A clinic I work with moved off its old booking software last week. New system, new website, new calendar. All of it planned, tested twice, switched over on a Wednesday morning.
The part that took the longest had nothing to do with the new system. It was the 298 people sitting in the old one.
That is the thing nobody budgets for when switching software. You plan the new tool. You compare the pricing. You test the booking flow. Then you look at the old account on the day you are about to cancel it, and realise it is holding years of who your customers are, when they last came in, and why.
What actually lives in your old system
Pull up whatever you are about to leave and look properly. In this clinic's case, the old account was holding:
- 298 real clients, going back to mid-2025
- 414 completed appointments, each with a date
- Which practitioner each person saw
- The reason they first came in, in their own words
- Health questions they had answered on intake
None of that is in the new system unless somebody moves it. And "somebody" is usually the owner, at 11pm, a week after the switch, when the old subscription has already been cancelled.
The order that makes this survivable
We did it in a deliberate sequence, and the sequence is the whole lesson.
Export before you decide anything
Get the data out of the old system while you still have a paid, working login. Exports are the one thing that gets harder every single day you wait. We pulled two exports — one of the people, one of the appointments — before anything was switched off.
Keep the old system alive for 30 days
The clinic's old website and booking pages are still running right now, hidden from the public. Nothing points to them. Nobody can book through them. But they exist, and they will keep existing until mid-October.
That is not indecision. That is the rollback. If something in the new setup turns out to be broken in week two, there is somewhere to go back to. The cost is one more month of a subscription you were cancelling anyway.
Decide what you are NOT moving
We deliberately left the health-screening answers in the old export file. They are sensitive, the new system is not set up to hold them safely, and nobody needed them to run a clinic next week.
Moving everything is not the goal. Moving what the business runs on is the goal. Write down what you left behind and why, so that in six months you know it was a decision rather than an accident.
What did the customer history actually buy?
This is the fair question, and it has a concrete answer.
Because the visit history came across, the clinic now knows who has been coming for two years and who came once and vanished. It knows which of its 76 Google reviews came from people already on the list — 33 of them — so the review-request campaign skips those people instead of asking them for a review they already wrote.
None of that is possible from a fresh start. A fresh start means every customer is a stranger on day one.
The bit that surprised the owner
Five appointments were already booked in the old system for dates after the switch. Real people, real dates, several weeks out.
They did not automatically move. Why would they? Two different companies' software, no connection between them.
We caught it because someone read the old calendar one more time before pulling the plug. Check your forward bookings before you switch anything off. It is a five-minute job, and the alternative is a customer standing outside your door on a day your new system says you are closed.
The takeaway
Switching software is not a technology project — it is a moving day, and your customer history is the box nobody labels.
If you are looking at a new system this quarter: export first, keep the old one breathing for a month, and write down what you chose to leave behind.