Home services dispatch: how to schedule and route techs efficiently

Scheduling vs dispatch

Scheduling and dispatch get used interchangeably, but they are different jobs, and mixing them up is behind a lot of bad days.

• Scheduling happens when the customer calls or books online. The CSR picks a date and an arrival window based on the job type and how much room is left that day.
• Dispatch is deciding which technician goes to which job, in what order, and then adjusting all day as jobs run long, parts are missing, emergencies come in and techs call out sick.

In a two-truck shop, the owner often does both from a phone. As you grow past a handful of techs, the two roles start to separate: CSRs fill the schedule and a dispatcher (sometimes a dedicated person, sometimes a service manager) owns the board. The handoff between them is where things go wrong. If CSRs book without knowing what the board looks like, the dispatcher spends the day cleaning up promises they didn't make.

A useful way to think about it: scheduling decides how much work goes into a day, and dispatch decides how that work gets done. Get the first one wrong and no amount of clever dispatching fixes it.

What inefficient dispatch costs

Dispatch waste is easy to miss because it doesn't show up as a line item. It shows up as techs who were "busy all day" but closed fewer jobs than they should have, overtime that crept up, and callbacks nobody tracked back to the wrong tech being sent.

Illustrative example: a residential HVAC and plumbing company runs 8 service techs. Suppose each tech loses about 45 minutes a day to avoidable drive time, crisscrossing town because jobs were dispatched in booking order. That is 6 hours a day across the team, or roughly one full tech's worth of capacity. If an average service call generates $400 in revenue and that recovered time allowed even one extra call a day across the whole team, that's about $8,000 a month over a 20-day month, before counting fuel, vehicle wear and overtime.

Now add first-time fix rate. If one in ten jobs needs a second visit because the tech didn't have the skill or the part, every one of those return trips is a slot that could have gone to a paying customer, plus a customer who is less happy than they would have been.

These numbers are hypothetical, and yours will differ. The exercise is worth doing with your own data because it tells you whether dispatch is a small annoyance or one of the bigger levers in the business.

Start with capacity planning

Capacity planning means knowing how many hours of each kind of work your team can actually deliver each day, and booking against that number instead of an open calendar. It is the single most useful thing most shops can do for dispatch.

A simple version you can set up in a spreadsheet or in your field service software:

1. Count real available hours. Techs on shift, minus time off, training, warehouse runs and the first and last stretch of the day that goes to loading and paperwork. Eight-hour shifts rarely mean eight bookable hours.
2. Estimate job durations by type. A maintenance visit, a no-cool diagnostic, a water heater swap and a drain clearing each have a typical range. Use your own job history, not guesses; most field service platforms let you report on time on site by job type.
3. Add an allowance for drive time. Based on your service area. A dense suburban market and a rural one look very different.
4. Hold back a buffer. In peak season, keep some share of the day unbooked for emergencies and overruns. Many shops start somewhere around 10% to 20% and adjust based on how often the buffer gets used.
5. Close the day when it's full. When booked hours hit capacity, CSRs offer the next available window or a waitlist, instead of squeezing in one more.

That last step is the hard one culturally. Turning away a same-day call feels like leaving money on the table. But booking a job you can't reach means a missed arrival window, an angry customer and a tech working late, which often costs more than the job was worth. Capacity planning also pairs with your phones: if you're answering every call quickly (see what missed calls cost a home service business), you need a schedule that can absorb what comes in.

Match the job to the right tech

Sending the closest available tech is the instinct. Sending the right tech is usually what pays. A few attributes are worth recording for every technician and checking on every assignment:

Factor	Why it matters	How to capture it
Skills and certifications	A commercial rooftop unit, a tankless water heater or a gas line needs someone qualified for it	Skill tags on each tech; required skill on each job type
Truck stock	The right tech without the right part is a second visit	Standard truck inventory; note likely parts on the booking
Job type and sales opportunity	Replacement-age equipment may warrant a tech comfortable presenting options, or a comfort advisor	Equipment age and history captured at booking
Customer history	Repeat customers and members often prefer the tech they know; callbacks should go to someone who can close them out	Notes on the customer record
Location and start point	Techs who take trucks home start the day in different places	Home addresses or start points on file

Good booking questions make matching possible. "What's going on with the system, how old is it, and is anything leaking or making noise?" gives the dispatcher far more to work with than "AC not working." A short script for CSRs, tuned per trade, pays off at the dispatch board and in first-time fix rate.

Route by area, not by booking order

The simplest way to cut drive time is to stop dispatching jobs in the order they were booked. Instead:

• Zone your service area. Split it into a few regions and, where you can, assign techs a home zone for the day.
• Steer the schedule toward zones. When a customer is flexible, CSRs can offer windows on days when a tech is already nearby.
• Sequence each tech's day. Order jobs so the route moves outward and back, or across the zone, rather than bouncing between ends of town.
• Use arrival windows, not exact times. Windows of two to four hours give the dispatcher room to sequence efficiently. Exact appointment times should be the exception, priced or reserved for situations that really need them.

Route optimization tools can help, especially past 10 or so techs, but routes are only as good as the job durations and skill rules feeding them. A perfectly optimized route that sends an apprentice to a complex diagnostic is still the wrong route.

Handling emergencies and jobs that run long

Every plan breaks by mid-morning. The question is whether your team has rules for what happens next or makes it up under pressure.

Define what counts as an emergency. For example: no heat when it's below freezing, no cooling for an elderly or medically vulnerable customer in extreme heat, active water leaks, sewage backups, and anything involving a gas smell (where the customer should first be told to leave the house and call the gas utility or 911). Write the list down so CSRs aren't deciding on the fly.

