Key takeaways
- Good traffic control KPIs cover four areas: service delivery, safety and compliance, commercial performance and assets.
- Every KPI needs a written formula, a data source, an owner and a review rhythm, or it becomes an argument rather than a measure.
- Leading safety indicators such as inspections completed and near-misses reported tell you more, sooner, than injury rates alone.
- The billable-hours ratio and docket-to-invoice days connect field discipline directly to profit and cash.
- Set targets from your own baseline and trends; industry benchmarks for these numbers are rarely published and rarely comparable.
The most useful traffic control KPIs measure four things: whether you deliver the shifts you promise, whether your sites are safe and compliant, whether each shift makes money and turns into cash, and whether your equipment earns its keep. Twelve numbers cover it: fill rate, on-time starts, no-shows, inspections, near-misses, fatigue and credential exceptions, billable hours, margin, invoice lag, debtor days and utilisation.
Most operations managers already feel these numbers. They know when the roster is thin, when a client is slow to pay or when a depot is short of signs. Turning that feel into a measured number does two things: it shows you a problem a month before it becomes a crisis, and it ends arguments about whether something is really getting worse.
How to set up KPIs that people trust
Before the list, four rules. Without them, KPIs become a monthly spreadsheet nobody believes.
- Write the formula down. "On-time start" means nothing until you define on time: crew on site, set-up complete or first vehicle held? Pick one and keep it.
- Name the source. Every number should come from a record: roster, docket, inspection form, payroll or invoice. If it comes from someone's memory, it is an opinion.
- Give each KPI an owner. The allocator owns fill rate. The accounts lead owns debtor days. The safety coordinator owns inspection completion.
- Set targets from your own baseline. Industry benchmarks for these measures are rarely published and rarely comparable between companies. Measure for a quarter, then improve on your own trend.
A final warning: any number you reward will be gamed. If supervisors are praised for low incident counts, reports may dry up. Reward reporting and fixing, not just the absence of bad news.
Service delivery KPIs
These tell you whether clients get what they booked.
1. Shift fill rate
Formula: shifts filled with a correctly qualified worker ÷ shifts requested, for the period.
Track unfilled shifts by reason: no available worker, fatigue limit, missing qualification, late booking. A falling fill rate is the earliest warning that your roster depth is not keeping up with demand. Our guide to rostering traffic controllers covers how to build that depth.
2. On-time start rate
Formula: shifts where the crew was on site and ready at the booked time ÷ shifts worked.
Use the docket or check-in timestamp, not the supervisor's recollection. Late starts are the most common client complaint, and they often trace back to depot loading, unclear site addresses or crews coming off another job.
3. No-show and late-withdrawal rate
Formula: shifts where a rostered worker did not attend or withdrew inside your notice period ÷ shifts rostered.
Break it down by worker and by day of week. A few individuals usually account for most of it. See our article on reducing no-shows and late cancellations for practical fixes.
Safety and compliance KPIs
Work on or next to a road used by traffic is high-risk construction work. Injury counts alone are too slow and too random in a small company to steer by, so lean on leading indicators. Safe Work Australia publishes guidance on measuring WHS performance that is worth reading alongside this list.
4. Inspection completion rate
Formula: required site inspections recorded on time ÷ inspections required.
In NSW practice, work-site traffic control is inspected at the start and end of every shift and at suitable intervals during it, plus weekly and after-care inspections on longer jobs. Count what your project documents require and whether each one was recorded, by whom and when. Gaps here are both a safety risk and an evidence gap if something goes wrong.
5. Near-miss and hazard reports per 1,000 hours worked
Formula: (near-miss and hazard reports ÷ hours worked) × 1,000.
Counter-intuitively, you want this number healthy, not near zero. A crew that reports vehicles entering the work zone, abuse and close calls is giving you the information to fix the site. A sudden drop often means people have stopped reporting, not that the roads became safer. Our guide to near-miss reporting in traffic control explains how to make reporting easy.
If you also track lost time injuries, the common frequency rate formula is lost time injuries × 1,000,000 ÷ hours worked. Treat it as a lagging measure and expect it to swing wildly when hours are low.
6. Fatigue exceptions
Formula: shifts that breached your company fatigue limits, or needed a documented approval to exceed them ÷ shifts worked.
Your limits should come from a risk assessment; many operators start from something like 12 hours per day, 50 hours per week and 10 hours' rest between shifts. SafeWork NSW's Code of Practice "Managing the risk of fatigue at work" commenced on 20 February 2026, so check your approach against it on the SafeWork NSW website. The free roster fatigue checker is a quick way to test a week's roster against your limits.
7. Credential exceptions at allocation
Formula: allocations blocked or corrected because a worker lacked the required card, induction or licence ÷ allocations made.
The target for credentials on site is always 100%, so the useful number is how often your process catches a problem before the shift. In NSW, an employer must not direct a worker to do traffic control work without the right SafeWork NSW card, and workers who have not done traffic control work for two years need refresher training. A rising exception count means your register is out of date.
Commercial and cash KPIs
These tell you whether the work makes money and how quickly it turns into cash.
8. Billable-hours ratio
Formula: controller hours invoiced ÷ controller hours paid.
