Starting from the conceptualization of ‘Cluster Index’ (CI), Villani et al. [16, 17] implemented the ‘Dynamic Cluster Index’ (DCI), an algorithm to perform the detection of subsets of agents characterized by patterns of activity that can be considered as integrated over time. DCI methodology makes possible to shift the attention into a new dimension of groups of agents (i.e. communities of agents): the presence of a common function characterizing their actions. In this paper we discuss the implications of the use in the domain of social sciences of this methodology, up to now mainly applied in natural sciences. Developing our considerations thanks to an empirical analysis, we discuss the theoretical implications of its application in such a different field.