Cost-to-serve: small orders and delivery terms that eat margin

Two customers can buy the same products at the same gross margin and leave you with very different profit. The difference is the cost of serving them: how many orders they place, how often you deliver, how urgent they are, and how long they take to pay. Gross margin does not show any of that.

This is article 4 in a series of five. If you are reading in order, start with Why your average margin lies, which shows how a single blended margin hides what individual customers and products earn. Article 2, The long-tail products that quietly cost you money, looks at the product side. Article 3, Passing on UK cost rises without losing customers, covers pricing. This article takes the next step: from gross margin to what is left after serving each customer. Article 5, What a Business Health Check looks at, and what owners learn, explains how the whole method fits together.

What cost-to-serve actually includes

Cost-to-serve is the cost of everything you do for a customer after the product has been made or bought. In a manufacturing or wholesale business it usually has five parts:

• Order handling: entering, confirming and invoicing each order, plus the queries that follow.
• Picking: the time to find and assemble the items. Ten lines of one unit each take longer than one line of ten.
• Packing: cartons, pallets, labels and paperwork, which are often similar for a small order and a large one.
• Transport: vehicle, fuel and driver time per drop, and whether the drop is shared with other customers.
• Credit: the cost of waiting for payment, and of chasing it when it does not arrive.

Most of these costs are fixed per order or per drop, not per pound of sales. That is why small orders hurt: the cost does not shrink to match the order value.

An illustration

The numbers below are invented for illustration only. They are not drawn from any real business.

Customer A places four orders a year of £5,000 each: £20,000 of sales at a 30% gross margin, so £6,000 of gross profit. Customer B places 120 orders of £250 each: £30,000 of sales, also at 30%, so £9,000 of gross profit. On the usual report, B looks the better customer.

Now assume each order costs £60 to handle, pick and pack, a delivery costs £40 for A's large drops and £18 for B's small ones, and B pays after 60 days while A pays after 30. Treat the cost of waiting as 8% a year on the money outstanding for the days each customer takes to pay, again as an assumption.

• Customer A: £240 of order costs, £160 of delivery and about £131 of credit cost leave around £5,470 of profit, or roughly 27% of sales.
• Customer B: £7,200 of order costs, £2,160 of delivery and about £395 of credit cost come to £9,755, against £9,000 of gross profit. B loses about £755 a year.

The customer with 50% more sales is the one that loses money. Nothing in the gross margin report would have shown this. It is the same effect described in article 1, applied to customers instead of products.

Why payment terms belong in the calculation

Credit is a real cost, and the UK is moving to make slow payment harder to sustain. The government response to its 2025 late payment consultation, published in March 2026, states that late payment costs the UK economy £11 billion a year, and that affected business owners spend on average 86 hours a year chasing invoices. It sets out an intention to introduce maximum payment terms of 60 days, with limited exemptions, and to make a right to statutory interest at 8% above the Bank of England base rate a requirement of commercial contracts. The proposed change starts with 60 days and is not expected before 2027, so it is an intention, not yet a rule.

The rules that already exist matter too. GOV.UK states that for business-to-business debts the statutory interest rate is 8% plus the Bank of England base rate, unless the contract sets a different rate. It also lists fixed sums that can be claimed as recovery costs: £40 for debts under £1,000, £70 for debts from £1,000 to £9,999.99, and £100 for £10,000 or more. These are reference points, not advice on what you should do with a particular customer. They do show how seriously the law treats the cost of waiting.

The practical point is simpler. A 60-day customer ties up twice as much of your cash as a 30-day customer for the same sales, and that cash has a cost whether or not you borrow to cover it.

Four decisions that change the answer

Once you can see cost-to-serve by customer, four levers are usually worth considering:

1. Minimum order value. Set it at the level where the average order covers its own handling cost with margin left over. Existing customers may need a transition period.
2. A delivery charge or a threshold for no-charge delivery. A threshold keeps the benefit for customers who order enough to justify it and nudges others to consolidate.
3. Order days. Fixed delivery days per area reduce part-empty drops. Urgent orders can be offered at a stated premium.
4. The price of time. If a customer wants longer payment terms, treat the extra days as a priced item, not a default.

Choose one change, not all four. A single clear change is easier to explain to customers and easier to measure afterwards.

What to check this month

• List your 10 smallest orders from the last quarter and estimate what each one costs to handle, pick, pack and deliver.
• List the 10 customers with the most complicated terms: frequent drops, urgent orders, delivery at no charge, long payment terms, regular returns.
• For each, compare gross profit with your estimated cost-to-serve. Mark any that fall below zero or below your target.
• Estimate the cost of payment terms using a rate you are comfortable defending, such as your overdraft rate.
• Pick one change from the four above, decide how you will explain it, and set a date to review the result.

An estimate is enough to begin. The aim is to find the customers where the gap between gross margin and real margin is widest, and decide what to do about one of them.

If you would like to see this view for your own business, the Business Health Check looks at margin by customer and by product, and where cost-to-serve changes the picture.