Transport is the line item most businesses estimate rather than measure, and it picked a bad decade for that. The American Transportation Research Institute put the average cost of operating a truck at a record $2.336 per mile in 2025, up 3.4% year over year, with costs excluding fuel climbing faster still at 4.2% (Fleet Owner).
A transport management KPI turns that pressure into something you can act on. These eight cover the full picture: what a mile costs you, how full your vehicles are, how long loads take, and whether the carrier you’re paying is actually performing. Each comes with a formula and a 2026 benchmark.
The Bottom Line
- Cost per mile is the anchor metric. The 2025 US average hit $2.336 and is projected near $2.51 for 2026, roughly $216,000 per truck per year.
- Load utilization is usually the fastest win available, because a half-empty vehicle costs almost exactly what a full one costs.
- Driver wages ($0.818/mile) and fuel ($0.482/mile) dominate the cost structure, but repair and maintenance rose fastest in 2025 at 8.6%.
- Track four to six of these at a time rather than all eight, chosen against whichever cost line is hurting most right now.
Metrobi drivers are rated 4.97/5
Trusted by local businesses for:
- Background-checked professionals
- Specialized in business deliveries
- Same drivers for consistency
- 4.97/5 average delivery rating
What transport management KPIs cover
Transport KPIs measure the cost and reliability of moving goods: vehicles, routes, drivers, carriers, and fuel. They apply whether you own trucks, hire a carrier, or both.
The first boundary to draw is against delivery performance. Whether a customer got their order on the promised day is a delivery question, measured as on-time delivery rate and covered in depth in the on-time delivery KPI guide. Transport KPIs ask the upstream question: what did that trip cost, how full was the vehicle, and was the carrier reliable? You can run excellent on-time delivery while losing money on every route, and only transport KPIs will tell you.
The second boundary is against cost. Freight is one component of what it costs to fulfil an order, and the whole-order version of that figure is cost per order in the distribution KPI set. Use cost per mile to decide how to move a load, and cost per order to decide whether the customer receiving it is profitable. Likewise, if loads are going out with the wrong contents, the measurement you want is order accuracy from the order management KPI set, since no amount of transport optimization fixes a correctly delivered wrong box.
Transport is one of six families in a complete operational scorecard, and the supply chain metrics and KPIs framework shows how they connect. What follows is the transport column in detail.
1. Cost per mile
Formula: total transport costs / total miles driven
This is the anchor. Everything else in transport either explains this number or follows from it.
ATRI’s 2025 figure of $2.336 per mile is the marginal cost of operating a Class 8 truck, broken down as driver wages at $0.818, fuel at $0.482, truck and trailer payments at $0.404, and repair and maintenance at $0.215. The 2026 projection is roughly $2.51 per mile, or about $216,000 per truck annually. Regional spread is real: the Northeast ran $2.52 against the South-Central US at $2.23 (Fleet Owner).
If you run lighter vehicles your absolute number will be lower, but the structure holds: labor first, fuel second, vehicle cost third, maintenance fourth and rising fastest. Calculate yours monthly and include everything, meaning wages, fuel, insurance, lease or depreciation, maintenance, and tires. The common mistake is excluding the costs that don’t arrive as a monthly bill, which flatters the number by 20% or more and makes in-house delivery look cheaper than it is.
Cost per mile is also the right basis for the own-versus-outsource question, since it converts a fixed-cost commitment into the same units as a quoted rate. Whether those costs sit on your balance sheet or your monthly expenses is a separate decision with its own tax and cash consequences, and the differences between capex and opex matter before you compare a lease quote against a per-mile rate.
Regional cost spread should also feed into where you place capacity. The gap between the Northeast and the South-Central US is nearly 30 cents a mile, which is part of why distribution keeps concentrating in mid-South freight corridors. Operators running delivery routes around Nashville and comparable markets work from a materially different cost base than their Northeast equivalents, and a cost-per-mile target copied from a national average will mislead you in either direction.
2. Cost per delivery and cost per stop
Formula: total transport costs / number of deliveries completed
Cost per mile is the wrong unit for local operations. A route with 30 stops in 40 miles has a cost per mile that looks catastrophic and a cost per stop that may be excellent. If most of your work is urban multi-stop, cost per stop is the metric that reflects reality.
