What should be included in fulfilment cost per order?

Shipping at your real rate including surcharges, packaging including void fill and branded inserts, pick and pack labour at fully loaded cost, an apportioned share of storage, and the cost of returns spread across all orders. The last two are the ones normally left out, and together they usually add 30% or more to a naive postage-only figure.

How do I account for returns in fulfilment cost?

Spread the total cost of returns across every order rather than only the returned ones, because you cannot know in advance which order will come back. A return typically costs around three times a normal fulfilment once inbound postage, inspection labour and any markdown are counted, so a 10% return rate adds roughly 30% to the average.

What is a good fulfilment cost per order?

It depends far more on category and order value than on efficiency, so the useful test is the ratio rather than the absolute figure. Under about 15% of average order value is comfortable; above 25% the product economics need unusually high margins or strong repeat purchase to work. A £6 fulfilment on a £15 order is a problem; the same £6 on a £90 order is not.

Why is cost per order better than cost per unit?

Because most of the cost is triggered by the order rather than the item. Adding a second unit to an existing box costs seconds of labour and almost no extra postage, while a second order costs a full pick, pack and shipment. It is why raising units per order is usually the cheapest saving available.

How do I apportion warehouse storage to each order?

Take the monthly cost of the space — rent, utilities, racking and any warehouse management software — and divide by orders shipped that month. It converts a fixed cost into a variable one, which is legitimate for decision-making as long as you remember the cost does not actually fall when volume does, and recheck the apportionment whenever volume changes materially.

How to calculate fulfilment cost per order

Most people count postage and call it fulfilment. Add pick labour, packaging, the share of storage and the cost of returns, and the real figure is commonly half again as large.

Published 12 August 2026

What belongs in the number

Fulfilment cost per order is everything spent moving one order from a shelf to a doorstep, divided by orders. The parts people forget are the ones that do not arrive as an invoice.

Shipping. The carrier charge, at your actual negotiated rate, including surcharges. Fuel and residential surcharges are routinely left out of estimates and routinely appear on the bill.

Packaging. Box or mailer, void fill, tape, label, and anything branded. Branded tissue and inserts are marketing, but they are paid for out of the same order.

Pick and pack labour. The time to locate, pick, pack and manifest, at fully loaded cost — wage plus employer taxes plus the unproductive share of a shift. A picker paid £12 an hour costs closer to £15, and does not pick for sixty minutes of every hour.

Storage. Rent, utilities and racking apportioned across the orders that pass through. It is a fixed cost being made variable, which is legitimate as long as the divisor is honest.

Returns. The return postage, the inspection time, the repackaging, and the write-off on anything unsellable. This is the single largest omission in most estimates.

Returns are the part that breaks the model

A return does not cost one outbound shipment. It costs the outbound shipment you already paid for, the inbound postage, the labour to inspect and restock, and — often — a markdown because the item can no longer be sold as new.

Spread that across all orders rather than counting it only on returned ones. At a 10% return rate where each return costs three times a normal fulfilment, returns add roughly 30% to the average order's fulfilment cost. In apparel, where return rates of 25% and above are ordinary, returns can cost more than the original fulfilment did.

This is why category matters more than efficiency here. A warehouse cannot optimise its way out of a 40% return rate; only sizing information, better photography and honest descriptions move that number.

Why per-order beats per-unit

Most fulfilment cost is per order, not per item. Picking a second unit from the same shelf adds seconds; shipping it in the same box adds almost nothing. The expensive events are opening a new order, packing a new box and paying a new postage charge.

That has a direct consequence: raising units per order is usually the cheapest available saving. A business at 1.2 units per order that reaches 1.6 has cut its fulfilment cost per unit by roughly a quarter without renegotiating anything. Bundles, quantity breaks and free-shipping thresholds all attack that number.

Working it out honestly

Take a full month. Add every cost above for that month, including returns processing and the storage apportionment, and divide by orders shipped. Use a month rather than a week, because returns lag shipments and a short window flatters the figure.

Then compare it against average order value. Fulfilment above about 15% of AOV is worth investigating; above 25% and the product economics usually only work if margins are unusually high or the item drives repeat purchases.

Common questions

What should be included in fulfilment cost per order?: Shipping at your real rate including surcharges, packaging including void fill and branded inserts, pick and pack labour at fully loaded cost, an apportioned share of storage, and the cost of returns spread across all orders. The last two are the ones normally left out, and together they usually add 30% or more to a naive postage-only figure.
How do I account for returns in fulfilment cost?: Spread the total cost of returns across every order rather than only the returned ones, because you cannot know in advance which order will come back. A return typically costs around three times a normal fulfilment once inbound postage, inspection labour and any markdown are counted, so a 10% return rate adds roughly 30% to the average.
What is a good fulfilment cost per order?: It depends far more on category and order value than on efficiency, so the useful test is the ratio rather than the absolute figure. Under about 15% of average order value is comfortable; above 25% the product economics need unusually high margins or strong repeat purchase to work. A £6 fulfilment on a £15 order is a problem; the same £6 on a £90 order is not.
Why is cost per order better than cost per unit?: Because most of the cost is triggered by the order rather than the item. Adding a second unit to an existing box costs seconds of labour and almost no extra postage, while a second order costs a full pick, pack and shipment. It is why raising units per order is usually the cheapest saving available.
How do I apportion warehouse storage to each order?: Take the monthly cost of the space — rent, utilities, racking and any warehouse management software — and divide by orders shipped that month. It converts a fixed cost into a variable one, which is legitimate for decision-making as long as you remember the cost does not actually fall when volume does, and recheck the apportionment whenever volume changes materially.