Service dispatch and technician scheduling field guide for HVAC

Dispatch is the day, not a name on a grid

Service dispatch is matching the right technician to the right job at the right time, with the right parts and the right information, so the truck rolls and the customer gets served. That is the whole job in one sentence. Everything else is detail about how you make that match hold up across a dozen calls and three trucks while the phone keeps ringing.

Good dispatch fills the day and raises billable hours. The tech moves from one job to the next with short drives between them, shows up with the part already on the truck, and fixes it on the first trip. Bad dispatch does the opposite. It sends a maintenance tech to a compressor-down call, routes a truck across town and back, and forgets to tell the customer when to expect anyone. The work still happens. It just costs you two trips, a callback, and an afternoon of windshield time you cannot bill.

The board does not care how good your technicians are if the assignments are wrong. A strong crew run by weak dispatch produces a mediocre day. The money is in the match and the sequence, and that is decided in the office before anyone turns a wrench.

What is dispatching in field service?

Dispatching in field service is assigning jobs to technicians and sequencing them across the day, then keeping that plan current as the day changes. Three questions sit underneath every assignment: who goes, where they go, and when they get there. Answer those for every call on the board and you have dispatched the day.

The part people miss is that dispatch is not a morning event. You do not build the board at 7 a.m. and walk away. A job runs long, a customer cancels, an emergency drops in, a part is not in stock after all, and every one of those moves the board. Dispatch is the act of keeping the assignments true to reality from the first call to the last truck back at the shop.

Scheduling and dispatching get used as the same word, but they pull apart in practice. Scheduling is putting a job on a future day and a time window. Dispatching is deciding, on the day, which tech actually takes it and in what order. The same call gets scheduled once and dispatched once, sometimes by two different people, and the handoff between them is where a lot of shops drop the ball.

How do you match the right technician to the job?

Match the skill and certification the job needs to a tech who actually has them, then check that the tech is reasonably close. Skill first, geography second. A nearby tech who cannot do the work is not a save, it is a wasted truck roll and a second dispatch when the right person finally goes out.

The match runs deeper than journeyman versus apprentice. A controls problem on a rooftop unit needs someone who reads a wiring diagram and talks to a thermostat, not your strongest brazer. A chiller or a VRF system needs the tech trained on that equipment, and on commercial work that often means the manufacturer's training, not just general experience. Refrigerant work needs EPA Section 608 certification, and the type matters: Type I for small appliances, Type II for high-pressure systems, Type III for low-pressure, Universal for all of it. Send a tech without the right card and the work is not legal, never mind whether they can do it.

Get the match right and you get a first-time fix. Get it wrong and you get a callback, which costs you a second truck roll, a second parts pull, and a customer who now wonders whether you know what you are doing. The skill match is the cheapest callback prevention there is, and it happens for free in the office.

How do you reduce windshield time?

Cluster the work by area so each tech runs a tight loop instead of crossing town between calls. Windshield time, the hours a tech spends driving between jobs, is time you pay for and cannot bill. The average service tech loses a large share of the day to it, often 30 to 40 percent of the shift behind the wheel, and every minute of that is margin walking out the door.

The fix starts at scheduling. When you book a call, look at where the tech is already going to be that day and that area, and try to slot the new job into a route, not a random open slot across the map. A job booked into the wrong corner of the territory drags a truck out of its loop and back, and the drive eats the whole appointment's profit before the tech touches a tool.

Drive time is also where the schedule quietly fails. A dispatcher who sequences by appointment time alone, ignoring the map, builds a day where the tech zigzags across the territory chasing time windows. Sequence by geography inside each window instead. Group the morning calls on the north side, the afternoon calls on the south, and let the time windows flex within reason. The customer cares about a two-hour window. The tech cares about not driving past the next job to reach a farther one first.

The dispatch board

The dispatch board is the grid of technicians down one axis and time across the other, with each job sitting in the slot where a tech is doing it. It is how a dispatcher sees the whole day at a glance: who is free at 2 p.m., who is buried, which truck is near the new emergency, and where the holes are that you can still fill. If you cannot see the day, you cannot run it.

