Why drive time is the quiet cost
Every service business sells the same scarce thing: a technician's hours on site. Time spent driving between jobs is time that cannot be billed, and it is paid for twice, once in wages and once in fuel and wear on the vehicle. Yet most owners can tell you their average ticket to the dollar and have only a rough feeling for how many hours a week their crew spends behind the wheel.
As of October 2026, the tools for fixing this are ordinary. GPS tracking and route optimization are standard parts of field service software, and an on-the-way text driven by live location is something customers now expect. The software is the easy part. The harder part is the set of decisions around it: how you carve up your service area, what appointment windows you promise, how you fit emergencies into a planned day, and how you track vehicles in a way your team accepts as fair.
This guide covers all of that. It starts with what a wasted mile actually costs, moves through four ways to route a day and the rules that make routes hold together, then deals with GPS tracking practice and employee notice, the link between live location and arrival texts, and the few numbers worth watching. The platform referenced is Run with Jarvis. Any figure that is not a published plan number is illustrative arithmetic and labeled that way.
What a mile really costs
Start by pricing drive time for your own shop. The arithmetic below uses invented round numbers. Substitute yours.
The time cost. Say you run four technicians, and each spends about two and a half hours a day driving. Say better routing trims thirty minutes a day from each. That is two hours a day across the crew. Over 22 working days, it is 44 hours a month.
Forty-four hours is not automatically 44 hours of extra work, because saved minutes arrive scattered across days and technicians. Be conservative and assume only a third of it turns into usable job time, roughly 15 hours. If a typical job takes an hour and a half on site, that is room for about ten more jobs a month. At an illustrative average ticket of $280, that is $2,800 in monthly revenue from the same crew, the same vans and the same payroll.
The vehicle cost. Say each van covers 40 miles a day. Four vans over 22 days is 3,520 miles a month. If tighter routes remove one mile in eight, that is 440 miles. At an illustrative all-in running cost of $0.70 a mile for fuel, tires, maintenance and depreciation, that is about $308 a month.
The cost you cannot see. A technician who arrives late and flustered does worse work and sells less. A customer who waited through a four-hour window and saw nobody is unlikely to call again. Neither appears in the arithmetic, and both are consequences of the same routing problem.
Two things stand out when you run your own numbers. The time saving is usually worth far more than the fuel saving, so the goal is more jobs, not just fewer miles. And small daily improvements add up, because they repeat on every working day for every technician.
The related question of how far you should travel at all, and what to charge when you do, is covered in the service radius and drive-time pricing guide.
Four ways to route a service day
There is no single correct way to route. There are four common approaches, and each fits a different kind of demand.
| Routing approach | How it works | Best for | Trade-offs |
|---|---|---|---|
| Fixed territories | Each technician owns a geographic area and takes the jobs in it | Steady, predictable demand over a wide area, recurring routes | Uneven workload when one area is busy and another quiet; weak for emergencies |
| Nearest available technician | Each new job goes to the closest qualified technician at that moment | Emergency and same-day work where response time wins the job | Can pull technicians across town repeatedly; planned work gets disrupted |
| Zone days | Routine work in each zone is booked only on set days of the week | Maintenance, inspections and non-urgent jobs that can wait a few days | Customers wait for their zone day; needs discipline at booking |
| Dynamic optimization | Software sequences each day's jobs for least drive time and re-sequences as jobs change | Mixed workloads with many stops per technician per day | Depends on accurate durations and addresses; technicians see the order change |
Fixed territories are simple and build familiarity. A technician who works the same neighborhoods learns the buildings, the parking and the customers. The weakness is rigidity. When the north side is slammed and the south side is quiet, a strict territory rule leaves one technician overbooked and another idle.
Nearest available is the natural rule for emergencies. The customer wants someone now, and the closest qualified technician is the right answer. Used for everything, though, it produces a day of zigzags, because each decision is made without regard to the next one.
Zone days are the most underused idea in routing. If routine work in the east zone is only offered on Tuesdays and Thursdays, every east-zone job lands on a day when a technician is already there. The customer waits a little longer and the route is tight by construction.
Dynamic optimization is what the software does: take the day's jobs and work out the sequence with the least driving, then adjust as cancellations and additions arrive. It is powerful, and it is only as good as the information you give it.
Most businesses end up with a blend. Routine work is planned by zone and sequenced by the optimizer. Emergencies go to the nearest qualified technician. Territories exist loosely, as a default, and bend when the workload is uneven.
The rules of good routing
Whatever blend you choose, a handful of rules decide whether routes hold together in practice.