The useful property is that it exposes stop density. Two routes with identical mileage and wildly different stop counts have very different economics, and cost per stop is the number that makes the case for consolidating a thin route or adding a customer to a dense one.
Watch the fixed-cost split. A large share of cost per stop is driver time and vehicle time that you pay for whether or not the stop exists, which means the marginal cost of adding a stop to an existing route is far below your average cost per stop. That gap is where route profitability decisions live.
3. Load utilization
Formula: actual load (weight or volume) / maximum vehicle capacity x 100
Load utilization measures how much of what you paid for you actually used. It’s usually the fastest available win in transport, because the cost difference between driving a full vehicle and a half-empty one is small, while the revenue difference is enormous.
Measure against whichever constraint binds first. Dense goods hit a weight limit before a volume limit. Bulky light goods, including most floral and bakery freight, hit volume or cube limits long before weight. Measuring against the wrong constraint produces a flattering number and no insight.
A sustained reading below 70% usually means one of three things: delivery windows are too tight to consolidate loads, your route planning is static while demand moved, or you’re running a fixed schedule where demand would support a variable one. All three are fixable without buying anything.
4. Fuel efficiency and fuel cost per mile
Formula: total miles / total gallons consumed, and total fuel spend / total miles
Fuel averaged $0.482 per mile for heavy trucks in 2025, barely moving from 2024. That stability did not hold into 2026: diesel averaged $5.01 through late September 2026 and set a record of $6.529 in the week of September 21, a 36.9% increase on 2025 (Fleet Owner).
That volatility is the argument for tracking fuel separately rather than letting it hide inside cost per mile. When a single input swings nearly 40%, a blended number stops telling you whether your operation got worse or the market did.
Track fuel efficiency by vehicle and by driver. The spread between your best and worst driver on the same route in the same vehicle is routinely 10% to 15%, and it comes down to idling, acceleration, and speed. That’s a coaching conversation with a measurable payback, not a capital expense.
5. Average transit time
Formula: sum of (delivery timestamp – dispatch timestamp) / number of shipments
Transit time measures how long a load actually takes, which matters for two reasons beyond customer promises.
First, transit time is the input to your delivery promise. If you quote two days and your actual average is 1.9 days with a standard deviation of 0.8, you will miss regularly even though your average looks fine. Track the variability, not just the mean.
Second, inbound transit time variability is what drives safety stock. Unreliable inbound freight forces you to hold extra inventory as insurance, which shows up as carrying cost in the inventory management KPI set rather than as a transport problem. Many businesses pay for an unmeasured transport problem on their inventory line without ever connecting the two.
6. On-time pickup rate
Formula: pickups completed in the agreed window / total scheduled pickups x 100
On-time pickup is the leading indicator that on-time delivery is the lagging indicator of. A load collected late is almost always delivered late, and you find out hours earlier.
The target is 95% and up, matching the delivery standard it predicts. It’s the highest-value KPI in this list for anyone who currently only tracks delivery outcomes, because it turns a post-mortem into a warning. A missed pickup at 9am is a problem you can still solve at 10am.
This is also the metric that most reliably separates a carrier’s sales pitch from its operation. Carriers tend to report delivery performance and stay quiet about collection reliability.
7. Truck turnaround time
Formula: departure timestamp – arrival timestamp, at each site
Turnaround time measures how long a vehicle sits at a location before it can leave, and it’s the KPI most likely to reveal that your transport problem isn’t actually a transport problem.
When turnaround runs long, the cause is usually on the dock rather than on the road: congested receiving, no appointment scheduling, paperwork handled sequentially instead of in parallel, or a single loading bay serving too many vehicles. Those are warehouse and facility issues, measured as receiving efficiency and dock-to-stock time in the warehouse KPI set.
The economics are straightforward. A driver waiting is a driver being paid at roughly $0.818 per mile equivalent to produce nothing, and detention charges from carriers turn your dock congestion into a line on their invoice.
8. Carrier performance score
Formula: a weighted composite of on-time delivery, on-time pickup, damage rate, and billing accuracy per carrier
If you buy transport rather than own it, this is the KPI that lets you compare providers on something other than the rate they quoted.
