APIANT.ai / Stories / Fitness studio
Customer story · Fitness studio · Mindbody
A studio owner wanted to see which members were drifting away. He asked his AI client for the Mindbody reports, and had seven working by the next afternoon.
A boutique fitness studio runs everything on Mindbody: classes, memberships, sales. The owner connected Mindbody to APIANT one evening and asked his AI client for the reports the next morning.
Apps in this story
The challenge
Mindbody records every visit. The owner wanted one answer out of those visits: which members are drifting away. The rules: count only visits where the member signed in; a member turns amber after a week without one, and red after two.
He also wanted a weekly report: each member's colour, classes attended and current contract, plus the intro offers that started that week.
- 16minutes to four versions of his member report, the last one running in 20 seconds
- 12automations built in two days, every one from his AI client
- Read-onlyevery report reads Mindbody and changes nothing in it
The journey
-
Day 1, evening
Connected within the hour
He signed up, connected Mindbody with its API key, then connected his AI client to APIANT, all in about twenty minutes. From then on he built nothing in the browser: every automation was asked for, built and run from his terminal.
-
Day 2, morning
The member traffic light, four versions in sixteen minutes
The ask, in his own step notes: sort members into green, amber and red by days since their last signed-in visit, counting only real visits and setting aside class-pass and intro-offer holders.
- Version 1 pulled the last 60 days of visits and sorted members. It ran in under a second.Works
- Version 2 added each member's status, so only active members were scored.Works
- Versions 3 and 4 moved to looking members up one by one, and version 4 narrowed that to only the members who might be slipping, amber and red, and skipped the greens. It ran in 20 seconds.Works, 20 s
-
Day 2, lunchtime
The weekly intro-offer report
Visits plus sales, to find new intro-offer clients and look each one up. It pulled three pages of sales in 34 seconds.
Works, 34 s
-
Day 2, lunchtime
Upgraded from free
He upgraded from free and went straight back to building.
Upgraded from free
-
Day 2, early afternoon
Five more reports in seven minutes
- All-time classes per client, month by month.
- The current member list.
- Member history in batches.
- Extra pages of visits per member.
- Classes counted once per class.
Every run succeeded.
What he built
| Report | What it does | Result |
|---|---|---|
| Member traffic light | Green, amber, red by days since last visit, active members only | Works, 20 s |
| Weekly intro-offer report | New intro-offer clients from visits and sales | Works, 34 s |
| Classes per client | All-time classes, month by month | Works |
| Current member list | Visits plus five pages of sales | Works |
| Member history | Each member's history, in batches | Works, four batches |
| Visits per member | Extra pages of visits for each member | Works |
| Classes per member | Each class counted once | Works |
Twelve automations in all, counting the earlier versions of the first two.
Why it worked
- 01
He asked in plain English.
The steps, the loops and the Mindbody calls were the agent's job; his job was saying what "drifting away" means for his studio.
- 02
Every version was tested before he relied on it.
Each report was run against his real Mindbody data from his AI client, and each new version followed the last within minutes.
- 03
Read-only by design.
Every report reads from Mindbody and changes nothing in it.
More stories
- A yoga studio wanted every Mindbody sale in its membership register, sorted by its own rules. It went live on day three and sorted 21 paid sales by the afternoon of day four.
- A training studio wanted every new client's first package sale sent from Mindbody to its own systems. The first live sale went through half an hour after signup.
- A gym owner wanted ActiveCampaign to know what Mindbody knows. New clients now arrive hourly through the day, tagged with what they actually do.
What would you ask for?
Details are taken from the account's own run history and shared without the studio's name.