Oh, ok, I see what you mean.
I think I would use 90 days to be inactive.
I guess numbers of accounts that take a 90 day break and return to be active later can be known?
My bet is most 90 day inactive accounts rarely return, but maybe you can show those numbers, too.
Perhaps the type of transactions have an effect, too.
Of those that made a post (since the fork to hive?), and were active for XX days, XX are currently active.
Those that only made jsons for XX days and are still currently active.
Maybe show the average number of days it takes for formerly active accounts to become inactive?
How many were active for at least XX days and went on to be active XX days later.
Does that help?
If an account was active for 30/60/90 days, then took a minimum 30/60/90 day break, how many are currently active in the last 90 days?
What is the longest inactive streak to come back and be active in the last 30 days?
What are the accounts with the least inactive days?
Which accounts are the most active across the various columns?
Json, transfers, posts, comments, etc.