Cost per delivery: building the number that prices your service

Ask a delivery operator what a single drop costs and most will quote a number lifted from a competitor's rate card, or from whatever a marketplace app happens to pay per completed order. Neither number comes from their own ledger, and neither survives a fuel price rise, a failed delivery, or a van due for renewal.

Price a service below its real cost and the business quietly funds every drop out of cash it never budgeted for. Price it without knowing the real cost at all, and there is no way to separate a good month from a lucky one.

Start with the cost nobody disputes. Diesel in the UAE has sold at AED 4.30 a litre since 31 August 2026, and petrol at AED 3.69 a litre (GlobalPetrolPrices.com, citing Emirates General Petroleum Corporation data, retrieved 2026-09-06). Run a route's distance and drop count through the shipping cost per unit calculator and fuel alone already puts a floor under the price, before a single dirham of wages, tolls or depreciation is added.

Cost per delivery is the total fuel, tolls, driver time and vehicle depreciation for a route, divided by the drops that route actually completes, not the number a competitor's rate card or a marketplace app happens to pay. What follows builds that number properly, component by component, and shows where most delivery pricing quietly loses money.

Key Takeaways

• Diesel at AED 4.30 a litre and petrol at AED 3.69 a litre (31 August 2026) are the easiest cost to price in, and the easiest to under-price once fuel moves again.
• Salik tolls now run on dynamic pricing of AED 0 to 6 per gate across Dubai's ten toll points, and a multi-drop route can cross several in a single day.
• Last-mile delivery can account for up to 53% of the total cost of moving a shipment, which is why the delivery leg needs its own costing rather than a share of a freight rate.
• A failed delivery is not a zero-cost miss — it is a second trip, and the redelivery cost belongs in the price, not in the loss column.
• Vehicle depreciation is a per-delivery cost once it is spread across a vehicle's working life and drop volume, not a lump sum written off once a year.

Fuel and tolls are the visible costs, and still underpriced

Last-mile delivery already accounts for up to 53% of the total cost of moving a shipment (Wikipedia, Last-mile (transportation), retrieved 2026-09-06), which is exactly why the delivery leg needs its own costing rather than a share of a wider freight rate.

A van doing 45 drops across a mixed urban route covers roughly 120 km a day, including backtracking between zones. At AED 4.30 a litre and a realistic 12 litres per 100 km for a loaded panel van, that route burns close to AED 62 in diesel before it crosses a single toll gate.

Tolls add up fast once dynamic pricing is in the mix. Since 31 January 2025, Salik has charged between AED 0 and AED 6 a gate depending on the time of day, across ten toll points in Dubai (Wikipedia, Salik (road toll), retrieved 2026-09-06). A route that clears four gates during the morning peak can add close to AED 24 a day — a cost that barely registers on one trip and adds up fast over a month of routes.

Divide AED 86 of fuel and tolls across 45 drops and the visible cost alone already sits around AED 1.90 a delivery, before the driver, the vehicle, or a single failed attempt enters the calculation.

Driver time is a cost per hour, not a cost per drop

A driver's cost does not stop at the wage on the payslip. All-in cost — salary, visa, insurance, and time spent waiting at a loading bay or re-attempting a locked gate — is what actually funds the route, paid whether the driver completes 30 drops or 55.

If a driver's fully loaded daily cost runs to AED 160 across an eight-hour shift, that is roughly AED 3.55 per drop on a 45-drop day, and closer to AED 5.30 per drop on a slower 30-drop day covering the same distance. Drop count, not the wage itself, is what moves this number — which is exactly why route density belongs in the pricing conversation, not just the driver's pay grade.

Vehicle depreciation is a cost per delivery once you spread it properly

A panel van bought for AED 90,000 and run for five years or 250,000 km, whichever comes first, depreciates at roughly AED 0.36 a kilometre. Over a 120 km route that is about AED 43 a day, or AED 0.96 per drop on 45 deliveries — a cost most operators fold into "overheads" instead of pricing into the drop itself.

