One in 200 UK companies became insolvent in the last 12 months. Run that against a ledger of 150 customers and the odds of a failure this year are better than even. Here is the maths.
Over the past 12 months, one in every 200 companies in England and Wales entered insolvency.
Put like that, it sounds like someone else's problem. Half a percent. A rounding error.
So run it against your own ledger. The answer is less comfortable, and it explains why so many finance teams are surprised by a bad debt they assumed would never happen.
The Insolvency Service publishes company insolvency figures every month. In September 2026, 1,882 companies became insolvent in England and Wales, 4% fewer than in August and 5% fewer than in September last year. Over the 12 months to September, the insolvency rate was 49.9 per 10,000 companies on the register. That is one in 200.
Some context:
So the UK is not in an insolvency crisis, and 2026 has been slightly quieter than the three years before it. But the global picture is worse. Atradius expects business insolvencies worldwide to rise 5% this year, and for the UK to stay near historically high levels.
If every customer has a one in 200 chance of failing in a year, the chance that at least one of them fails depends on how many you have.
With 150 active customers, the odds are better than even that one of them fails in any given year. With 500, expect two or three.
These figures are a simplification. Your customers are not average. Larger, older companies fail less often than start-ups. Construction, retail and hospitality account for the most failures, so a ledger weighted towards those sectors runs higher. And customers in the same sector tend to get into trouble at the same time.
But the direction holds. Over a few years, one of your customers will almost certainly fail. The real questions are which one, and how much they will owe you on the day.
The first you cannot know. The second you can decide in advance.
Three features of how UK companies fail make it hard to see coming.
Most failures give no public warning. In September, 1,420 of the 1,882 insolvencies were creditors' voluntary liquidations, three in four. Those are directors closing the doors themselves. There is no winding-up petition in the Gazette for a supplier to spot first. The petitions that do appear are rising: compulsory liquidations, where a creditor goes to court, were up 25% on a year earlier.
Published accounts are out of date. Small companies file accounts up to nine months after their year end, and many file abbreviated versions. By the time a weak balance sheet is public, it can be well over a year old.
And the causes are slow, then sudden. We have read the administrators' reports for 433 recent UK administrations. The reasons they give are rarely a single shock:
Most failures are a squeeze that builds over months, followed by a lender or HMRC deciding it has waited long enough. The squeeze is visible if you know where to look. The ending usually is not.
Credit scores matter, but they lag. These signals often move first.
Payment drift. A customer who always paid in 45 days now pays in 70. Run your aged debt report and look for who has started paying later, not just who has stopped paying.
Disputes that appear from nowhere. Queries over invoices that would once have been paid without comment can be a way of buying time. Sometimes the dispute is pure tactics, as one of our clients found when a developer discovered a dispute only after a winding-up petition was filed.
Hiring. A firm that is recruiting has confidence and cash. One that has quietly stopped replacing leavers is telling you something its accounts will not show for months. UK employers have now cut staff for more than two years in a row, so this matters across the economy, not just at the margin.
Changes at the bank. A new lender, a new charge registered at Companies House, or a switch to invoice finance can all be perfectly normal. They are also worth a conversation.
What your sales team hears. The people who visit customers usually know before the credit report does.
There are three practical ways to limit what a customer failure costs you.
Set credit limits deliberately, customer by customer, and review them when the signals change rather than once a year.
Watch concentration. A single customer worth a quarter of your sales is upside and exposure in the same envelope.
Transfer the risk. Trade credit insurance sets a limit on each customer, watches them while you trade, and pays out most of the debt if an insured customer becomes insolvent or simply does not pay. It replaces an unpredictable loss with a known premium, and the insurer's credit team tends to spot trouble earlier than any one supplier can.
If you would like a feel for how risky a particular sector or company looks right now, our free Risk Index is a quick place to start. If one in 200 has made you think about a particular customer, talk to us.
