Worked scenario
Two hundred members, and nobody knows who stopped coming
Membership businesses lose people long before they cancel, and attendance is the signal that is recorded and never read.
3 min read
A member joins, attends four times a week, then twice, then stops. The membership continues billing until it lapses, at which point the centre discovers it has lost them.
The decision precedes the cancellation
By the time someone cancels, they stopped attending months earlier. Retention effort at renewal is aimed at a decision that was already taken.
Which drop matters
A fall relative to that member's own pattern, not an absolute threshold. Someone who always came twice a week and now comes once is a signal; someone who has always come once is not.
Cohorts tell you about the centre
Retention by joining month reveals whether a particular intake or promotion produced members who stayed. Most centres measure joins and cancellations separately and never connect them, so they repeat promotions that fill the gym with people who leave.
What changes
- Attendance aggregated per member, not just logged at the door
- Members whose attendance has fallen surfaced while they are still members
- Contact made on a pattern rather than at renewal
- Retention measurable by joining cohort
Questions about anything here, or a situation this does not cover? contact@anantatechhub.com

