Dispatch KPIs: drive time, utilization and on-time rate

Why dispatch needs its own numbers

Dispatch decides how a field team spends its day. Two companies with the same techs, trucks and call volume can have very different results because one board is planned well and the other is a scramble. Revenue and close rates measure the outcome; dispatch KPIs measure the engine that produces it.

Dispatching is a specialist role. U.S. building equipment contractors, the HVAC, plumbing and electrical firms, employed about 14,490 dispatchers in May 2025^[2], and the median dispatcher outside police, fire and ambulance earned $50,340 a year^[3]. In most smaller shops there is no dedicated dispatcher at all: in 2023, 58.3% of plumbing and HVAC contractor establishments had fewer than five employees and 76.8% had fewer than ten^[1].

That is why the KPI set must be short and easy to pull. Six well-defined numbers reviewed every week beat twenty numbers nobody trusts. Start with whatever your field service software already records reliably, fix the data habits that break the rest, and add a number only when someone has agreed to act on it.

The core dispatch KPIs

Add two supporting numbers if you have the data: same-day fulfillment rate (same-day requests completed today / eligible same-day requests) and cancellation rate (jobs cancelled or rescheduled by you / jobs booked). Both show whether the board is realistic.

Drive time per job

Drive time is the cost dispatch controls most directly. Measure it as minutes from "en route" to "arrived," per completed job, by tech and by day. Miles matter for fuel, but minutes limit capacity.

• Track it daily and review it weekly, by tech and by zone.
• Separate the first drive of the day from drives between jobs; the first drive depends on where techs start.
• Watch for outliers: one cross-town job can double a tech's drive time for the day.
• Expect it to vary by season and by trade; compare a week with the same week last year, not with another company.

Drive time is also a safety number. Transportation incidents were the most frequent type of fatal work event in 2024, accounting for 38.2% of all occupational fatalities^[4]. Fewer and calmer miles are good for people as well as margins.

Utilization

Utilization shows how much of a tech's paid day is spent on jobs. Use on-site job hours divided by paid hours. Some shops use billable hours instead of on-site hours; pick one and stick with it.

Low utilization usually has a few causes: too much drive time, late first arrivals, long gaps between jobs, parts runs and end-of-day paperwork. High utilization sounds good, but very high numbers sustained for weeks often mean no room for emergencies and a rising number of late arrivals. Look at utilization and on-time rate together.

On-time arrival rate

Define on-time as arriving within the window the customer was promised. Not within 15 minutes of an internal target; within the customer's window. Pull it from arrival timestamps against the booked window.

• Report it for the first job of the day separately; late first jobs point to start-of-day problems.
• Report late arrivals with and without an advance notice; a late arrival with a heads-up text is a lesser failure than one with none.
• Break it down by job type: if diagnostics run late, the time standard is probably too short.

Jobs per tech and schedule fill rate

Jobs per tech per day is the number owners watch most, and the easiest to misread. It depends heavily on job mix: a day of tune-ups produces more jobs than a day of diagnostics, and an install day produces one. Report it by job type, or alongside the share of install and diagnostic work, so a fall in the number is not mistaken for a fall in effort.

Schedule fill rate looks forward instead of back. It compares hours already booked for the next three days with the available hours in your capacity plan. A low fill rate on Monday morning tells the office to run call-down lists, offer flexible maintenance customers earlier slots or push marketing in quiet zones. A fill rate near the top of the range tells CSRs to protect emergency capacity and offer specific next-available windows rather than squeezing jobs in. It is the one dispatch KPI that lets you act before the week goes wrong.

Who sees which numbers

Different people need different views. Share numbers in a way that helps each person do their job, and avoid turning the board into a leaderboard that punishes techs for problems dispatch created.

• Owner or general manager: the six core KPIs weekly, plus trends by month and season.
• Dispatch lead and CSRs: schedule fill rate daily, on-time rate and drive time by zone weekly.
• Service manager: first-time fix rate, callbacks and utilization by tech, for coaching.
• Techs: their own numbers, explained in a one-to-one, with the team averages for context.

Keep the dashboard to one screen: the six numbers, this week against the four-week average, with a short note on the change agreed in the last review. If a number is not discussed in the review for a month, remove it.

Setting targets

Industry averages for these KPIs are hard to find and rarely use the same definitions, so do not borrow a number from a vendor brochure. Pull eight to twelve weeks of your own data, set the baseline and aim for steady improvement. Set targets by season, because a peak cooling week and a quiet spring week are different businesses.

A 30-minute weekly KPI review

1. 01

   Pull the numbers

   The six core KPIs for last week, by tech, with the previous four weeks for comparison.

2. 02

   Spot the biggest change

   Which number moved most, in which direction, for which tech or zone?

3. 03

   Find the cause

   Look at the jobs behind it: a time standard that was wrong, a zone with too many cross-town calls, a tech without the right parts.

4. 04

   Agree one change

   Pick one specific change for the week, with an owner.

5. 05

   Check last week's change

   Did the change you made last week move the number you expected? Keep it or drop it.

Example: one week on the board

Example (fictional): a six-tech company's weekly review showed the figures below. Drive time had risen, and on-time arrivals had dropped for two techs in the south zone.

The jobs behind the numbers showed that CSRs had booked south-zone repairs on days when those techs were in the north for maintenance. The change was a booking rule: south-zone repairs go to the south tech or wait for the south zone day unless urgent. The next week, drive time dropped back toward the average and on-time arrivals recovered.

Common KPI mistakes

• Using averages only; look at the worst days and worst techs too.
• Changing definitions mid-year, which breaks every trend line.
• Rewarding a single KPI, which invites gaming (for example, jobs per day at the expense of first-time fix).
• Measuring techs on numbers dispatch controls, such as drive time on a badly planned board.
• Building a dashboard without a routine to review it.

For the inputs that drive these numbers, see capacity planning, GPS fleet tracking and the service call checklist.