One number,
built from the public record.

What goes into it

The score reads a company's most recent filed accounts and turns them into thirteen financial ratios: profitability measures like operating margin and ROCE, liquidity measures like the current ratio and acid test, and funding measures like gearing and interest cover. The ratios are grouped into three components:

• Profitability (42%): does the trade itself make money.
• Asset quality (38%): can the balance sheet meet its obligations.
• Funding (20%): how much the company leans on borrowed money.

The three components combine in a nonlinear model that rewards balance and penalises a weak leg, because companies rarely fail on every measure at once. The full formula, weights and validation evidence are published in the methodology whitepaper.

Signals outside the accounts

Accounts are filed up to nine months after the year-end, so the score is also adjusted by live signals from the public record: CCJs from HMCTS court records, winding-up petitions and insolvency notices from The Gazette, charges, and overdue filings. An active winding-up petition forces the risk rating to 10, whatever the accounts say.

How to read it

Higher is better on the 0–100 gauge. The paired risk rating works the other way: 1 is the lowest risk, 10 the highest. A company with too little usable filing data shows as Unscored rather than guessed at, and every profile states its data quality so you can see how much the filing supports the number.

One more piece of context matters: the profile also shows a sector percentile, so you can see whether the score is weak for that trade or weak in absolute terms. A thin margin is routine in distribution and alarming in software.