Using AI safely with guest data, then filling the quiet sessions
Before any dashboard or AI drafting, we close the gap where guest data was going into personal AI tools.
The problem
A bathhouse and sauna venue keeps guest names, phone numbers and health questionnaire answers in its booking system. Staff had started pasting booking exports into personal AI tools to write promotions, and the owner could not see where that data went. At the same time, weekday daytime sessions were running well under half full.
Data privacy first
- Company AI accounts on a business plan, so guest data is not used for training.
- Names and phone numbers masked before AI drafts anything.
- Health questionnaire answers left out of the data AI works from.
- A person approves every message before it goes out.
What the dashboard is for
Bathhouse and sauna bookings
One weekly view of session occupancy by day and time, first time guests who come back, and no shows, so the manager can see which sessions need filling and who to invite back.
What the numbers say
Overall occupancy is 67% against a 75% target. Weekday sessions from 9am to 3pm have room for 383 more guests every week.
- Session occupancy67%Target 75%
- Guest visits4,476Last four weeks
- First time guests returning27%Within 60 days, target 35%
- No shows and late cancels4.6%Under the 6% limit
How FDX uses it
Privacy first, then the data, then the drafts
Next steps it points to
- Quiet sessionsInvite members and past guests to a weekday daytime session. AI drafts the offer from booking history and the manager approves it before it goes out.
- First visitsSend every first time guest a thank you and a return invitation within a week, drafted with names and numbers masked.
- WeekendsOpen a waitlist for sold out sessions so late cancellations are filled.
ExampleExample case study with illustrative sample data and stock photos from Unsplash. Not a real client or real results.
Next exampleConstructionSeeing margin fade while there is still time to recover it
The dashboard is one part of the work.
It gives the team one trusted view of the numbers. The bigger gain is AI working from that same data to draft the follow ups, with a person approving every step.
When AI drafts from your data, personal details are masked before the model sees them. Onshore hosting is available, and your data is not used for training.