What are you trying to do?
-
-
-
Eg why do you care about churn? Are you measuring it for curiosity? Trying to predict income? Etc
- Show replies
New conversation -
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
you also need to keep in mind the total subscriptions you have and not just the delta. simplest measure is unsubs / existing user base.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
In mobile games we count churn as “% of players who played in period N but not period N+1” for whatever time period you care about. Tracking that separately from new players is important if different time cohort spend/behave differently, which they do for us
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Isn't that just (Ever Subscribed)/(Currently Subscribed)? I'd argue you're probably interested in survival functions, s(d, t) = P(unsub by t+d | sub at t). You could maybe get at it with (unsubs last month)/(active subs last month) for some notion of active.
-
Much better metrics could be pulled by measuring individuals and then aggregating. "Of the cohort of people who have been active in the last n weeks, how many drop?"
End of conversation
New conversation -
-
-
not an expert here but would naively guess some kind of cohort drop-off analysis? (img from https://andrewchen.co/the-easiest-spreadsheet-for-churn-mrr-and-cohort-analysis-guest-post/ …)pic.twitter.com/YaChUZIscY
-
I guess, given the less uniform nature of email lists, I would use this side-by-side with something focusing more on the months themselves including # posts, some kind of rating or traffic-per-post analysis, etc..?
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.