How to Switch Taxi Dispatch Systems Without Losing a Day’s Work
The Short Answer
Switching dispatch systems takes one to three weeks for most UK fleets. Export your passenger list, account customers, driver and vehicle records and forward bookings first; build zones and tariffs in the new system and test them against last month's real jobs; train drivers before go-live, not on the day; then cut over on a quiet weekday morning with both systems available. Never cut over on a Friday.
The fear that keeps operators on a system they dislike is not the cost of the new one — it is the image of a Saturday night where nothing dispatches and the phones do not stop. That is a legitimate fear and it is entirely avoidable, because every bad migration fails in one of about five predictable ways.
Week 1: get your data out
Start here, because it is the step most likely to be slow and least within your control. Request, in writing, an export of:
- Passenger records — name, phone, saved addresses, and any notes ("front door only", "assistance needed", "no dogs").
- Account customers — company details, billing addresses, cost centres, agreed rates and PO requirements.
- Drivers and vehicles — licence numbers, badge and plate expiry dates, insurance and MOT dates, vehicle types and accessibility.
- Forward bookings — every scheduled and recurring job past the planned cutover date. School runs and hospital contracts are what hurt if they are missed.
- Historic jobs — at least 12 months, for reporting continuity and any dispute that surfaces later.
Clean it before you import it
A migration is the one moment you get to throw away a decade of duplicate passenger records and drivers who left in 2021. De-duplicate on phone number, drop anything with no activity in two years, and fix the addresses that never geocoded properly. Importing mess costs more later than cleaning it now.
Week 1–2: build and test the new system
- Zones first. Draw them to match how your drivers actually think about the town, not the council ward boundaries. Set adjacency so cascades go somewhere sensible.
- Tariffs next. Rebuild every one, including the awkward ones — airport runs, waiting time, bank holidays, school contracts, the account rate you agreed in 2019 and never wrote down.
- Then test them against reality. Take fifty real jobs from last month, price them in the new system, and compare to what you actually charged. Differences are either a tariff bug or a pricing decision nobody documented. Both are worth finding now.
- Set up payments. Connect Stripe, run a live card payment for £1, refund it, and confirm it appears where you expect.
- Configure compliance alerts with the expiry dates from the export, and check the first alert actually reaches a human.
Week 2: train the people
Controllers and drivers need different things, and both need them before go-live rather than during it.
| Who | What they need | When |
|---|---|---|
| Controllers | Two hours hands-on taking real bookings in a test environment | 3–5 days before cutover |
| Drivers | The app installed, logged in, and one practice job completed | A week before, in person if possible |
| Office/accounts | One settlement run and one account invoice, start to finish | Before the first month end |
Get the driver app installed early and make it a condition of the next shift. The most common go-live problem is not software — it is fifteen drivers at 6am who never downloaded the app and cannot log in.
Cutover: pick the right morning
- Choose a Tuesday or Wednesday morning. Never a Friday, never the start of a bank holiday, never the week the schools go back.
- Enter forward bookings into the new system in advance and reconcile the two lists the night before.
- Keep the old system readable for at least a month. You are not going back to it, but you will want to look things up.
- Have the vendor on the phone for the first two hours — not on email, on the phone.
- Run a manual fallback for the first day: a printed driver list with mobile numbers, so a controller can allocate by phone if anything stalls.
“The firms that have a bad switch are almost always the ones who cut over on a Friday because that was when the contract ended. Move the contract, not the cutover.”
The first week after
Expect a dip. Controllers will be slower for three or four days and drivers will grumble for about a week; both are normal and both pass. What is not normal, and what you should escalate immediately, is jobs failing to dispatch, fares pricing wrongly, or payments not reconciling.
Watch these four numbers daily for the first fortnight:
- Unallocated jobs per day — should fall to near zero by day five.
- Average time from booking to driver accepted.
- Card payments that failed or were not captured.
- Driver support calls to the office — the best single proxy for how the app is landing.
We migrate your passengers, accounts and forward bookings as part of onboarding.
Talk to Us About SwitchingWhat it costs you to wait
Operators routinely stay three or four years longer than they meant to, because the migration feels risky and the current system is merely annoying. It is worth being concrete about the other side of that ledger: the jobs lost to a passenger app you do not have, the controller hours spent on work that should be automatic, and the renewal price that goes up every year on a contract you have stopped negotiating.
If you are still at the comparison stage, the 24-point checklist covers what to test in a demo, and what dispatch software actually costs covers the pricing models. If you want to see the destination first, our taxi dispatch software page shows the console itself.
Frequently Asked Questions
- How long does it take to switch taxi dispatch systems?
- One to three weeks for most fleets under 50 vehicles. The software setup is rarely the bottleneck — the time goes on getting a clean data export from your current provider, rebuilding tariffs, and training drivers before go-live.
- Will I lose my booking history when I switch?
- Not if you export it first. Request passenger records, account customers, driver and vehicle details, forward bookings and at least 12 months of historic jobs in writing before you give notice. You have a right to the personal data under UK GDPR, though not necessarily in a convenient format.
- When is the best time to go live on a new dispatch system?
- A Tuesday or Wednesday morning. Avoid Fridays, bank holiday weekends and the week schools return — you want your quietest predictable period, with the whole working week ahead of you to resolve anything that surfaces.
- Do drivers need retraining on a new dispatch system?
- Some, but less than operators fear. Get the app installed and one practice job completed a week before cutover; the accept-navigate-complete flow is similar across platforms. Budget more time for controllers, who need a couple of hours hands-on before go-live.
Written By
Cabica
Taxi and private-hire software
Cabica builds dispatch, driver and passenger software for independent taxi and private-hire operators.