@paulag I'm glad you provide some numbers o key metrics we should be looking at and transparency into the methodology by which you arrived to them.
While everyone gets excited over new account adds, the rubber meets the road with churn because it lets you know how many of those adds you keep. Based upon what the numbers you provided are saying, it seems Steemit is still a bit of a sausage factory with a retention problem. I'm curious if in this report the new adds were normalized to remove what would be considered new bots and sock puppet accounts from new adds (or if that is even possible). I believe you did something in the past to determine a more real number of active users. It would be cool to see if there is a way to determine churn among that group as opposed to the whole.
I did something before where I removed the list of known bots at the time, but to be honest the post was laughed at because there are thousands of bots and I had a few hundred. this task at the moment is impossible.
@paulag The bot problem is pretty bad. I remember when I joined back in July 2016, and it was pretty bad even then. Steemit did not have a real registration review process, and each account was receiving the equivalent of like $7 USD in STEEM at signup when STEEM was valued at like $0.20 or so. People were registering hundreds and thousands of accounts, just to get the free STEEM. It was insane.
Now, it's much harder to do that, and accounts are created for different reasons. It would be cool if at some point we could see what the real churn rate is among active users, but who knows if that could ever be teased out.