Key takeaways
- Measure a handful of dispatch KPIs well rather than twenty badly: call-to-dispatch time, on-time arrival, drive time share, first-time completion and recalls.
- Define every KPI with a formula and a data source before you set a target, then baseline for four weeks.
- Separate emergency and routine work in every metric; mixing them hides problems in both.
- Dispatch is a skilled role with real cost: the median dispatcher wage was $50,340 in 2025[1], so give the board tools and numbers that make that time count.
- Review daily in a five-minute huddle and weekly with the trend, and change one thing at a time.
Why the board needs its own numbers
Most plumbing shops track revenue, average ticket and close rate. Fewer measure the dispatch board itself, even though it decides how many jobs a tech can reach, how long customers wait and how much paid time is spent driving. Many plumbers travel to a different worksite every day, and plumbers are often on call for emergencies, so evening and weekend work is common[2]. That makes the board a moving puzzle, and puzzles improve only when you can see the pieces.
Dispatch is also a paid skill. O*NET reports a median dispatcher wage of $50,340 a year, or $24.20 an hour, in 2025, with the top 10% above $78,140[1]. Good dispatchers earn their pay by protecting tech time; KPIs show where they are doing it and where the system gets in their way.
The core dispatch KPIs
Start with these. Each one has a clear formula and comes from data most scheduling systems already record.
| KPI | Formula | Why it matters | Watch out for |
|---|---|---|---|
| Call-to-dispatch time | Minutes from call answered to tech assigned, split by emergency and routine | Fast assignment wins emergency work and calms anxious callers | Clock starting at booking instead of at the call |
| On-time arrival rate | Jobs where the tech arrived inside the promised window divided by all jobs | The promise customers judge you on | Windows widened to make the number look good |
| Drive time share | Paid drive hours divided by total paid tech hours | Driving is paid time with no billable work | Ignoring the first and last drive of the day |
| Jobs per tech per day | Completed jobs divided by tech days worked, by job type | Shows capacity and how routing affects it | Comparing install days with service days |
| First-time completion | Jobs finished on the first visit divided by all jobs | Return trips for parts cost a whole extra visit | Planned multi-visit jobs counted as failures |
| Recall rate | Jobs needing an unplanned return for the same problem within a set period | Quality and diagnosis problems show up here | Recalls booked as new jobs |
| Same-day capacity used | Emergency slots used divided by emergency slots reserved | Tells you whether you reserve too much or too little | No reserved capacity at all, so nothing to measure |
Add two operational counts beside them: cancellations and reschedules per week, and the share of those gaps you refill. The routine for refilling is covered in filling same-day schedule gaps.
Call-to-dispatch time
This is the most direct measure of how quickly the shop responds. For emergencies, the clock should start when the call is answered and stop when a named tech has accepted the job. For routine work, measure from call to a confirmed appointment.
- Record the reason for slow assignments: no tech available, waiting on a callback from the on-call tech, address outside the service area, or the office was busy.
- Track after-hours separately. If the on-call tech takes long to call back, the fix is the on-call process, not the daytime dispatcher.
- Watch how often urgent callers are told “we’ll call you back.” Each of those is an open loop that can end with the customer calling a competitor.
After hours, the number is only measurable if the call is answered and logged. An SI Employee such as Ava can take the details of an urgent call and reach the on-call technician in the order you approve, which gives the night shift the same timestamps as the day.
On-time arrival and the promise window
On-time arrival only means something next to the window you promised. A shop that promises an eight-hour window and arrives on time almost always is not necessarily serving customers better than one that promises two hours and hits most of them.
Record the window, not just the appointment
Store the start and end of the promised window on every job so arrival can be compared with it automatically.
Use the arrival timestamp
Use the moment the tech marks arrived on the job, not when the office notices. Make arriving a one-tap step on the tech’s phone.
Split by window type
Report first-call-of-the-day jobs separately. They should be close to perfect; later windows depend on earlier jobs running to plan.
Count the heads-up
Track how often customers got an on-the-way message before arrival. Most people do not answer calls from unknown numbers[3], so a text from your saved business name does the work a call cannot.
Drive time and routing
Drive time is paid time. Average hourly earnings at plumbing, heating and AC contractors were $42.38 in September 2026[4], so a tech who drives an extra 40 minutes a day costs the shop meaningful money every week before any fuel or vehicle wear.
- Measure drive time share for each tech and each day of the week. Big differences usually point to routing habits or zone design.
- Cluster routine jobs by area where the customer’s window allows it. Save cross-town moves for emergencies.
- Start the day close to home for techs who take vans home, and plan the last job near their route back.
