How to calculate and read your server's stickiness ratio
DAU/MAU is borrowed from product analytics, and most communities cite it wrong. Here is a clean procedure plus how to interpret the number once you have it.
How to compute it
— DAU: unique members who sent a message or reacted that day (lurker-views don't count on most platforms anyway)
— MAU: unique active members across the trailing 30 days
— Divide DAU (averaged over the month) by MAU
What the bands mean
A ratio of 0.10 means the average member shows up ~3 days a month. Consumer apps celebrate 0.20+; thriving communities often sit 0.15-0.25. Below 0.10 you have a notification-board, not a community — people check in, not live in.
Why the denominator lies
If you count lurkers as MAU on Telegram (where you can't), your ratio collapses artificially. Keep numerator and denominator on the same definition of 'active' — message-or-react — or the number is meaningless.
Discord vs Telegram wrinkle
Discord gives you presence and message events; Telegram gives you almost nothing per-user in groups. On Telegram, proxy stickiness with daily-unique-senders / monthly-unique-senders from your own logging bot.
The caveat
Stickiness rewards chatty communities and punishes high-signal low-frequency ones (a release-notes channel can be healthy at 0.05). Read it against intent.
Open question: should stickiness be measured per-channel rather than per-server, since members live in 2-3 channels, not all 40?
Server Signal
@ServerSignal
How to calculate and read your server's stickiness ratio
Этот пост опубликован в Telegram-канале Server Signal. Подписаться можно по ссылке: @ServerSignal.