What Drivers Actually Cost You

Ask an outcall operator what a driver costs and you get a number per run, or a number per hour.

Both are the wrong unit. A driver does not cost you per run — they cost you per shift, and what you get back is the runs that shift produced. The gap between the two is where the money is.

The number nobody keeps

Utilisation = time spent on runs ÷ time on shift

Most operations have never computed it, because the second half is on a rota and the first half is nowhere.

When it does get computed, it usually comes in lower than anyone expected, and the shape is worse than the number: idle time clusters in the same hours every week, and those hours are not the ones people assume.

Cost per run is a trap

Cost per run = shift cost ÷ runs completed

This number improves when you assign more runs to fewer drivers, which is exactly what makes turnaround worse. Optimise it and you will slowly strangle the thing it was supposed to protect.

Use it alongside turnaround, never alone. The pair tells you whether you are efficient or just short-staffed:

Cost per run down + turnaround up   = short-staffed, and the guest is paying for it
Cost per run down + turnaround flat = a real gain
Cost per run up   + turnaround down = you added coverage. Was it worth it? See below

The two places the money actually goes

One — paid time where nobody is moving.

Not all idle time is waste. A driver positioned near where the next call will come from is doing something. A driver parked at the office because nobody knows where anyone is, is not.

The distinction is entirely a records problem. Without live status you cannot tell the two apart, so every hour looks the same on the rota and every hour costs the same.

Two — runs that should not have been assigned.

A run given to the driver who was easy to reach rather than the one who was close. A second car sent because nobody could confirm the first had returned. A run that arrived so late the booking cancelled, which you paid for and got nothing back.

These do not show up as a cost line. They show up as "a busy night that did not make money."

The cheapest driver is often the most expensive

Rates vary, and the instinct is to fill shifts with the cheapest hands available.

A driver who does not update status       costs you every dispatcher call to find them
A driver who does not know the area       costs you minutes on every run
A driver who is unreliable on Fridays     costs you the whole Friday, not one shift

The unit that matters is cost per completed run at an acceptable turnaround, not the hourly rate. A driver at a higher rate who returns clean status and knows the roads is usually cheaper by that measure — often by a wide margin.

What to measure before changing anything

Runs per shift, by driver and weekday
Utilisation, by hour of the night
Turnaround split three ways   phone→confirmed, confirmed→depart, depart→arrive
Cancellations attributed to lateness

The third line is the one that finds the money. If confirmed→depart is the largest of the three, the problem is not driving and not headcount — it is that nobody knows who is free. That is a records fix, not a hiring one, and it is close to free.

Coverage: how many is right

There is no general answer, but there is a general method.

1  Plot bookings by hour, over four weeks
2  Plot driver-hours over the same window
3  Find the hours where bookings peak and coverage does not
4  Move one shift. Do not add one yet

Move before you add. Most rooms are not under-staffed overall; they are staffed to a shape that no longer matches demand — usually because the rota was set when the business looked different and nobody revisited it.

⚠️ Adding a driver to fix a records problem makes the records problem permanent. You will have bought your way to the same turnaround with more cost, and the underlying cause will be harder to see because the symptom is gone.

When a run loses money

Worth computing once, for a normal run:

Contribution = booking value × your margin
             − driver cost for that run
             − the cost of what that driver could not do meanwhile

The third term is the one people leave out. A long run to the edge of your area is not just its own cost; it removes a driver from the map for an hour during which two closer bookings had to wait or go elsewhere.

Some areas do not pay. You will not know which until the record exists.

The system we built

tasteck is a booking and analytics system for night venues and outcall operations, built by people who ran them for sixteen years and grew from ¥200 million to ¥1.2 billion a year — six-fold, by attacking the business with systems rather than by pushing harder on sales.

What it does here: a dispatch board with status on every run — requested, confirmed, dispatched, arrived, returning, done — updated by drivers from their own screen, so the board is current without the dispatcher phoning anyone. Driver shifts submitted and confirmed, so shift hours and run time sit in the same place and utilisation is a query. Bookings carry the area, so run distribution by area is visible. Settlement broken out by line, including transport, so a run's cost is not buried in a total.

What it does not do: it is not a routing engine and computes no optimal route. It does not handle driver payroll or vehicle costs. It does not decide your coverage. What it supplies is the record that makes the decision arguable.

The output no one else produces

tasteck outputs the maximum you can spend on each marketing channel next month, as an amount in your currency.

Driver cost and channel spend are the same decision seen twice. A channel that delivers bookings at the far edge of your area costs more to serve than one delivering bookings next door — and a per-booking acquisition figure cannot see that. The ceiling is computed from the lifetime value of the guests each source actually delivered, which is the only view that nets out what serving them cost.

Nothing else in the nightlife category produces that figure.

Ask it from ChatGPT

tasteck connects to ChatGPT over MCP: ask your numbers as a question and the answer comes back in the chat — utilisation by hour last Friday, which area takes longest to reach, which driver returns status cleanly.

Multi-language is built in, the operating surface itself, with your language set during onboarding.

From $34 a month for up to two locations. Thirty days free on every plan, cancel any time. → Pricing

Start this week

1. Compute utilisation by hour for one week. Time on runs ÷ time on shift.
2. Split turnaround three ways. If confirmed→depart is the biggest, stop hiring and fix the board.
3. Plot bookings against driver-hours over four weeks. Look for shape, not total.
4. Move one shift into the gap. Measure for two weeks before adding anyone.
5. Write down what a run to your furthest area actually returns. Some of them do not.

Read next

• Outcall Dispatch: Turnaround Time Is the Whole Business
• What a Quiet Phone Is Costing You
• Night Venue Scheduling: Why the Roster Never Matches the Room
• No-Shows: The Money That Never Walks In
• Outcall Dispatch Software

---

On benchmarks. No target utilisation rates, cost-per-run figures, or coverage ratios appear in this guide. We do not have a dataset broad enough to publish them, and they vary enormously by city, density and how bookings arrive. One week of your own utilisation, split by hour, will show you a gap no industry average can.