Financing changes the arithmetic further. A van on a loan carries interest on top of depreciation, and a van kept two years past its economic life carries maintenance costs that eat the saving from not replacing it. Either way, the vehicle belongs in the same per-drop calculation as fuel and driver time, not a separate capital budget nobody revisits.

A failed delivery is a second trip, not a missed one

This is the cost most rate cards leave out entirely. A delivery that fails on the first attempt does not disappear from the books — it returns as a second trip carrying its own share of fuel, tolls and driver time, usually with no second fee to cover it.

Take a route where one drop in ten needs a second attempt. On the numbers above, a redelivery adds back close to AED 6 in fuel, tolls and driver time, effectively doubling that drop's true cost. Spread across the route, a 10% failure rate adds roughly AED 0.60 to the average cost of every delivery on it, not just the ones that fail.

This matters most on fixed-price contracts. A service priced flat per drop, agreed before the failure rate was known, absorbs every redelivery as pure margin loss. This differs from cost per unit shipped, which allocates freight, duty and warehousing into a landed cost per unit for a business moving product. A delivery operator has no unit changing hands, only a route, a vehicle and a driver's day — the failure cost sits entirely on that side of the ledger.

Putting the build together

<!-- [CHART: Cost-per-delivery build-up by component — fuel, tolls, driver time, depreciation, failed-delivery loading] -— ``` Cost per delivery build (45-drop route, illustrative): +---------------------------------+-----------------+ | Component | Cost per drop | +---------------------------------+-----------------+ | Fuel | AED 1.38 | | Tolls | AED 0.53 | | Driver time | AED 3.55 | | Vehicle depreciation | AED 0.96 | | Failed-delivery loading (10%) | AED 0.60 | | Total cost per delivery | AED 7.02 | +---------------------------------+-----------------+ ``` That AED 7.02 is the floor, not the price. Add the margin the business needs to fund growth, bad debt and the next vehicle, and only then does a per-drop or per-contract rate make sense. Deciding where that margin sits, and whether the fix is a higher rate or tighter route density, is the decision this number exists to support — which is why it belongs in the [service pricing review](/income/accelerator/scale/growth-strategy/services), not in a spreadsheet nobody revisits. ## Frequently asked questions ### How do I calculate a true cost per delivery for my own fleet? Add fuel, tolls, fully loaded driver cost and vehicle depreciation for a representative route, then divide by that route's actual drop count. Load in a failure-rate adjustment from your own redelivery data, not an industry average, since failure rates vary sharply by area and delivery window. ### Should head office overheads sit inside the cost-per-delivery figure? Not in the operational figure used to price a route day to day — that number should stay clean enough to compare route to route. Add a separate overhead allocation when setting the contract price, so operational efficiency and overhead recovery are not confused with each other. ### How often should this number be recalculated? Whenever fuel prices move materially, at every vehicle renewal, and at least once a quarter regardless. A cost-per-delivery figure built on last year's fuel price and a written-off van is not a live number, and pricing off it is pricing off history. ### Does the failure rate really need its own line? Yes, because it moves the total more than most operators expect. A route running at a 5% failure rate and one running at 15% can carry identical fuel and driver costs and still have a meaningfully different true cost per drop. ## The bottom line The number that should set a delivery service's price is not the one on a competitor's rate card, and it is not the flat fee a marketplace app happens to pay. It is fuel, tolls, driver time and vehicle depreciation, divided by the drops a route actually completes, with the redelivery cost loaded back in rather than written off as bad luck. Build that number once, from your own routes, and it holds up under a fuel price change, a slow month, or a fleet renewal in a way a rate copied from someone else's business never will. Everything else — the margin, the contract terms, the decision to add another van — sits on top of it. *Figures were verified via WebFetch on 6 September 2026 against the sources cited above; WebSearch was unavailable this session (its budget was exhausted), so driver-time, depreciation and failed-delivery figures are shown as worked arithmetic on stated assumptions rather than sourced industry averages, since no verifiable UAE-specific figure could be found for either.*