A whiteboard does this for one or two trucks. Past that it falls apart, because the board has to change constantly and a whiteboard does not travel to the tech in the field or update itself when a job runs long. A digital board lets you drag a job from one tech to another, slide it later when the morning call runs over, and watch the conflicts light up before they become a double-booking. FieldOS gives you that techs-by-time board with drag-and-drop, so the dispatcher moves work around the day and every tech's phone updates with the change at the same moment.

The board is only as good as what is on it. Realistic durations, accurate skills, and a live status from the field are what make it a planning tool instead of a wish. A board full of one-hour guesses is a board that lies to you all afternoon.

Realistic job durations and not overbooking

Stop booking everything as a one-hour job. A maintenance visit, a no-cool diagnostic, a compressor change, and a full system startup are not the same length, and pretending they are is how the board falls two hours behind by lunch. Build a real duration for each job type from your own history, and book the slot to match the work, not to a round number that makes the grid look tidy.

Underestimating durations is the quiet killer. The board says the tech finishes at 11 and the next customer is told to expect someone at 11:30, but the job was always a two-hour job, so now every appointment behind it is late and the tech is rushing work to catch up. Rushed work is callback work. The schedule was never real, so the day was never going to hold.

Leave slack. A board packed to 100 percent has no room for the diagnostic that turns into a repair, the emergency that has to go in, or the drive that ran long. Many shops aim to keep planned billable work around 70 to 80 percent of available hours and leave the rest as buffer, because a fully booked day is a day where the first surprise breaks every appointment after it. The slack is not idle time. It is the room that lets the day absorb reality.

How do you handle emergency and same-day calls?

Reserve capacity for emergencies before they happen, because they will happen and they are where the cash and the loyalty are. A no-cool call in July or a no-heat call in January is the customer at their most motivated to pay and most likely to remember who showed up. Miss it because the board was full and you handed that customer, and their future maintenance agreement, to the competitor who had a slot.

The practical move is to hold part of each day open for same-day work instead of booking every hour solid. Some shops designate a flex tech or a flex slot, a truck whose day is built to be interrupted, so an emergency does not have to blow up four other appointments to get covered. The rest of the board stays intact and only the flex capacity absorbs the hit.

When the emergency is bigger than your reserve, you triage. A commercial customer under contract with a response-time clock comes before a routine residential call that can move to tomorrow without much pain. Call the customer you are going to bump before they call you, give them a real new window, and most will take it fine. The no-show happens when you go quiet, not when you reschedule honestly.

Tracking technician skills and certifications

You cannot dispatch by skill if you do not know who has which skill. Keep a real record of each tech's certifications and competencies: EPA 608 type, any state or local trade license, manufacturer training on the specific equipment lines you service, and the practical strengths and gaps that do not show up on a card. Some techs are strong on controls and weak on heavy mechanical, and the board should know that even though no certificate says it.

Certifications also expire and licenses lapse, and an expired card is a compliance problem the day an inspector or a customer asks. Track the expiration dates, not just the fact that the tech once passed. The cost of a lapsed EPA card is not theoretical when it is the reason a refrigerant job has to stop.

When the skills live in the same system as the board, the match gets easier and harder to get wrong. FieldOS lets you tag each tech with their certs and competencies so that when you assign a refrigerant job or a controls call, the right people are obvious and the wrong assignment stands out. The goal is that skill match is a glance, not a memory test the dispatcher fails at 4 p.m. on a busy Friday.

Staging parts so the truck rolls ready

Send the tech with the parts the job needs. The second trip, the one where the tech diagnoses it, drives back for a part, and returns, is pure loss: double the drive, half the billable, and a customer who waited an extra day. On a known job, where the call says the capacitor is blown or the customer reported the model and the symptom, the part should be on the truck before the tech leaves the shop.

Truck stock is the first line. A well-set truck carries the common failure parts for the equipment in your territory: capacitors, contactors, common motors, igniters, fuses, the usual filters and belts. What it cannot carry is everything, so the dispatch handoff has to flag the job-specific part the truck will not have, and someone has to confirm it is pulled and loaded. A part that is in stock at the shop but not on the truck did not help anyone.

