Switching field service software without losing your customer history
The export is the easy part. It takes an afternoon. The hard part is the month either side of it, and the things that were never in the export to begin with.
I have done this move. What follows is the order I would do it in again.
Pull everything before you tell anyone you are leaving
Run your exports while your account is fully paid, fully active and nobody is annoyed with you. Not after the cancellation email. Not on the last day.
This is not about expecting anyone to behave badly. It is that a downgraded or lapsed account can lose features, export can sit on a pricing tier, and support gets slower once you are on the way out. What each platform will actually give you is worth knowing before you start.
Pull customers, job history, invoices and the price book, and put them somewhere that is not the new system and not the old one. A folder on your own machine, backed up. That folder is your insurance for the whole move.
Open the files. Actually open them.
An export you have not opened is a rumour.
Open the customer CSV and check the row count against what the platform says you have. Check that addresses survived, and specifically that customers with more than one property still have all of them — multiple addresses are the single most common thing to arrive mangled, because a spreadsheet is a bad shape for one customer with three houses.
Then look at what is missing. On most exports it will be job photos, job notes, attachments, and any message thread you had with a customer through the platform. That is normal, and it is also where most of what your shop actually knows is written down.
Decide what you are willing to lose
You will not move all of it. So decide deliberately rather than finding out later.
For us the answer was: customers and their history move, invoices move as a record, photos do not, and old job notes get triaged. I went through the customers we work on repeatedly — the ones with the awkward boiler, the crawlspace shutoff, the dog — and re-typed the notes that mattered into the new system by hand. It was an evening. It was worth it.
Everything else stays in that export folder. Not migrated, just kept, in case someone asks in two years.
Run both for a month
Pick a date. From that date, every new job goes in the new system. Every job already open finishes in the old one.
Do not migrate work in progress. A job that is half-invoiced in one system and half-scheduled in another is how you double-bill somebody or fail to bill them at all.
That means paying for both for a month. It is the cheapest part of the whole exercise and it is the part people try to skip.
Leave the open invoices where they are
Anything already sent to a customer stays in the old system until it is paid. The payment link works, the customer recognises the invoice, and nobody has to explain anything.
Re-issuing a live invoice out of a new system with a new look and a new payment link is a good way to have a customer sit on it for another three weeks, or call you to ask whether it is real. You did not switch software to get paid slower.
The crew is the actual migration
The data move is a technical job with a right answer. Getting five people to stop opening the old app is a different kind of problem, and it is the one that decides whether this works.
Two things helped. Pick the date and say it out loud, more than once, before it arrives. And on that date, take the old app off the phones. Not “stop using it” — off. If it is still on the home screen, somebody tired at 4pm on a Friday will tap the one they know.
You still have to quote jobs during the changeover
This is the gap nobody plans for. For a few weeks you are half in one system and half in another, and a customer is still standing in front of you waiting for a number.
That is worth having a tool for that does not care which platform you are on. I built Dead Hand partly for this: your rates and your margins live on your phone, you build the quote on the step, and it does not know or care what is happening with your CRM this month. It is $15 a month and it is completely independent of whatever you are migrating.
Whatever you use, have something. The changeover is exactly when a shop starts guessing at prices again, and guessing is expensive in a way that does not show up until the job is done.
The order
- Export everything while the account is healthy, not on the way out.
- Open the files and count the rows. Check multi-address customers.
- Decide what you are willing to lose, and hand-carry the notes that matter.
- Pick a cutover date. New work in the new system from that day.
- Finish open jobs and open invoices where they started.
- Take the old app off the phones on the date.
- Keep a way to quote that does not depend on either system.
- Cancel only once the last old invoice is paid.
Cancel last, and in writing
Get the cancellation confirmed in writing, with a date, from a person. Keep it. Check the next statement, and the one after that.
I would not have thought to say that a couple of years ago. I say it now because I have had a subscription set to renew that I had asked twice to stop, and the thing that settled it was having the email.
Common questions
How long does it take to switch field service software?
Plan on a month with both systems running, not a weekend. The export takes an afternoon. What takes the month is finishing the jobs that are already open in the old system and getting the crew off the old app.
What does not transfer when you switch field service software?
Usually job photos, job notes and attachments, message threads with customers, and anything a customer approved inside the old portal. Exports tend to cover customers, jobs and sometimes invoices. The knowledge in the notes is the part shops miss.
What should I do about open invoices when switching?
Do not migrate them. Let the old system finish collecting what it already sent, and start clean in the new one on a fixed date. Chasing a re-issued invoice through a new payment link is how you lose money and confuse a customer at the same time.
Disclosure: Friendly is mine. Everything above is how I would move a shop onto anything, including something that is not. It opens next year; leave your email.