# [2019.01.23] EDA I haven't done an exploratory data analysis for quite a while. The project seems to be interesting. I even have a kind of déjà vu now since there is a service not in a great need of a recommender system. And the first question that we need to answer is how to involve people in a type of activity new for them. Not how to retain --- that would be too simple since it's a direct application of recommenders. But first of all, how to involve. That, of course, can have a dead simple answer: use the same techniques as in online advertising. After all, promoting any new activity is very similar to showing banners in a hope of growing conversion.