- Check whether supply-house runs are counted as drive time. If they are frequent, the fix is van stock, not routing.
Example (fictional): a five-tech shop logs 200 paid tech hours in a week, 46 of them driving. Drive time share is 23%. After zoning routine work into north and south halves for a month, the shop logs 40 drive hours out of 200, or 20%, and fits in several more jobs without adding hours.
First-time completion and recalls
These two belong to dispatch as much as to the field. Many return trips start at booking: the wrong tech for the job, no photos of the water heater label, no note that the faucet is a specific brand. Good intake questions and clear job notes raise first-time completion before the tech leaves the shop.
- Tag every return visit with a reason: parts not on the van, wrong skill, customer not home, scope larger than booked, or genuine recall.
- Review the top reason each week and fix the step that caused it, often a missing intake question or a gap in van stock.
- Count recalls by tech and job type privately and use them for coaching, not public ranking.
The standard for notes is set out in job notes that help the next visit.
Weekends, nights and coverage
Coverage KPIs matter because customers’ schedules and your techs’ schedules do not line up. On an average weekday in 2025, 80.5% of employed people worked, compared with 30.3% on weekends and holidays[5]. Many homeowners are easiest to reach and visit outside weekday hours, exactly when the board is thinnest.
Track calls received, jobs booked and jobs run by day of week and time band. If weekend demand is steady and the board turns most of it away, that is a staffing or on-call question with numbers behind it rather than a hunch.
Setting targets without gaming them
Any KPI that is tied to praise or pay will be optimized, sometimes in ways that hurt customers. Wider windows improve on-time arrival; booking emergencies as routine work improves response time; closing jobs before the notes are written improves jobs per day. Protect the numbers with a few rules.
- Pair each KPI with a guardrail: on-time arrival with average window length, jobs per tech with recall rate, call-to-dispatch time with the share of calls classified as urgent.
- Set targets from your own baseline, not from a number heard at a conference. Improvement against last quarter is the honest measure.
- Change definitions only in writing, and restate past months when you do, so the trend stays comparable.
- Reward the team for the trend rather than individuals for single weeks.
Who owns which number
KPIs without owners become wallpaper. Give each one a person who explains it at the weekly review and proposes the next change.
- Dispatcher or dispatch lead: call-to-dispatch time, same-day capacity used and the cancellation refill rate.
- Service manager: first-time completion and recalls, because both depend on diagnosis, van stock and training.
- Operations or office manager: on-time arrival, drive time share and the review routine itself.
- Owner: the small set of numbers that link dispatch to revenue, such as jobs per tech per day and booked hours against available hours.
A review routine that sticks
- Daily huddle, five minutes: yesterday’s on-time rate, any emergency that waited too long, today’s reserved capacity.
- Weekly review, thirty minutes: the trend for each core KPI, the top reason for return trips, one change to test next week.
- Monthly: compare against the baseline, retire any KPI nobody used, and confirm definitions have not drifted.
Keep the dashboard small enough to read on a phone. If a number has not changed a decision in two months, drop it. Moving from a whiteboard makes most of these automatic; see moving to digital scheduling.
Frequently asked questions
What is a good on-time arrival rate for a plumbing company?
There is no single benchmark because it depends on the window you promise. Baseline your own rate for a month with the window recorded on every job, then set a target that improves it without widening the windows.
Should emergency and routine jobs be measured together?
No. Emergencies follow different rules for response and arrival, so report them separately. Mixing them makes both numbers harder to act on.
How many dispatch KPIs should a small shop track?
Five to seven is plenty: call-to-dispatch time, on-time arrival, drive time share, jobs per tech per day, first-time completion, recalls and same-day capacity used.
Where does the data come from?
Mostly from the scheduling system: call or booking timestamps, promised windows, tech status changes and job outcomes. The key is making the tech's arrived and completed taps quick and consistent.
References
O*NET OnLine (U.S. Department of Labor), BLS 2025 wage data (2025). National Wages 43-5032.00 - Dispatchers, Except Police, Fire, and Ambulance (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2025). Occupational Outlook Handbook: Plumbers, Pipefitters, and Steamfitters (opens in a new tab) ↑
Pew Research Center (2020). Most Americans don't answer cellphone calls from unknown numbers (opens in a new tab) ↑
U.S. Bureau of Labor Statistics, Current Employment Statistics (2026). Average hourly earnings of all employees, plumbing, heating, and air-conditioning contractors, not seasonally adjusted (CEU2023822003) (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2025). American Time Use Survey, Table 4: Employed persons working and time spent working on days worked, 2025 (opens in a new tab) ↑