You’re pricing a job for a main contractor you haven’t worked for before. The enquiry looks solid, the drawings are decent, and the commercial vetting amounts to one phone call — to a bloke who subbied for them two years ago. “They’re slow, but they pay.” That sentence — second-hand, two years stale, impossible to check — is the entire credit assessment on a contract that could be worth more than your van fleet.
Here’s what almost nobody in a mid-market office seems to know. If that contractor is any real size, it has been filing its payment record with the government twice a year since 2017. The record is public. It is free. It takes about two minutes to read.
The report they’d rather you didn’t read
The rules are called the Reporting on Payment Practices and Performance Regulations, and they work like this. Any UK company big enough to clear two of three thresholds — turnover over £36 million, a balance sheet over £18 million, or more than 250 employees — must publish a report on a government website every six months, within thirty days of the period ending. Not filing is a criminal offence, for the company and for its directors. That net catches the national housebuilders and most of the main contractors a mid-market firm is likely to work under.
And the report isn’t marketing. It’s numbers: the average number of days the firm takes to pay an invoice; what share of invoices it settled within thirty days, within sixty, and beyond; and — the one I’d look at first — the percentage of invoices it failed to pay within the terms it had agreed. You’ll find it by searching for “check when large businesses pay their suppliers” — it’s a government service, on gov.uk, with a search box. Type in the customer’s name.
Then read it against what you’ve been assuming. The firm’s standard terms are in there, so you can learn they pay on sixty days before you’ve ever seen the contract. Across all large UK businesses the average time to pay is around thirty-two days, but the average is not the point — the spread is. Some firms pay in under three weeks. Others average close to double that and still miss a share of their own deadlines. Those two customers are not the same customer, and they shouldn’t get the same price.
Since last year, retentions too
If you’ve read anything of mine you’ll know about retentions — the slice of each payment, usually a few per cent, that the customer holds back for months or years as security against defects, and that has a habit of never coming home. Here’s the newer part. For financial years starting on or after April 2025, these same reports must say whether the firm puts retention clauses in its construction contracts, the standard percentage it holds, and how the money gets released. The first reports carrying those answers started landing at the end of last year, and more arrive every month.
Think about what that means. The question every subcontractor asks too late — what will they hold, and when will I actually see it — now has a published answer, signed off by a director, before you’ve priced a single brick.
Why nobody reads it
Not laziness. The information has no home. Watch how a tender actually gets decided in most firms: the enquiry lands, the estimator prices the work, and the judgement on the customer happens in the MD’s head, built from reputation and one phone call. The payment record lives on a website that is nobody’s job to open. And a fact that isn’t in the system where the decision gets made may as well not exist.
The firms that get this right treat it as a field, not a task. Somewhere in whatever runs their quoting there is a customer record, and on it: the customer’s standard terms, their published average days to pay, their published late percentage, their standard retention, and the date somebody last checked. When the estimator opens the job, the cash cost of a sixty-day payer is on the same screen as the margin. Then you can actually do something — price the wait in, negotiate terms with your eyes open, or decline politely and let a competitor make the loan instead.
This is the sort of place purpose-built software earns its keep precisely because it’s small. It is not a module. It is not a platform. It’s one screen on the customer record and one habit: no tender goes out until the screen is filled in. If your take-on process lives in Outlook and a spreadsheet, the check might still happen once — and then it’s a printout in a lever-arch file, invisible by the time anyone prices the next job for the same slow payer.
Where this doesn’t help
If your work is mostly domestic — homeowners, small landlords, little developers — forget it. Those customers sit under the thresholds and file nothing, and your credit control is deposits and staged payments, as it should be. And even for the big names, keep the data in its place. It’s self-reported, backward-looking, and up to six months old. A clean record is not a guarantee: a contractor can pay beautifully right up until the week it stops paying anyone at all, and construction loses more firms to insolvency than any other sector. The report tells you how a customer behaves. It doesn’t tell you whether they’ll survive. Keep the credit limits. Keep watching your own ledger.
The direction of travel, though, is one way. For financial years starting this January, large companies have to put these same payment figures into the directors’ report — the formal annual document the board signs — and the late-payment bill working through Parliament, which I’ve written about before, wants capped terms and mandatory interest on top. How your customers pay is becoming public record. The information has been sitting there, free, under a law that’s nine years old. The only question is whether your process has anywhere to put it. Get in touch.