Talk at Community Leadership Summit 2014, Portland, OR, USA.
Large communities, and specially those collaborating to produce some sort of common result, are complex and difficult to understand. It is very difficult to track what's happening in them, to react early to problems, and to help them to grow healthy.
Fortunately, community analytics come to rescue, enabling the community manager to get a knowledge of what's happening that is, in many cases, very difficult to obtain without their help. I will explain how analytics can be used not only to improve this knowledge, but to track the effect of corrective measures, to detect early deviations from intended evolution, and to provide a more person-oriented management of the community.
I will use as examples real cases from free / open source software development communities.