The £22,000 figure has done its job. It is in the trade press, in policy papers, in the introductions to vendor pitch decks. As a headline, it works. As an actionable number, it is incomplete. The same £22,000 means three different things in three different businesses. Reading the figure as a single number leads finance leaders to under-estimate exposure in some cases and overestimate it in others. The deconstruction below makes the figure operational.
Where the headline number comes from
The £22,000 figure is drawn from the UK Industry Benchmark dataset and represents the average annualised cost of late payment for a UK business in the sample, blended across sectors and business sizes. The methodology is published. The number is defensible. Its weakness is in its averaging, not in its construction.
The line-item breakdown
Inside the £22,000, four cost components are doing the work. The proportions vary by business size and sector, but the categories are stable.
Financing cost. The working capital tied up in late receivables, multiplied by the relevant cost of capital. For most mid-market UK businesses in 2026, the cost of capital sits between 7 and 12 per cent, so a £200,000 receivables overhang at 60 days adds £2,000 to £3,300 to this line.
Time cost. The hours spent chasing late payments, managing exception queues and handling supplier disputes. At an FD or AP manager's loaded cost, two hours a week on chase activity equals £6,000 to £10,000 a year. The figure is often higher in businesses that have not invested in AP infrastructure. We have written about the hidden cost of late payments nobody measures in more detail.
Fraud exposure cost. The probability-weighted cost of AP fraud, calibrated against the volume of bank-detail changes the business processes manually. Where the business processes 20 bank-detail changes annually with manual verification, the expected loss runs between £2,000 and £5,000. This line is rising fast for the reasons set out in why AP fraud will explode in the AI era.
Churn cost. The supplier turnover attributable to late payment, captured through replacement cost, capacity premium and pricing degradation. This is the hardest component to quantify but often the largest. Industry estimates run between £4,000 and £12,000 annually for a mid-market UK business.
The four components add up to a range that brackets £22,000 cleanly. Where individual businesses sit higher or lower depends on which components dominate their profile.
The cases where the real number is much higher
Three patterns produce a real cost that exceeds the headline.
Concentrated supplier base. Where the top fifty suppliers represent more than 60 per cent of spend, the churn cost line accelerates. The replacement cost of a strategic supplier paid late is much higher than the average supplier.
Recent fraud incident. Businesses that have absorbed a fraud loss in the prior 24 months tend to under-invest in process changes that prevent the next one. The expected-value fraud line therefore stays elevated. A single £80,000 loss in year one materially raises the rolling annual cost.
Construction or agency sector. Both sectors run elevated lateness against their own suppliers, so the cost line is structurally higher than the cross-sector average regardless of business size.
What proportion is recoverable through behavioural visibility
Not all four lines are equally addressable. Visibility into payment behaviour changes the calculus on each one differently. The structural mechanism sits in payment behaviour should be underwritten, not promised.
Financing cost. Modestly recoverable. Better forecasting and earlier exception handling reduce the working capital tie-up, but the underlying receivables balance is determined by contract terms and customer behaviour, not by visibility alone.
Time cost. Substantially recoverable. The hours spent on chase activity collapse when the business operates against verified behavioural data. Recovery of 60 to 80 per cent of this line is realistic.
Fraud exposure cost. Substantially recoverable. Continuous identity verification and behavioural baselining reduce the expected loss line materially. Recovery of 70 to 90 per cent of this line is realistic in businesses currently running manual verification. Supplier verification and payment behaviour are converging into the discipline that does this work.
Churn cost. Variable recovery. Network-level visibility into the buyer's own behaviour reduces the supplier-attrition rate, but does not eliminate it. Recovery of 30 to 50 per cent of this line is realistic.
A blended recovery of 50 to 65 per cent across the £22,000 is a defensible estimate for most mid-market businesses, which lands the recoverable figure between £11,000 and £14,000 per year.
A self-assessment
The figure becomes more useful when each business calculates its own. Three inputs are sufficient: average daily AP exception queue size, annual bank-detail change volume, and the share of spend in the top fifty suppliers. Plug those into the four components above and the headline figure becomes specific.
.jpg)
.jpg)