Quick answer: Look at bookings by source and branch, cancellations, driver activity, revenue by pricing band, driver balances, and expiring documents. Once a week. That's it. Six things, maybe twenty minutes, and you'll catch most problems before they get expensive.
Okay. Now the longer version.
Picture a fleet owner on a Monday morning. Coffee's going cold. The dashboard's open, and it's full: charts, tables, little coloured arrows pointing up and down. Where do you even look? Most people check total bookings, nod, and close the tab. I get it. But that one number is a bit like checking your bank balance and calling it financial planning. It tells you where you are, not how you got there. So here's what I'd look at instead, using any decent taxi fleet management software that has proper reporting built in.
1. Where bookings come from
Not how many. Where from.
Phone, app, website, corporate accounts, partner work. If you've got more than one branch, split it by branch too. Say the app numbers keep creeping up week after week. Good sign, your passenger app is catching on. Now say phone bookings are flat and app bookings aren't moving at all. That usually means something's stuck, maybe nobody's telling customers the app exists, or the first booking is clunky enough that people give up and just ring the office again. One branch running behind the others? Worth a chat with whoever manages it. This week, not in three months.
2. Cancellations (and when they happen)
A cancellation before any driver accepts is no big deal. A cancellation after a driver's already driving to the pickup? That one costs you. Fuel, time, and a driver who's now in the wrong part of town for the next job. Don't just count them. Look at the pattern. Lots of cancellations around 6pm? Specific zone? Mostly one booking source? A cluster usually points at something fixable, like long waits in one area or a fare that looks higher than people expected once they see it on screen.
3. Driver activity
How many drivers were actually out working this week versus last week? And are jobs getting picked up quickly, or sitting around unassigned? Here's the bit people get wrong. This isn't about catching anyone out. If a driver who was busy last month has gone quiet, that's a reason to pick up the phone and ask how things are going, not to send a warning. Maybe they're fed up with something. Maybe they're ill. Maybe a competitor offered them a better deal. You'd rather know early.
4. Revenue by pricing band and zone
Total revenue is one number. Revenue split by pricing band and zone is a whole story. Maybe one zone brings in loads of bookings but the margin's thin, so that fixed fare probably needs another look. Maybe your peak pricing kicks in exactly when you set it, but bookings in that window aren't any higher than normal, which means the surge isn't catching extra demand at all, it's just charging people more for the same thing. (Customers notice that, by the way.)
You can only spot this if the report's broken down. A single big total hides everything.
5. Driver balances and unpaid invoices
Boring. Yes. Do it anyway.
Small mismatches are easy to fix in week one. Leave them for a quarter and suddenly nobody remembers what happened, the driver thinks you owe them money, you think they've been paid, and it turns into a proper argument. A quick weekly glance at balances and adjustments avoids most of that. It also matters for trust. Drivers stick around when the money side is clear. Nothing makes someone start looking for another firm faster than chasing their own earnings.
6. Documents that are missing or about to expire
Takes thirty seconds. Check how many driver and vehicle documents are missing, close to expiry, or already expired. Catch an insurance renewal today and it's a quick reminder. Miss it and you've got a driver who can't legally work next Tuesday, probably the busiest night of your week, because that's how it always goes.
A small bonus: support tickets
If you track them, scan through quickly. One complaint about a pickup point is just a bad day. Four complaints about the same pickup point in a fortnight is a pattern, and patterns are where the fixable problems hide.
Making it a habit
Same day, same time, every week. Fifteen to thirty minutes. The point of repeating it is that you learn what "normal" looks like for your fleet, and then anything odd jumps out. If you only look now and then, every figure just looks like a figure. Keep a scrappy note while you're at it. Two lines is fine. "Cancel up in the east zone, check wait times, fixed." After a few months that note is worth more than any dashboard, because it shows what you actually tried and what worked.
When something looks wrong
Don't panic. One strange week could easily be weather, a football match, or three drivers on holiday at once. Two or three weeks of the same movement? Now it's a trend, and it deserves a proper look. Check the neighbouring numbers first. Cancels jumped? Look at wait times and zones before blaming anything.
What CabTaxi covers here
CabTaxi's reporting includes bookings by source, branch and period, plus driver records, weekly invoices, ledgers, document status and ride-linked support tickets, all inside the same platform. So your Monday review doesn't mean opening five different tools. Want to see what your own weekly check would look like? Book a live CabTaxi demo and bring a typical week from your fleet. We'll walk through it together.
Frequently Asked Questions
- When checking fleet reports, how often should I do it?
Most operators are covered by weekly suits. Daily is loud and monthly is not quick enough to detect a problem before it gets big.
- What number is the most important?
If you look at bookings split by source and branch, this will tell you more than just bookings, as long as the business is a business.
- May driver activity data be used to monitor driver activity?
Use it to identify those who may require assistance, not as a way to police people. Close supervision will chase away good drivers.
- What's the difference between a trend and a single occurrence?
An odd week may be a weather week, or a special event. If that movement is appearing for a couple of weeks, investigate it thoroughly.
- Is an analyst required for this?
No. Most operators need only a small set of numbers, which are checked every week on the same day.
- Why bother to divide revenues by pricing band and zone?
It provides you with an insight into exactly where your money is going, and if your pricing strategy is achieving what it's supposed to.
