Free subscription audit
I took over a process that had stopped working and turned it into an automatic check of who has a free subscription and whether they should still have it. It has been running since early 2026 across the US, Spain, the UK, Ireland and France.
- Automated:BigQueryCollects the accountsEmployee accounts and free subscriptions, on both the old and the new billing model, with subscription details and basic account data.
- Automated:AutomationChecks ambassador locationsFor each ambassador, it finds the other accounts at the same location, which are also entitled to a free subscription.
- Automated:SlackCompares with the employee listOn every run. Accounts with no match are marked as possibly ready for removal.
- Automated:EmailSends me the reportEvery two weeks: employee accounts to remove, subscriptions to renew or cancel, and ambassador subscriptions waiting to be added.
- A person decides:MeReview and apply every change
The problem
Some Booksy accounts don't pay for a subscription: employee accounts, accounts on free subscriptions, and businesses at the same location as a Booksy ambassador. All of this lived in a file nobody kept up to date, and nobody checked whether free subscriptions should end. So some accounts that should have been paying weren't, people entitled to a free subscription got it late and then needed refunds, and former employees could keep using their employee accounts.
My role
When the process came to me, I chose to extend and automate it myself. I built it alone and I'm the one who uses the report. I also stay in touch with the ambassador managers in each market to keep the ambassador list current.
Key decisions
It only reports, it never deletes. Matching can be wrong, and removing a real employee's account costs more than a short delay. I make every change by hand.
It matches on more than one field. Slack and account data don't always agree, so it compares email in several formats, for example with a dot, plus first and last name.
What can go wrong
People sometimes entered random details on their accounts, which caused false matches. This was very rare, and newer account creation rules have mostly removed the problem. Because a person reviews every change, a false match doesn't remove anything.
Results
- First full update in the US: about 50 employee accounts and 80 free subscriptions removed.
- Other markets: roughly 5 to 20 free subscriptions and up to 15 employee accounts each.
- Over 100 free subscriptions ended across all markets. At a minimum of $30 a month each, that's at least $3,000 a month in subscriptions that are no longer free.
- The check now repeats by itself, and former employees don't stay on the list unnoticed.
Done by hand, a check like this simply wouldn't happen.