Where the call gives you the model and the symptom, dispatch can stage the likely part before the tech arrives. That is the difference between a one-trip fix and a two-day repair, and it is decided in the office by whoever reads the call notes and stocks the truck, not by the tech standing at the unit wishing they had brought the right motor.

Customer communication and the ETA text

Tell the customer when the tech is coming, and tell them again when the tech is on the way. A confirmed appointment with an on-the-way text cuts no-shows and cuts the calls into your office asking where the tech is. The customer who knows a tech is fifteen minutes out is home with the gate open. The customer who heard nothing since booking is at the store, and your tech is standing on an empty porch burning a slot.

The on-the-way message is the highest-value one. It turns a two-hour window into a real arrival, and it lets the customer do the small things that make the visit go: pen up the dog, move the car, clear the path to the equipment. It also sets the tone before the tech arrives, which matters more on a first-time customer than most owners admit.

Automate it so it does not depend on a busy dispatcher remembering. FieldOS can fire the on-the-way notification when the tech marks the job en route, with the name and an arrival window, so the customer hears from you without anyone in the office lifting a finger. Fewer no-shows, fewer where-is-my-tech calls, and a customer who feels handled instead of forgotten.

Re-dispatching when the day changes

The board you built at 7 a.m. is wrong by 10. A job ran long, a part was not there, a customer canceled, an emergency dropped in. Re-dispatching is the work of keeping the board true to what is actually happening, and it is most of what a dispatcher does after the morning.

The trigger to watch is the job that runs over. The moment a tech is going to blow past the next appointment, you have a decision: push the next job later, move it to a tech who is free, or call the customer and reschedule. The worst choice is to do nothing and let the tech show up late to every remaining call, because that turns one long job into five unhappy customers. Make the move while there is still time to make it cleanly.

A live board is what makes this possible. When the field updates status in real time, on the way, on site, finished, the dispatcher sees the overrun as it forms instead of finding out when the next customer calls angry. A stale board that nobody has touched since the morning is how you double-book a tech, because two people both think a slot is open that filled an hour ago.

GPS and knowing where the trucks are

When an emergency comes in, the question is who can get there fastest, and you cannot answer it from a guess about where everyone is. GPS on the trucks tells you who is actually nearest and actually free, so you dispatch the closest qualified tech instead of the one you assume is close. The assumption is usually wrong, because the morning never went exactly as planned.

Live location also catches the day drifting. A tech who has been parked at one address for three hours is either on a job that blew up or on a long lunch, and either way the board needs to know. You are not babysitting. You are seeing the day clearly enough to make the next call right.

FieldOS shows tech location alongside the board, so dispatching the nearest available person to a same-day call is a look at the map, not a round of phone calls asking where everyone is. Pair that with the skill tags and you are picking the closest tech who can actually do the job, which is the only version of nearest that helps.

Scheduling the non-urgent work to fill slow days

Not every job is an emergency, and the routine work is what keeps the trucks full on the slow days. Maintenance visits, planned installs, and any non-urgent repair are the flexible inventory you use to fill the gaps that demand calls do not. A shop that only books reactive work lives feast to famine, slammed one week and idle the next.

Use the backlog deliberately. The slow stretch in spring and fall, between the cooling and heating seasons, is when you schedule out the maintenance agreements and the deferred installs so the techs stay billable when the emergency phone goes quiet. That is also where a preventive maintenance program earns its keep: a book of scheduled PM visits is a buffer of work you can place into the slow days on purpose. See the HVAC preventive maintenance program guide for how to build and route that book.

Estimated and approved installs are the other lever. A signed install can move within a range of days, so it slots into the open capacity instead of forcing an overtime crunch later. Tie the estimate to the schedule so approved work flows onto the board as fill, not as a surprise. The estimating and bidding guide covers pricing those jobs so the work you schedule actually makes money.

What is first-time fix rate?

First-time fix rate is the share of service calls resolved completely on the first visit, with no return trip needed. It is the single metric that tells you whether your dispatch is actually working, because a first-time fix is the product of the three things dispatch controls: the right tech, the right parts, and the right information all arriving together.

