) classification of 'individual recognition,' we first discuss different case studies that exemplify the categories of 'familiar recognition' and 'class-level recognition.' Then, through the analysis of the invertebrate literature, we illustrate eight key properties that characterize 'true individual recognition' systems. We are confident that the proposed framework will provide opportunities for exciting discoveries of the cognitive abilities in invertebrates.