Every operation with trucks has one number somebody watches. It’s usually the diesel price, and it gets checked far more often than it needs to be. Last year that number behaved itself. Almost nothing else did.
The American Transportation Research Institute publishes an annual breakdown of what it actually costs to run a truck, built from carriers’ own books rather than estimates. The 2026 edition covers 2025, and it landed on a record. The average cost to operate a truck came in at $2.336 per mile, up 3.4 percent on the year. Take fuel out of it and the picture gets worse instead of better: $1.854 per mile, up 4.2 percent.
Fuel and driver pay both rose slower than inflation. So the money went somewhere, and it wasn’t the pump.
Where it actually went
The increases landed almost entirely in the lines nobody puts on a dashboard:
- Tolls, up 13.2 percent
- Repair and maintenance, up 8.6 percent
- Driver benefits, up 6.6 percent
- Tires, up 6.4 percent
Every one of those is a bill that shows up after the truck has already run. You can watch the fuel price in the morning and change how you buy. You find out about the tire and the toll and the brake job later, in an invoice, usually in a month where three of them arrive together.
The one cost you were watching is the one that stayed put.
ATRI’s sample is for-hire carriers, so if you run your own trucks you’re not in it. That doesn’t buy you much, because your parts counter, your tire vendor, your benefits broker and the toll authority all charge you the same way they charge everybody else. The input prices are shared. Only the visibility is different, and the private fleet usually has less of it, because the cost of a delivery is buried inside the cost of the goods.
It’s also worth knowing what carriers did in response. They cut truck counts by 2.4 percent, left about 10 percent of the trucks they kept sitting without a driver, and cut non-driver staffing by 7.8 percent. Even after all that, operating margins in truckload and refrigerated came in under 1 percent. There’s no cushion in the rate you’d pay to hand your deliveries to someone else, because the carrier is absorbing the same increases you are and has already run out of places to trim.
Why you keep the trucks anyway
Companies that run their own fleets are asked every year why they bother, in the National Private Truck Council’s benchmarking survey. The top answer isn’t cost. It’s customer service, at 49 percent, well ahead of controlling transportation costs.
That’s the right answer. A truck you own shows up when you said it would, and the driver knows the customer’s name and which door actually opens. That’s difficult to buy from anybody, at any rate, and it’s usually the reason your customer stays with you instead of the distributor two towns over who’s forty cents cheaper on the case.
None of that changes the arithmetic though. The fleet is a service you’re giving away inside the price of the goods, and services you give away are the ones that never get measured.
The measure most fleets are using is the wrong one
Nine out of ten fleets in that same survey track cost per mile as their headline number. For a carrier, that’s exactly right, because a carrier sells miles. You don’t. You sell delivered orders, and the miles are just what it took to get there.
Which means cost per mile can look perfectly healthy while the thing you actually sell is losing money. Two stops on the same route can be identical on a mileage report and nothing alike in real life, because one is six minutes at a loading dock and the other is forty minutes in a downtown alley where the receiver is at lunch and the elevator is out. Same miles, same fuel, very different day.
The unit that matters to a distributor is cost per stop, and after that, cost per customer. Once you have those two, questions that used to be arguments turn into arithmetic. Whether that account justifies a twice-weekly delivery. Whether the minimum order is set anywhere near the right place. Whether the route that everybody complains about is genuinely the bad one, or just the loud one.
The play to run this week
You can get a usable first version of this without a project, and without asking anyone to fill in a spreadsheet.
- Pick one route and one week. Add up what it cost you to run: driver hours including overtime, fuel, and a per-mile allowance for maintenance and tires. Use the ATRI figure if you don’t have your own, it’s close enough to start.
- Count the stops that week. Divide. That’s your average cost per stop, and it’s the first time most operations have ever seen the number.
- Now rank the stops by how long the driver was actually there. If your drivers work from paper, you’ll have to ask them, and they will know. If arrival and departure are captured on a phone, you already have it.
- Take the three slowest stops and put them next to what those customers spend with you. That comparison is the whole exercise, and it usually surprises somebody in the room.
Nothing there requires new trucks or a new system. It requires knowing when the truck arrived and when it left, which is the one piece of information a printed route sheet has never been able to give you.
That’s the part Driver Delivery handles inside Workd, and it’s new in the platform. Routes are built and sequenced in the system, the driver works the day from a phone, and arrival, departure, delivered quantities and proof all land on the record as the day happens. The cost per stop stops being a research project and starts being a report.
Either way, run the exercise. Ten minutes with one route tells you more about next year’s delivery costs than watching the diesel price ever will.
You don’t need cheaper trucks. You need to know which stops are paying for them.
Sources
- An Analysis of the Operational Costs of Trucking, 2026 Update · American Transportation Research Institute, July 2026, covering calendar year 2025
- Benchmarking Survey Report · National Private Truck Council, annual survey of private fleet operations
- Driver Delivery in Workd · routes, driver app, and delivered-quantity capture