Get the durations right. A route built on jobs that are scheduled for an hour and take two will collapse by lunchtime. Review your job types and set realistic durations, including the time to park, unload and write up. Routing accuracy depends more on honest durations than on clever sequencing.
Clean the addresses. An address that points to the wrong side of town, or to the middle of a ZIP code, throws off every calculation after it. Confirm the address when the job is booked, not when the technician is circling the block.
Promise windows you can keep. A narrow window pleases the customer and constrains the route. A wide one frees the route and annoys the customer. A reasonable middle is a two-hour arrival window for routine work, with an on-the-way text that narrows it on the day. Promise what your routes can deliver, then use live location to tell the customer more precisely.
Reserve room for emergencies. If every hour of every technician's day is booked with planned work, the first emergency call wrecks someone's schedule. Hold a small amount of capacity each day for urgent jobs. If no emergency arrives, fill it with a job pulled forward from tomorrow.
Decide the emergency insert rule in advance. When an urgent call comes in, who goes? The nearest qualified technician who can absorb it with the least disruption to promised windows. And what happens to the job they were heading to? It gets a call or a text with a new time, immediately. Writing this rule down stops the argument from happening on the phone.
Match skills before distance. The nearest technician is not the right one if they cannot do the job. Routing by distance alone sends people to work they have to walk away from, which is the most expensive drive of all.
Start and end the day sensibly. If technicians start from home, the first job should be near home and the last job on the way back. If they start from the shop, load the van the night before so the morning does not begin with an hour in the yard.
Stock the van to avoid the parts run. A mid-day trip to the supply house is a routing failure that no optimizer can fix. Knowing what is on each van, and booking jobs accordingly, saves more miles than re-sequencing stops.
Let the plan change, but not constantly. Re-sequencing in response to a cancellation is useful. Reshuffling a technician's afternoon every twenty minutes is demoralizing. Set a rule for how late in the day the order can change.
How these rules play out when several technicians share one board, including who assigns what and how the day is adjusted, is covered in the guide to CRM and dispatch for multi-tech operations.
Routing is won or lost at booking
It is tempting to think of routing as something dispatch does each morning. In reality, most of the route is decided days earlier, at the moment an appointment is offered.
When a customer calls for a routine job and is offered "any time this week", they pick whatever suits them, and the result is a scatter of jobs across the map. When they are offered "Tuesday or Thursday, when we are in your area", the same job lands next to three others.
That means whoever answers the phone, a person or an AI receptionist, needs to know the routing rules: which zones are served on which days, how many emergency slots are held, how long each job type takes. A booking system that offers openings without regard to location will undo any optimization that follows.
On Run with Jarvis, smart job booking and the calendar sit in the same system as GPS tracking and route optimization, so the appointment that gets offered and the route that gets driven are working from the same information. The argument for keeping those together, and the cases where separate tools still make sense, is in all-in-one platform vs disconnected point tools.
What GPS tracking is actually for
GPS tracking has a reputation as surveillance. Used well, it is closer to a shared map, and its benefits go to the customer, the dispatcher and the technician as much as to the owner.
Honest arrival times. The office knows where the van is, so the customer can be told the truth.
Faster emergency response. Dispatch can see who is actually closest, not who they think is closest.
Fewer interruptions for technicians. When the office can see a van is ten minutes out, nobody needs to call the technician to ask. That is a real benefit to someone trying to drive.
Proof of service. A recorded arrival time and time on site settles disputes about whether someone showed up, and it supports your side if a charge is contested.
Better durations. Actual time on site, collected over weeks, tells you how long each job type really takes. That feeds directly back into scheduling.
Safety. If a technician does not answer and has not moved in an hour, someone should check.
Framed this way, tracking is infrastructure for running the day, not a tool for catching people out. How you introduce it determines which of those it becomes.
GPS tracking etiquette and employee notice
Tracking people's movements is sensitive, and handling it badly costs you trust that is hard to rebuild. The following is general operational guidance, not legal advice. Rules on employee monitoring vary by state and locality, and you should confirm your obligations with an attorney before you begin.
Tell people, in writing, before you start. Explain what is tracked, during which hours, who can see it, how long the data is kept, and what it will and will not be used for. A written policy that each employee acknowledges is far better than a mention in a meeting. Some jurisdictions require notice or consent, which is one of the things to check with counsel.
Track the work, not the person. Limit tracking to working hours. If technicians take company vehicles home, decide and state whether tracking continues after hours, and lean toward not doing so. Tracking a personal vehicle or a personal phone raises additional issues and deserves specific legal advice.
