Social networks have become an important tool in behavioural ecology. But unlike other quantities, networks are inferred rather than directly measured. Depending on the way this is done, social networks can have a large uncertainty that previous methods haven’t been able to quantify. We have developed a general Bayesian framework that solves this problem and conducts Bayesian social network analysis to make the most of hard-won behavioural data. If you’re interested, you can read more about the framework in our preprint and get hands on with BISoN in our R package bisonR.