This is the most important profit number most operators never calculate. Unbilled travel, standby, induction, written-off cancellations and credited docket disputes all push it down. Even a few points of improvement flows almost straight to profit.
9. Gross margin by client and shift type
Formula: (revenue − direct labour − direct equipment and vehicle costs) ÷ revenue, grouped by client and by time band.
Averages hide problems. A client whose jobs are always short, late-cancelled or extended without approval can be loss-making while the company average looks fine. Night and weekend work priced as "day rate plus a bit" often shows up here too.
10. Docket-to-invoice days
Formula: average days from shift end to invoice issued.
Every day here is a day added to your cash gap. The fix is almost always in the field: dockets signed on site, checked the same day. Our guide to digital timesheets and dockets shows what a clean docket includes.
11. Debtor days
Formula: (trade debtors at month end ÷ revenue for the last three months) × the number of days in those three months.
Track it overall and for your top clients. Rising debtor days with steady revenue means clients are paying you later, which shows up as a wage-week crunch soon after.
Asset KPI
12. Equipment and vehicle utilisation
Formula: days an item was on a job ÷ days it was available, by item type and depot.
Measure utes, TMAs, VMS boards, arrow boards and sign sets separately. Low utilisation on expensive plant is a capital problem; very high utilisation with frequent hire-ins may mean you should buy. Also track items unavailable for repair or inspection, because availability is part of the story. A reserved item sitting on a ute that never left the depot is not "used", so count actual deployment from dockets rather than bookings. Our guide to equipment utilisation reports explains how to set the denominator honestly before you decide to buy more.
The 12 traffic control KPIs on one page
| # | KPI | Owner | Review | Warning sign |
|---|---|---|---|---|
| 1 | Shift fill rate | Allocator | Weekly | Falling trend, more "no worker" reasons |
| 2 | On-time start rate | Operations | Weekly | Late starts clustered by depot or client |
| 3 | No-show and late withdrawal | Allocator | Weekly | Same workers repeatedly |
| 4 | Inspection completion | Safety | Weekly | Missing mid-shift or end-of-shift records |
| 5 | Near-miss reports per 1,000 hours | Safety | Monthly | Sudden drop to near zero |
| 6 | Fatigue exceptions | Operations | Weekly | Exceptions becoming routine |
| 7 | Credential exceptions | Allocator | Monthly | Rising count, expired inductions |
| 8 | Billable-hours ratio | Finance | Monthly | Falling while revenue grows |
| 9 | Gross margin by client and band | Finance | Monthly | One client or band consistently low |
| 10 | Docket-to-invoice days | Accounts | Weekly | Creeping past a few days |
| 11 | Debtor days | Accounts | Monthly | Rising with steady revenue |
| 12 | Equipment utilisation | Depot | Monthly | Low on costly plant, or constant hire-ins |
Print it, stick it on the operations room wall, and review it at the same meeting every week.
Mistakes to avoid when measuring
- Too many numbers. Twelve is plenty. Add more only when one of these is under control.
- Lagging-only safety measures. An injury rate tells you what already happened.
- Definitions that change. If "on time" changes meaning each month, the trend is meaningless.
- Data entered after the fact. KPIs built from paperwork reconstructed on Friday afternoon are fiction. Capture on site, with timestamps.
- Measuring without acting. Every review should end with one or two actions and an owner.
Next steps
Pick four KPIs, one from each area, define them in writing, and run them for a month before adding more. Most of the data already sits in your rosters, dockets and inspection forms. TrafWise captures shift times, signed dockets, inspections and fatigue checks as the work happens, which makes these numbers something you read rather than rebuild.
Frequently asked questions
What KPIs should a traffic control company track?
Start with a small set across four areas: service (shift fill rate, on-time starts, no-shows), safety and compliance (inspections, near-miss reporting, fatigue and credential exceptions), commercial (billable-hours ratio, gross margin, invoice lag, debtor days) and assets (equipment utilisation). Track them weekly or monthly from real records.
What is a good shift fill rate for traffic control?
There is no published industry benchmark that fits every operator. Measure your own baseline first, then look at the trend and the reasons for unfilled shifts. A falling fill rate usually points to roster depth, short-notice bookings or fatigue limits, and each needs a different fix.
Are lost time injury rates a good safety KPI?
They are useful but lagging: they count harm after it happens, and in a small company one injury swings the number wildly. Pair them with leading indicators such as inspections completed on time, near-misses and hazards reported, toolbox talks held and fatigue exceptions. Those show whether your controls are working before someone is hurt.
How often should operations managers review KPIs?
Operational measures such as shift fill, on-time starts, no-shows and fatigue exceptions are worth a weekly look. Commercial and asset measures such as margin, debtor days and utilisation usually suit a monthly review. Safety leading indicators should be visible weekly and reviewed formally with workers and management regularly.
Where does the data for traffic control KPIs come from?
Mostly from records you already keep: the roster, signed dockets, inspection forms, incident and near-miss reports, payroll, invoices and equipment logs. The quality of the KPI depends on those records being complete and timestamped, which is why digital dockets and rostering make measurement far easier.