The math is brutal in the other direction. Every call that does not fix on the first trip becomes two truck rolls, two drives, often a second parts pull, and a customer whose patience you spent. A shop sitting at a low first-time fix rate is paying for a hidden second fleet that exists only to clean up the first fleet's incomplete calls. Many operations target a first-time fix rate around 80 percent, with the best-run shops pushing past 90, but use your own baseline as the number to beat.

Dispatch moves this metric more than the techs do. Send the controls call to the controls tech, stage the part the call notes pointed to, and put the equipment history in the tech's hands before they arrive, and the first-time fix happens almost by default. Skip any one of those and you have built the callback before the tech left the shop.

The metrics that tell you dispatch is working

Dispatch quality is measurable, and a few numbers tell you most of what you need. Track them over weeks, not days, because a single day is noise. The point is the trend and where it breaks.

Billable hours, or technician utilization, is the headline: the share of paid hours that lands on a customer invoice. Industry averages often run in the 55 to 65 percent range, with strong shops up near 75 to 82, so know your number and watch which way it moves. Jobs per tech per day tells you throughput, but read it next to first-time fix, because cramming in more calls that turn into callbacks is not progress. Windshield time as a percentage of the day is the routing scorecard; when it climbs, your clustering slipped. And first-time fix rate, covered above, is the quality check on the whole match.

Read them together, not alone. High jobs-per-day with a falling first-time fix rate means you are booking too tight and rushing the work. High utilization with high windshield time means the techs are busy but half of busy is driving. The numbers only mean something in combination, and the combination is what tells you which lever to pull next.

What a good dispatcher actually does

A good dispatcher carries three maps in their head: the techs, the customers, and the territory. They know which tech is strong on controls and which one talks a nervous homeowner down, which customer needs the senior person and which site has a gate code that takes ten minutes to sort out, and how long it really takes to get from the north side to the south side at 4 p.m. None of that is on the grid, and all of it decides whether the assignment holds.

The job is judgment under pressure, not data entry. When the emergency hits and two trucks are equally close, the dispatcher who knows the techs sends the one who will actually fix it, not just the one the map favors. When a customer calls upset, the dispatcher who knows the account knows whether to bump the schedule or hold the line. That knowledge is why a strong dispatcher is worth more than the software and why the software exists to support them, not replace them.

The trap is treating the role as a junior seat. A weak dispatcher quietly costs you more than a weak tech, because every bad assignment multiplies across the day. Put a sharp, experienced person in the chair and give them a board that shows the truth, and the same trucks bill more hours doing the same work.

Scheduling software versus the whiteboard

The whiteboard works until it does not, and it stops working the day you cannot see the whole picture or push a change to the field without a phone call. A digital scheduling system pulls the board, the routing, the skills, and the customer notifications into one place that the office and the trucks both see at once. That last part is the difference: the whiteboard never leaves the wall, so the tech in the field is always working from yesterday's plan.

The honest comparison is not software versus no software. It is whether the system does the jobs a dispatcher actually needs: a techs-by-time board you can rearrange, skill tags so the match is visible, location so you dispatch the nearest, parts and history attached to the job, and automatic customer notifications. A tool that does only scheduling and makes you run routing and notifications somewhere else has just moved the problem.

FieldOS is built to hold those pieces together: the dispatch board, the technician skills and location, the job history and parts, and the customer ETA messaging in one workspace, so the dispatcher is not stitching three apps together while the phone rings. The win is not the software for its own sake. It is that the same crew and the same calls turn into more billable hours and fewer callbacks because nothing falls through the gap between tools.

Commercial contract dispatch and SLA response

Contract customers usually buy a response time, and that clock changes how you dispatch them. A commercial service agreement often promises a four-hour response on a critical call, sometimes faster on a priority site, and the contract spells out what counts as critical. When that call comes in, it jumps the queue ahead of routine residential work, because you signed up for that priority and the penalty for missing it is real, in money and in the renewal.