Build it from four components, weighted to what hurts you most:
| Component | Formula | Typical target |
|---|---|---|
| On-time delivery | On-time / total x 100 | 95%+ |
| On-time pickup | In-window / total x 100 | 95%+ |
| Damage rate | Damaged shipments / total x 100 | Under 0.5% |
| Freight bill accuracy | Correct invoices / total x 100 | 98%+ |
Billing accuracy is the component most often skipped and the one that most often pays for the exercise. Freight invoices carry errors at rates that would be unacceptable anywhere else in your business, and almost all of them favour the carrier.
Score quarterly, share the scorecard with the carrier, and the conversation changes character. A carrier presented with their own damage rate next to a competitor’s tends to improve it. A carrier told their service “feels unreliable” does not.
Carrier scorecards also matter for specialized freight, where the cost of getting it wrong isn’t a late delivery. If any part of your load is regulated, the compliance requirements around it belong on the scorecard alongside service. Anyone moving cleaning chemicals, aerosols, certain batteries, or alcohol above a threshold is subject to the rules covered in the guide to transporting hazardous materials, and a carrier’s documentation record there deserves as much weight as their on-time rate.
Transport management KPI benchmarks at a glance
| KPI | Formula | 2026 reference point |
|---|---|---|
| Cost per mile | Total transport cost / total miles | $2.34 (2025 US avg), ~$2.51 projected |
| Cost per stop | Total transport cost / deliveries | Trend down, compare across routes |
| Load utilization | Actual load / capacity x 100 | 85%+, under 70% needs action |
| Fuel cost per mile | Fuel spend / total miles | $0.48 (2025 heavy truck) |
| Average transit time | Delivery time – dispatch time | Track variance, not just mean |
| On-time pickup rate | In-window pickups / total x 100 | 95%+ |
| Truck turnaround | Departure – arrival | Under 2 hours per site |
| Carrier score | Weighted service composite | 95%+ composite |
Industry guidance consistently recommends tracking four to six KPIs tied to your current biggest pain rather than all of them at once (usedatabrain, 2026). If freight cost is the problem, start with cost per mile, load utilization, and fuel. If reliability is the problem, start with on-time pickup, transit time variance, and carrier score.
Which transport KPI to start with
Pick by symptom.
“Delivery costs are eating the margin.” Cost per mile or cost per stop first, then load utilization. Those two together will tell you within a month whether you have a pricing problem or a density problem, and the answers are completely different.
“Deliveries keep slipping.” On-time pickup rate first, because it’s the leading indicator and nobody tracks it. Then transit time variance.
“We never know whether the carrier is any good.” Carrier performance score, including billing accuracy. Three months of data changes the negotiation.
“Drivers are always waiting around.” Truck turnaround time by site. Expect the answer to point at your own dock.
Whichever you pick, write the definition down. Cost per mile in particular has about five defensible calculations, and the one chosen is usually whichever excludes the most inconvenient costs.
Frequently asked questions
What is a transport management KPI?
A transport management KPI measures the cost or reliability of moving goods, covering vehicles, routes, drivers, and carriers. The core set is cost per mile, load utilization, fuel efficiency, average transit time, on-time pickup rate, truck turnaround time, and a carrier performance score.
How do you calculate cost per mile?
Divide total transport costs by total miles driven for the same period. Include driver wages, fuel, insurance, lease or depreciation, maintenance, and tires. The 2025 US average for a Class 8 truck was $2.336 per mile, with a 2026 projection near $2.51.
What is a good load utilization rate?
Above 85% of whichever capacity constraint binds first, whether weight or volume. Below 70% sustained usually means delivery windows prevent consolidation, route plans have gone stale, or a fixed schedule is running where demand would support a variable one.
Should I track cost per mile or cost per delivery?
Cost per mile for long-haul and line-haul work. Cost per delivery or cost per stop for local multi-stop routes, where a high cost per mile is normal and tells you nothing. Many operations usefully track both.
How many transport KPIs should a business track?
Four to six at a time, selected against your current worst cost or service problem. Tracking all eight produces a report nobody acts on. Review the set quarterly and rotate as the problem changes.
Putting it to work
Start with one month of honest cost per mile, including the costs that don’t arrive as a monthly invoice. Add load utilization against the right constraint. Those two numbers together explain most transport cost problems and cost nothing but the time to compile them.
Then add on-time pickup rate, because it’s the cheapest early warning system in logistics and almost nobody is watching it.
