Just a quick follow up: the other week I quickly threw together a model trying to extrapolate Evernote’s cohort churn and premium conversion based on some data shared in NYTimes and TechCrunch articles. I think the growth/churn estimates were decent, but I found a sneaky bug in the premium conversion method and in re-reviewing the NYTimes article there are some data points mentioned that seemed inconsistent. Without more data, I do not trust the premium conversion speculation anymore, so have removed the post and model.