I won't dispute that Evernote has grown since, uhh, 2009. But "registered users" is kind of a useless metric in the web and app world. How many of those people signed up and then left? How many were bots? (Evernote doesn't seem to attempt to restrict bot activity...they just require an email address and a password to create an account.) I'd be more interested in Daily Active Users. I also noticed this graph, while nearly 2 years old, also conveniently cuts off before the 2016 freemium changes/price increases that likely shed some of those users.
In terms of this discussion, how much of that user growth is directly attributable to customer recommendations? Or even more relevantly, how many of such recommendations are Android users? It's there that I would expect NPS scores to truly suffer. Two plus years of major platform-specific deficiencies in basic functionality, paired with ongoing pleas for a fix, doesn't make for a loyal fanbase.