Most shippers can quote what a carrier charges them per shipment. Far fewer can state their true last mile delivery cost per package — the fully loaded number that includes driver hours, fuel, vehicle wear, failed attempts, and the overhead sitting behind every route. That blind spot is expensive. According to the Capgemini Research Institute, last-mile delivery accounts for roughly 41% of total supply chain costs, and Statista data covering 2018 through 2023 shows the last mile’s share of total shipping cost rising from 41% to 53% worldwide.
Here is how to build the number properly, what it tends to look like, and which levers actually move it.
Start with cost buckets, not the invoice
A carrier invoice tells you what you were billed. It does not tell you what delivery costs. Whether you run your own vans or buy last-mile capacity, the same five buckets apply.
Labor
Usually the largest line. Include wages, payroll taxes, workers’ comp, benefits, paid time not on the road (yard time, pre-trip, debrief), and helper labor on two-person deliveries. The U.S. Bureau of Labor Statistics put the median annual wage for light truck drivers at $44,140 as of May 2024 — roughly $21 an hour before the loaded cost of employment, which typically adds 25–40% on top.
Vehicle and fuel
Split these. Fuel is variable and volatile: the U.S. Energy Information Administration projected an average of about $3.70 per gallon for regular gasoline and $4.76 per gallon for on-highway diesel in 2026. Non-fuel vehicle cost — depreciation or lease, insurance, maintenance, tires, tolls, registration — is best expressed per mile. For heavier equipment, the American Transportation Research Institute’s latest operational costs analysis found the industry-average marginal cost of running a truck hit $2.336 per mile in 2025, up 3.4% year over year, with the non-fuel portion at $1.854. Vans run well below that, but the direction of travel is the same.
Failed attempts and rework
This is the bucket almost everyone omits. A failed delivery consumes a stop, then consumes a second one, plus customer service handling and sometimes a refund. A Loqate study of retailers put the average cost of a failed delivery at $17.78 per order. Even if your figure is lower, rework is never free.
Technology and dispatch
Routing software, telematics, scanning hardware, tracking notifications, dispatcher salaries, and exception handling.
Facility and fixed overhead
Cross-dock or warehouse handling, sortation labor, linehaul into the delivery market, rent, and management. Allocate this per package or per route — just allocate it consistently.
How to actually run the calculation
Work at the route level for one representative day, then divide.
Total route cost ÷ packages successfully delivered = cost per package
An illustrative urban route (plug in your own figures):
- Driver, 9 paid hours at $28 fully loaded: $252
- Fuel, 95 miles at 14 mpg and $3.70/gal: $25
- Non-fuel vehicle, 95 miles at $0.42: $40
- Dispatch, tech and tracking allocation: $18
- Warehouse handling at $0.55 x 90 packages: $50
- Fixed overhead allocation: $30
- Route total: $415
At 90 packages attempted, that is $4.61 per package. But if 5 of those 90 fail on the first attempt and each re-attempt plus service handling costs roughly $7.50, the route really costs $452.50 to deliver 90 packages — $5.03 each. Ignoring failures understated the number by about 9%. Run the calculation both ways and the cost of poor address data becomes impossible to argue with.
Now change one variable. Keep the same driver, same van, same eight-hour window, but load 120 packages instead of 90 by tightening the delivery zone. Route cost rises only to about $431, because most of it is fixed to the route rather than the package. Cost per package falls to roughly $3.59 — about 22% lower with no new technology and no wage cut.
What the ranges look like
Be skeptical of any single published benchmark. Public per-package figures are mostly vendor-sourced and rarely disclose what is included. What holds up across sources is the shape of the curve rather than a specific dollar amount:
- Density dominates. Dense urban and suburban routes with tight stop spacing come in dramatically cheaper per package than rural or long-tail routes, where a driver may burn 20 minutes between stops.
- Service level is a cost multiplier. Narrow appointment windows, same-day commitments, signature requirements, and two-person or threshold delivery all raise cost per package, sometimes by multiples.
- Parcel is not the floor. UPS and FedEx both announced 5.9% average general rate increases for 2026, and accessorial changes push effective increases higher for many shippers. Meanwhile Pitney Bowes reported U.S. parcel volume of about 23.1 billion in 2025, up 3.3% — demand is growing into a rate environment that keeps ratcheting.
If your cost per package looks unbelievably low, you have probably left out overhead, rework, or unpaid driver time.
The levers that actually move the number
| Lever | What it changes | Where the savings come from |
|---|---|---|
| Route density / zone consolidation | Stops per hour | Spreads route-fixed cost across more packages — the single biggest lever |
| Address validation at checkout | First-attempt success | Removes rework, refunds, and service tickets |
| Route sequencing and re-optimization | Miles and paid hours | Cuts fuel, overtime, and vehicle wear |
| Delivery window design | Route feasibility | Wider or flexible windows let the optimizer batch stops |
| Vehicle right-sizing | Fuel and capital | Stop paying for cube you never fill |
| Photo POD and access notes | Repeat failures | Fixes the same problem address permanently |
| Pooled last-mile capacity | Fixed cost exposure | Converts fleet fixed cost into variable cost per package |
Order matters. Density and first-attempt success are worth more than shaving pennies on fuel, and they are usually the cheapest to fix.
The Bottom Line
Cost per package is not a number you look up — it is a number you build, from labor, fuel, vehicle, rework, and overhead, divided by packages that actually arrived. Once you own that figure at the route level, decisions that used to be arguments become arithmetic: which zones to serve in-house, which to hand to a partner, how wide to set delivery windows, and what a same-day promise really costs you.
If you would rather compare your internal number against real market capacity, Go LTL runs last-mile delivery services, courier, and LTL freight across South Florida. Send us your volumes and zones and request a quote — we will price it against the cost per package you just calculated.
Recent Comments