Every service business knows what a job costs. Very few know what the space between jobs costs, and it is usually the second-largest labour expense in the company after the work itself.
The reason it hides is structural: drive time has no invoice line, no job number, and no customer attached. It is paid, it is real, and it shows up in the accounts only as a slightly disappointing gross margin nobody can quite explain.
The number, measured honestly
Take a normal week — not your worst, not your best. Count the jobs your team completed and the hours they were on the clock. Then estimate the hours actually spent on customer sites. The gap is drive, staging, and parts runs, and for most residential service businesses it lands somewhere between a fifth and a third of the paid day.
Put a wage against that and the number stops being abstract. A four-technician team at $34 an hour loaded, losing 90 minutes each per day, is spending roughly $51,000 a year on windshield time. That is not a rounding error — it is a fifth technician you are already paying for and cannot dispatch.
Where the hours actually go
When we look at real schedules, the loss almost never comes from one bad route. It comes from four smaller habits that each look reasonable in isolation:
- First-call-wins dispatching. Jobs get scheduled in the order they were booked, so Tuesday runs north, south, north again. Nobody chose this; the calendar just filled in arrival order.
- The emergency that reshuffles everything. One urgent call is normal. The problem is that inserting it usually means dragging three other jobs across town, and nobody re-checks the geography after the insert.
- The parts run. A technician arrives without the part, drives to the supplier, and drives back. Two trips, one job, and a customer whose morning was longer than promised.
- The unbooked return. Work that could have been finished on the first visit gets scheduled for a second one, and the second visit costs a full drive to recover an hour of labour.
Four changes, in the order they pay
- 01Give the week a shape before you fill it
Loose geographic days — north Monday and Thursday, south Tuesday and Friday, commercial Wednesday — are not a hard rule, they are a default. Bookings land in the right day without anyone optimising anything, and the exceptions become visible because they are exceptions.
- 02Make the emergency insert cheap
The cost of an urgent call is not the call; it is the reshuffle nobody re-checks. When the schedule shows travel between stops, whoever is dispatching can see instantly which of three technicians is genuinely closest rather than which one answers the radio first.
- 03Kill the parts run with the record, not with inventory
Most missing-part trips are not stocking failures — they are information failures. If the previous visit logged the model, the serial, and a photo of the plate, the truck leaves loaded. This is the single highest-return use of field notes in the business.
- 04Price and schedule around the second visit
Track how often a job needs a return trip and why. If the answer is "we did not know what we were walking into", the fix is upstream in the booking questions, not in the route.
CompleteDayWeekMonthOpenActiveClosedWarningOverdueShow Table
What routing software will and will not do
Dedicated route optimisation is real technology, and for a fleet running 15 or more vehicles with tight time windows it earns its cost outright. For a team of three to eight it usually does not, because the constraint is different: your schedule is not badly optimised so much as invisible. Nobody is choosing a poor route; nobody can see the route at all until the day is already running.
| Team size | What actually limits you | Where to spend first |
|---|---|---|
| 1–3 vehicles | One person holds the whole day in their head | A shared schedule everyone can see on a phone |
| 4–8 vehicles | The day is visible but travel is not | Geographic defaults and dispatch by proximity |
| 9–15 vehicles | Emergency inserts wreck the plan daily | Live reschedule with travel shown, plus tighter booking questions |
| 15+ vehicles | Genuine combinatorial routing problem | Dedicated route optimisation |
The compounding version
Recovering ninety minutes a day per technician does not simply save wages. It creates capacity you can sell at full margin, and it does so without hiring, advertising, or raising prices. In a good season that is the difference between turning work away and taking it.
It also makes the day quieter for the person dispatching, which matters more than owners expect. A schedule with slack absorbs one emergency without anyone making four apologetic phone calls.
You cannot route your way out of a schedule nobody can see. Visibility first, optimisation second — in that order, always.
Common questions
How much time do field technicians actually spend driving?
In a typical residential service week it runs between a fifth and a third of paid field hours once you include staging and parts runs. The reliable way to check yours is to compare paid field hours against hours genuinely spent on customer sites for one ordinary week.
What is the fastest way to cut drive time?
Give the week a loose geographic shape before bookings land in it — north days, south days, commercial days — so most jobs route themselves. It costs nothing, needs no software, and typically removes the worst zig-zag days immediately.
Do small service businesses need route optimisation software?
Usually not below about 15 vehicles. Under that, the binding constraint is visibility rather than optimisation: a schedule everyone can see on a phone, with travel between stops shown, recovers more hours than a solver would.
Why do technicians keep making parts runs?
Almost always because the previous visit did not record the model, serial, or a photo of the equipment. It is an information failure rather than a stocking failure, and it is fixed on the customer record rather than in the warehouse.