Decide what can be moved. Maintenance visits and non-urgent estimates are usually the first to slide. Members, callbacks and jobs where a customer took time off work usually aren't. Make this a ranking the dispatcher can apply quickly.

Call bumped customers immediately. The damage from a moved appointment comes mostly from the customer finding out late. A call or text as soon as the decision is made, with a new window and an apology, keeps most of them.

Watch for jobs running long. If a tech is an hour over the estimated duration, the dispatcher should know without having to ask, and should decide whether to reassign that tech's next job before the customer's window passes. Many shops lose the afternoon because nobody noticed the 10 a.m. job turned into a three-hour repair.

Emergency and after-hours pricing, if you charge it, should be explained when the call is booked, not on the invoice.

Rules for the dispatch board

The dispatch board is the live picture of the day: every tech, every job, every status. Whether it's a screen in ServiceTitan or a whiteboard in the shop, it works only if everyone treats it as the single source of truth.

1. One board, one owner. One person has final say on assignments at any given time. Techs can raise issues, CSRs can suggest, but moves go through the dispatcher.
2. Techs update status in the app. En route, arrived, working, done. If statuses lag, the board is fiction and the dispatcher is back to phoning trucks.
3. No side deals. A tech who agrees with a customer to "swing by Thursday" outside the system creates an invisible job. Every visit goes on the board.
4. Flag at-risk jobs early. Unassigned jobs, jobs with no confirmed part, and jobs whose window will be missed should be visibly marked, not discovered.
5. Review the next day before you close today. Ten minutes at the end of the day to check skills, parts and routes for tomorrow prevents a messy 7 a.m.

Keeping customers in the loop

For the homeowner, dispatch quality comes down to one question: did they know when you were coming? Clear communication covers a lot of operational imperfection.

• Confirm the day before with the window and a way to reschedule.
• Send an on-the-way message with the tech's name, ideally a photo, and an estimated arrival.
• Tell them early about delays. A text that says "running about 45 minutes behind, still on for today" prevents most angry calls.
• Close the loop after the visit with the invoice, any quote for future work and, when it went well, a review request (see how to ask for reviews) and a follow-up system for open quotes.

General information, not legal advice. Automated texts to customers in the US are subject to consent rules and carrier registration requirements. Get permission to text when the job is booked, honor opt-out requests immediately, and check the requirements with your texting provider.

Dispatch in ServiceTitan, Jobber and Housecall Pro

The common field service platforms all give you the basic tools: a scheduling calendar or dispatch board, job assignment, a technician mobile app with job status, and customer notifications. ServiceTitan is generally aimed at larger, multi-trade shops and offers more depth around capacity and dispatch, while Jobber and Housecall Pro are popular with smaller teams that want something simpler. Specific features such as route views, capacity settings, GPS tracking or automated arrival texts vary by product and plan, so check what yours includes before paying for an add-on.

Two practical points apply whichever platform you use:

• Set it up properly. Job types with realistic durations, skill tags on techs, service zones and notification templates. Many shops use a fraction of what they already pay for because nobody had time to configure it.
• The software doesn't make decisions. It shows the board and enforces some rules, but someone still has to notice the job running long, decide who to bump and call the customer. That ongoing judgment work, repeated dozens of times a day, is what wears down a dispatcher.

Where AI dispatching helps

AI dispatching is useful when the volume of small decisions outgrows the people making them. An AI agent connected to your field service software can watch the board continuously and take on routine work that otherwise falls through the cracks.

Task	Agent handles	Person decides
New job assignment	Suggests the best-fit tech based on skills, zone, capacity and parts	Dispatcher approves, especially for large or unusual jobs
Jobs running long	Spots overruns early and proposes a reshuffle	Dispatcher confirms moves that affect other customers
Customer updates	Sends confirmations, delay notices and new windows; handles routine reschedule replies	Unhappy customers and complaints go to a person
Emergencies	Flags likely emergencies from the booking notes and shows open capacity	Dispatcher decides priority and who gets bumped
Next-day prep	Checks tomorrow's jobs for missing skills, parts or information	Service manager resolves gaps

The key design choice is human-in-the-loop control: the agent does the watching, drafting and routine messaging, and a person approves anything that changes pay, affects a member or customer relationship, or involves safety. Good exception handling matters more than clever routing. An agent that knows when to stop and ask is worth more than one that tries to optimize everything.

This is the kind of work Tarvora does on our home services engagements. We build the agent inside the tools you already run, show you a free working prototype against your own board and job history first, and are paid from a share of the savings it measurably produces. If you're weighing where to start, our practical guide to AI implementation for small businesses covers how to pick a first process.

The numbers to track

Pick a handful of numbers, measure them for a month before changing anything, and review them weekly with whoever owns the board.

• Drive time per tech per day: total windshield time, or drive time as a share of paid hours. This is the most direct measure of routing.
• First-time fix rate: share of jobs completed on the first visit without a return trip. Low numbers point to skill matching, truck stock or poor booking information.
• Jobs per tech per day: by job type. Useful, but read it alongside first-time fix rate and revenue per job, since more short visits with more callbacks is not progress.
• On-time arrival rate: share of jobs where the tech arrived inside the promised window.
• Capacity utilization: booked hours versus available hours. Consistently near 100% with lots of overtime means you need more capacity or a firmer cutoff; consistently low means demand or booking is the problem, not dispatch.
• Reschedules and cancellations: how many were caused on your side, and why.

Without a baseline, you won't know whether a change helped or the weather just shifted. And keep the numbers honest: if techs are judged only on jobs per day, they'll find ways to close tickets that come back later.

Frequently asked questions