Shorts, Refusals, Redeliveries: What One Exception Costs

A case missing from a delivery. A chilled order refused at the door. A shop shut when the van arrives. Individually, delivery exceptions feel like noise — a few minutes of friction in a day of drops, sorted with a scribble on the delivery note. That instinct is exactly why they persist: because nobody prices a single exception end to end, nobody sees what the category costs.

This article does the pricing. It follows one exception from the doorstep through redelivery, credit note, warehouse rework and the payment delay it quietly triggers, then separates the two very different levers for dealing with exceptions — preventing them, and proving what happened when they occur. The numbers used are illustrative and labelled as such; the structure of the cost is the point, and it transfers directly to your own figures.

First, name the exception types

“Failed delivery” hides several distinct events, each with a different cost anatomy and a different fix:

• Short delivery — less arrived than was ordered, usually a pick or load error, sometimes a stock shortfall discovered late.
• Wrong item — the quantity is right, the product is not; a picking or substitution error.
• Doorstep refusal — the customer rejects some or all of the drop: condition, temperature, life dates, over-ordering, or a dispute about what was agreed.
• Failed drop — nobody there, premises closed, access impossible; the stock comes home untouched.
• Damage or condition change — the goods left fine and arrived unsellable.

Recording which of these happened — with a reason code, not a scribble — is one of the lowest-cost, highest-leverage improvements available, because everything later in this article depends on knowing the mix.

The anatomy of one exception's cost

Take a doorstep refusal of part of an order, and trace every cost it triggers. Assumptions are illustrative throughout — swap in your own rates:

Three things stand out in this example. The redelivery, not the doorstep moment, is the biggest line here — and where replacement goods can piggyback on a route already passing the door, that marginal cost falls, which is exactly why exception handling belongs inside route planning rather than outside it. The credit note drags office time from people who did nothing wrong. And the smallest-looking line, payment delay, scales with invoice value, not exception value: a £20 dispute can hold a £2,000 consolidated invoice out of your cash for weeks. In a low-margin operation, an exception can consume much or all of the margin of the drop it occurred on — run the arithmetic on your own gross margin per drop and see.

Two levers, often confused: prevention and proof

Exception programmes stall when these are treated as one thing. They are not.

Prevention attacks the causes: accurate picking and loading so far fewer shorts and wrong items board the van; delivery time windows planned around when the customer can actually receive, which is what reduces failed drops; life-date and temperature discipline so refusals lose their ammunition.

Proof attacks the ambiguity that remains: signature, photos, timestamps and recorded quantities at the doorstep. Proof does not stop the exception happening — what it stops is the second, more expensive argument about what happened, the one that turns a £20 short into a credited £200 because nobody can demonstrate otherwise. Electronic proof of delivery belongs in this column: it strengthens your position in disputes and speeds their resolution; the failure-reduction work happens upstream, in planning and picking.

Budget the two levers separately and the investment case for each becomes honest: prevention is priced against exception frequency, proof against dispute value and resolution time.

Measure the category, then manage it

Without reason codes the first metric is unavailable and the rest are guesswork — which is why the scribbled delivery note is not a small habit but the thing standing between you and managing this category at all.

Where RouteMagic fits

RouteMagic instruments the whole chain this article traced. Failed and partial deliveries are recorded at the visit with a reason code, then rescheduled or returned to the warehouse as a tracked movement rather than a mystery; returns are routed to the right stock condition and linked to the credit note they generate, so the paper trail this article priced assembles itself. Signature and photo ePOD sits on the order record for the proof side, and route planning with delivery time windows works the prevention side by helping plan deliveries within agreed receiving windows. The measurable outcome shows up in customer results: Covetrus, distributing animal-health products across the UK and Ireland, reports 87% fewer delivery delays since moving its rounds onto the platform.

Conclusion

The reason exceptions survive untouched in so many operations is that their cost is spread across four departments and two weeks — a few driver minutes here, a credit note there, an invoice paid late somewhere else. Assemble the pieces and a single refusal can cost a multiple of what the doorstep moment suggests — in the worked example, the second trip and the office rework carry most of the bill. The response is a measurement habit before it is an investment: reason-code every exception for a month, price your own version of the table above, and split the fixes honestly between prevention, which reduces how often exceptions happen, and proof, which reduces the dispute and administrative cost when one occurs. A month of reason codes and one priced table will tell you what the category costs in your own operation — and which lever deserves the next pound.