Dispatching contract work means knowing the agreement before the call comes, not reading it while the clock runs. Flag the contract accounts and their response terms on the board so the dispatcher sees the obligation the moment the call lands. A priority customer treated like a walk-in is a contract breach waiting to surface at renewal time, and the customer remembers the slow response long after they forget the discount they negotiated.

The exact response times, the priority tiers, and what triggers them vary by contract, so the agreement controls, not a rule of thumb. What stays constant is that contract response is a commitment you priced into the deal, and the dispatch board is where that commitment either gets honored or quietly missed.

What to document

Dispatch decisions are easier to improve when you can see what they were. Capture the factors that drove each assignment and the outcomes that resulted, so the patterns show up over weeks instead of staying buried in a busy memory. The record is also what trains the next dispatcher faster than any ride-along.

Log enough to reconstruct the day and to spot the trends: which tech took which job and why, the parts staged, the promised and actual arrival times, whether the job fixed on the first trip, and the drive time between stops. When callbacks cluster on one tech or one job type, the record tells you whether it is a skill gap, a parts problem, or a duration that was always too short on the board.

Common mistakes

• Sending the wrong skill to a job, so it needs a second tech and a callback.
• Failing to stage the job-specific part, so the tech makes a second trip for it.
• Overbooking the board to 100 percent with no slack for emergencies or jobs that run long.
• Sequencing by appointment time alone and ignoring the map, so trucks zigzag the territory.
• Skipping the on-the-way notification, so the tech arrives to a no-show and burns the slot.
• Letting the board go stale after the morning, so two jobs double-book one tech.
• Booking every job as one hour instead of a real duration built from your own history.
• Treating a contract account like a walk-in and missing the response-time commitment.

Field checklist

Standards and references

Dispatch is an operations discipline, not a code-governed one, so the references here are the practices the trade has settled on and the rules that touch the work, not a standard that dictates how you build a board. The framework worth knowing is field service management practice, which treats dispatching, routing, and capacity planning as the levers that turn the same crew into more billable hours.

One hard rule does cross into dispatch: refrigerant handling requires EPA Section 608 certification under the Clean Air Act, by type, and dispatching a tech without the right card to a refrigerant job is a compliance problem, not just a skill mismatch. State and local trade licensing applies the same way where your jurisdiction requires it. Beyond that, your service agreements are the binding documents, because the response times and priority tiers you committed to in a commercial contract govern how those calls get dispatched.

Treat the numbers in this guide as references to hold against your own operation, not as targets handed down from above. First-time fix rate, utilization, and windshield time vary with your job mix, your territory, and your contracts, so the benchmark that matters is your own baseline and the direction it moves. Stress the three that hold across every shop: match the right tech to the right job, minimize windshield time, and reserve capacity for emergencies.

Units and terms

Dispatch has its own vocabulary, and the same idea shows up under different names across software, contracts, and the shop floor. The terms below are the ones that carry weight on the board.

Dispatch and scheduling get used loosely, but scheduling places a job on a future day while dispatching decides, on the day, who takes it and in what order. Windshield time is the non-billable drive between jobs. Utilization, or billable hours, is the share of paid time that lands on an invoice. First-time fix rate is the share of calls resolved on one visit. An SLA is the response-time commitment in a service agreement. EPA 608 is the federal refrigerant-handling certification, carried by type.

Dispatching vs scheduling

Scheduling places a job on a future day and window; dispatching assigns it to a tech and sequences it on the day

Windshield time

The non-billable hours a technician spends driving between jobs

Utilization / billable hours

The share of a technician's paid time that lands on a customer invoice

First-time fix rate (FTFR)

The share of service calls resolved completely on the first visit, with no return trip

SLA / response time

The contracted time within which you must respond to a call, often tiered by priority

EPA 608

Federal certification required to handle refrigerant, carried by type (I, II, III, Universal)

Flex tech / flex capacity

A truck or slot held open to absorb same-day and emergency work without breaking the booked board

Related tools

Calculators and readiness checks for this work

Compare your options

FAQ

People also ask

• What is HVAC preventive maintenance?
• How do you estimate an HVAC job?
• What should be stocked on a service truck?
• How do you find good technicians in a labor shortage?
• What is an HVAC maintenance agreement?