Why do business credit scores differ between providers?
Business Credit Reports
Answer
Business credit scores can differ because providers use different source data, update schedules, matching methods and statistical models. They may also predict different outcomes over different horizons. Compare the legal entity, assessment date, financial period and model definitions before treating a difference as an error or choosing which score to use.
Check for an evidence difference before a model difference
One report may use more recent accounts or a different set of payment experiences. Another may cover a parent rather than the operating subsidiary. Those differences can matter before any score formula is applied.
- Match registration number and country.
- Compare standalone and consolidated accounts.
- Check assessment dates and source periods.
Reconcile the decision, rather than averaging the numbers
If the inputs match, read the model’s target outcome and risk bands. Define which model your policy uses for which decision. Escalate disagreements that would change a material approval and record how your team resolved them.
Illustrative example
A date mismatch can look like a model disagreement
Illustrative scenario: one score reflects the latest filing while another still uses an earlier financial year. Before replacing either provider, check whether the newer figures, filing timestamps and assessment refresh are present in both records.
What to check
- Same entity
- Same financial basis
- Comparable dates
- Model and policy mapping
Frequently asked questions
Does a difference mean one provider is wrong?
No. Different inputs or model purposes can produce different assessments.
Can payment samples differ?
Yes. Providers may receive different reported experiences.
Can company matching cause a difference?
Yes. Parent, subsidiary and similar-name records can be confused.
Should I use the newest score?
Recency matters, but so do model suitability and the underlying evidence.
Can I map scores with a simple percentage?
Not reliably without a validated comparison of outcomes and populations.
Should I average the scores?
Avoid averaging incompatible scales or model targets.
Can a model revision change a score?
Yes. Check whether a revision changed the scale, inputs or risk interpretation.
How should procurement compare providers?
Test matching, completeness, refresh, methodology and decision usefulness on the same company sample.
What do I do with a suspected data error?
Preserve the source evidence and use the provider’s correction process.
What belongs in the audit trail?
The reports, dates, model definitions, discrepancy and documented resolution.
Sources & further reading
Source references support the definitions. Process suggestions and examples are editorial guidance.
- Dun & Bradstreet: business credit scores and ratings
An example of model-specific score definitions; these are not universal scoring bands.
- Creditsafe: company scores and recommended credit limits
One provider’s explanation of its score and recommended limit methodology.
Business Credit Reports provides commercial company information. Apply your organisation’s policy and appropriate specialist review. Suggest a correction.
For more detail: Business credit scores explained →
Need help with a company report?
Talk to our team about company coverage, reports or data access.