Guides
Reducing the cost of getting the box to the door
Operators trying to cut delivery costs usually start with fuel and vehicle contracts, which are the two smallest and least controllable components. The larger costs are structural and mostly self-inflicted.
Cost per drop, not cost per kilometre
The useful unit is what it costs to place one box at one door. That figure includes the driver hour it consumed, not merely the fuel, and it is dominated by how many drops fit into an hour rather than by distance travelled.
Reframing it this way changes which levers look attractive. A route ten percent shorter saves ten percent of fuel; a route with three more drops in the same hour reduces cost per drop by considerably more, because the expensive input is time.
Where the time actually goes
Finding the door
The single largest recoverable loss in Gulf operations. A driver phoning for directions is stationary and interrupting the kitchen simultaneously.
Failed deliveries
Full cost, zero revenue, and often a redelivery that doubles it. Repeat failures at one address are a data problem, not bad luck.
Scattered stops
Drive time between drops rather than the drops themselves. Density is the variable, not total distance.
Late departure
A round leaving thirty minutes late meets worse traffic for its entire length, compounding a delay created in the kitchen.
Stops that should not exist
Paused customers still on the round, or terms that expired and kept generating deliveries.
The levers ranked by what they actually return
Precise delivery locations first. Capturing a pin at signup removes the largest recoverable loss and costs nothing per customer after the first collection.
Correctness of the stop list second. A round that includes nobody who paused and nobody whose term ended is shorter for free, and this is pure propagation rather than logistics.
Density third, which is a service-area decision rather than a routing one. Concentrating a distant community into specific days converts an uneconomic scatter into a viable round.
Routing optimisation last, and often not at all. At meal-plan scale an experienced driver on a familiar round is already close to optimal, and a solver that does not know the parking usually loses what it gains.
What not to cut
Insulated packaging in a hot climate is a food-safety control rather than a cost line, and reducing it to save money transfers risk to the customer. The same applies to shortening the delivery window past the point where the cold chain is comfortable.
It is also worth being honest that some cost is the product. Customers pay for delivery to their door at a predictable time; degrading that to save a small amount per drop tends to show up later as churn, which is far more expensive than the saving.
Common questions
- What is the biggest recoverable delivery cost?
- Time spent finding doors. In Gulf operations a driver searching for a building is the largest single loss, and capturing a precise location at signup removes most of it permanently.
- Is route optimisation worth buying?
- Usually last, if at all. At meal-plan scale an experienced driver on a familiar round is close to optimal already, and correctness of the stop list returns far more than shorter sequencing.
- Should delivery cost be measured per kilometre?
- Per drop is more useful. The expensive input is driver time rather than fuel, so a route with more drops per hour beats a shorter route with the same number of stops.
- Where should cost not be cut?
- Insulated packaging and the delivery window in a hot climate. Both are food-safety controls, and trimming them transfers risk to the customer while producing a saving that churn will exceed.
See how this works in practice
The home page walks through the same mechanics against a real admin loaded with a demonstration kitchen: production, packing, routes and the customer side.
Related reading
- GuidesWorking out how many customers you can actually serveMost operators estimate capacity from their kitchen equipment and then discover the real ceiling somewhere else entirely — usually in the delivery window or in the coordination workload.
- GuidesPreventing meal delivery address errorsAddress problems look like small individual incidents and are actually one of the largest recurring costs in a delivery operation, because each one consumes two people at once.
- GuidesDeciding which areas you actually serveAlmost no meal-plan business decides its service area. It accumulates, one reasonable exception at a time, until a driver is spending ninety minutes reaching three customers.
- GuidesThe numbers a meal-plan business should actually trackMost meal-plan operators track revenue and headcount because those are easy. The numbers that predict next quarter are operational, and nearly all of them are already sitting in records you keep anyway.
- SolutionsRoute planning and driver manifests for meal deliveryA meal-plan round is not a set of ad-hoc jobs appearing through the day. It is a known list, knowable the night before — which makes it a planning problem rather than a dispatch problem.