Online communities in the enterprise are designed to fulfil some economic purpose, for example for supporting products or enabling work-collaboration between knowledge workers.
The intentions of such communities allow them to be labelled based on their type - i.e. communities of practice, team communities, technical support communities, etc. Despite the disparate nature and explicit intention of community types, little is known of how the types differ in terms of a) the participation and activity, and b) the behaviour of community users. Such insights could provide community
managers with an understanding of normality and a diagnosis of healthiness in their community, given its type and corresponding user needs. In this paper, we present an empirical analysis of community types from the enterprise social software system IBM Connections. We assess the micro (userlevel) and macro (community-level) characteristics of differing community types and identify key differences in the behaviour that users exhibit in these communities. We further
qualify our empirical findings with user questionnaires by identifying links between the objectives of the users and the characteristics of the community types.