Embodiments of this disclosure relate to systems and methods for determining a set of one or more identified characteristics correlated with high performing projects. Methods include receiving communication data from a plurality of servers, the communication data associated with a plurality of conversations involving one or more users. The communication data is converted into a common format and used to generate a graph, the graph based upon characteristics identified in the communication data and users involved with the plurality of conversations. The communication data can be clustered according to the characteristics and the users, thereby generating one or more clusters around at least one of a characteristic and a user. User data and project data can be generated based on the one or more clusters and be used to determine the set of one or more identified characteristics correlated with high performing projects.