Explain the why. Lead with the customer benefits and the benefits to the technician: accurate arrival texts, fewer calls from the office, evidence when a customer claims nobody came.
Limit who can look. Location data should be visible to the people who dispatch, not to everyone in the company.
Use it for patterns, not for gotchas. Reviewing a week of routes to fix scheduling is good management. Confronting someone about a seven-minute stop is how you teach a crew to resent the system. If there is a real performance concern, deal with it as a performance concern, with the data as one input.
Be consistent. Apply the same rules to everyone, including managers who drive company vehicles.
Give technicians the benefit of it. When the data clears a technician in a customer dispute, say so. People accept tracking much more readily when they have seen it protect them.
How ETA texts depend on live location
The on-the-way text is the most visible product of GPS tracking, and it is worth understanding why location is what makes it work.
Without live location, an arrival estimate is a guess made by whoever is in the office, based on where the technician was supposed to be. It is wrong whenever the previous job ran long, traffic was heavy, or the technician stopped for parts. A wrong estimate is worse than none, because the customer rearranged their morning around it.
With live location, the estimate is calculated from where the van actually is at the moment the technician heads out. The customer gets a time that reflects reality, and the message goes without anyone in the office having to remember to send it.
On Run with Jarvis, automatic ETA and arrival texts are included on every plan alongside GPS tracking and route optimization, starting with Core. The wording, timing and what to say when a technician is running late are covered in the on-my-way ETA text guide.
There is a routing lesson in this too. If your arrival texts are regularly showing times outside the promised window, the texts are not the problem. The durations or the windows are.
What to measure
Four numbers tell you nearly everything about whether routing is working. Track them weekly.
Jobs completed per technician per day. The headline. Better routing should raise it without lengthening the day.
Drive time as a share of the working day. If a technician is on the clock for eight hours and driving for three, more than a third of the day is on the road. Watch the direction, not just the level, since a rural service area will always carry more driving than a dense urban one.
On-time arrivals. The share of jobs where the technician arrived inside the promised window. This is the number the customer feels.
Miles per completed job. Total miles divided by jobs done. It adjusts for busy and quiet weeks in a way raw mileage does not.
Take a baseline for a month before changing anything. Then change one rule at a time, zone days first is a good choice, and watch which numbers respond. Changing several things at once leaves you unable to say what worked.
Drive time also belongs in your job costing. A job that looks profitable on parts and labor may not be once the hour of driving each way is counted. The method is in the job costing and profit-per-job guide.
Routing across more than one location
Once you run crews from more than one base, routing gains another layer.
Decide where the boundary is, and let it flex. A job near the line between two locations should go to whichever crew can reach it with less disruption that day, not to whichever side of an arbitrary line it sits on.
Share emergency coverage. If one location's crew is fully committed and the other has a technician close by, the customer does not care which office the van belongs to.
Keep one calendar. Two locations with two separate schedules cannot see each other's capacity, and work gets turned away by one while the other has an open afternoon.
Compare like with like. Measure each location on the same four numbers, and remember that geography explains some of the difference.
The wider set of questions that come with a second base, from staffing to reporting, is covered in the multi-location operations guide.
Common mistakes
Optimizing the sequence and ignoring the booking. The best optimizer in the world cannot fix a day of jobs scattered across the map by an unconstrained booking process.
Dishonest durations. Scheduling every job for an hour because it looks tidy guarantees a late afternoon.
No emergency capacity. A fully packed schedule has no room for the most valuable calls you get.
Windows too narrow to keep. A promise you break is worse than a wider promise you keep.
Springing tracking on the team. Turning on GPS without a written explanation invites suspicion, and in some places may breach notice rules.
Using location data to nitpick. It turns a useful tool into a grievance.
Measuring miles and not jobs. Fewer miles with the same number of jobs is a modest saving. The prize is more jobs.
Changing everything at once. You will not know which change helped.
Where to start
Pull last week's schedule and mark every job on a map, one color per technician. The zigzags will be obvious. Then check your job durations against what the jobs really took.
Pick one change. For most businesses that is zone days for routine work, with a small block of emergency capacity held each day. Write the emergency insert rule down. Write the tracking policy down, have it reviewed, and share it with the team before anything is switched on. Take a month of baseline numbers, make the change, and watch jobs per technician per day.
GPS tracking, route optimization and automatic ETA and arrival texts are part of every Run with Jarvis plan. Plans are on the pricing page: Core $500, Pro $750 and Elite $1,200 per month, with zero setup fees, unlimited users and month-to-month terms. To look at your own routes and booking rules together, get in touch.
