We describe a usage study of NetscanTech , a system that generates and publishes daily a range of social metrics across three dimensions: newsgroup, author, and thread, for a set of approximately 15,000 technical newsgroups in Usenet. We bring together three interlinked datasets: survey data, usage log data and social accounting data from Usenet participation, to triangulate the relationship between various user roles and differential usage of social metrics in NetscanTech . We found our most frequent users focused on information related to individual authors far more than any other information provided. In contrast, users that visited less frequently focused more on information related to newsgroups and viewing newsgroup metrics. Our results suggest features that designers and developers of online communities may wish to include in their interfaces to support the cultivation of different community roles.