fitness-studio-ops
Spot members at risk of cancelling from an attendance CSV — reads visits for the last 30 days vs. the prior 30 days, flags members whose attendance dropped 50% or more (with 4+ prior visits, so already-inactive members are skipped), sorted by drop percentage. Use monthly to build your win-back call list.
$ skillbank install fitness-studio-ops@1.0.0 # pins exactly 1.0.0 · README + changelog included
Fitness Studio Ops
Overview
Members do not cancel out of nowhere — they fade first. This skill reads a member CSV (visits in the last 30 days vs. the prior 30 days) and flags members whose attendance dropped by 50% or more, sorted by drop percentage so the steepest falloffs get attention first. Members with fewer than 4 prior visits are skipped: they were already inactive, so flagging them would just add noise.
Run it monthly and work the at-risk list with the win-back offer ladder in references/winback-ladder.md: a week-1 check-in text, a week-2 free guest pass, and a week-3 discounted personal session.
Installation
1. Copy the fitness-studio-ops/ directory anywhere convenient. 2. Make the script executable (Unix):
chmod +x scripts/churn.py
No pip install needed — standard library only.
Configuration
The script reads a plain CSV with exactly these columns:
name,visits_last_30d,visits_prior_30d
Grace Hall,3,12
| Column | Format | |--------------------|---------------------------------------------| | name | Member name, non-empty. | | visits_last_30d | Whole number of visits, 0 or more. | | visits_prior_30d | Whole number of visits, 0 or more. |
Drop % = (prior − recent) ÷ prior × 100. Bad counts print a clear error to stderr and exit nonzero — never a traceback.
Inputs
| Command | Arguments | Returns | |------------|-----------------|------------------------------------------------| | churn.py | --csv <file> | At-risk members sorted by drop %, largest first. |
No arguments prints usage to stderr and exits 2. --help prints usage and exits 0.
Worked example
$ ./scripts/churn.py --csv members.csv
At-risk members — attendance dropped 50%+ with 4+ prior visits: 3
Member Prior 30d Recent 30d Drop
------------------------------------------------------
Grace Hall 12 3 75.0%
Leo Marsh 20 8 60.0%
Henry Ito 8 4 50.0%
Tested with 6 members: 3 flagged at risk, 3 correctly excluded — Ivy Jones (only a 10% dip), Kara Lane (no dip), and Jake King (3 prior visits, below the 4-visit activity floor, so he was already inactive). Math verified by hand on Grace Hall: (12 − 3) ÷ 12 × 100 = 75.0%.
Safety notes
The script reads your member CSV and prints to stdout — it never uploads, emails, or transmits anything. Member attendance data stays on your machine. Keep the CSV somewhere with proper access controls.
Version history
Changelog
1.0.0 — 2026-09-27
- Initial public release.
scripts/churn.py: member CSV → at-risk list for attendance drops of 50%+ (prior visits ≥ 4 required, so already-inactive members are not flagged), sorted by drop percentage.references/winback-ladder.md: week-1 check-in text, week-2 free guest pass, week-3 discounted personal session.- Zero dependencies; tested with a 6-member sample CSV (math hand-verified) plus bad-input cases.