How to Run a Business Credit Check: A B2B Buyer's Guide

A business credit check tells you what a company's payment behavior looked like before today. It does not tell you what it will look like in 90 days. That distinction is the entire argument for why credit programs built entirely around bureau pulls are structurally behind the risk they're supposed to catch.

That said, credit checks are table stakes. Before you extend trade credit, you need to know whether this company pays its bills. Here is what a business credit check actually covers, which sources are worth using, and what to do when the report gives you a number but not an answer.

What a business credit check actually shows

Business credit reports aggregate data from a few sources: payment data from trade lines (vendors who've reported the account), public records (court judgments, liens, bankruptcies), and bank data where available.

A typical report gives you:

• A credit score or risk rating. D&B's PAYDEX score, Experian Business's Intelliscore, or Equifax's Payment Index. These are blended scores, not uniform standards. A PAYDEX of 80 is not the same as an Intelliscore of 80. Comparing scores across bureaus without understanding the methodology will give you a false sense of precision.
• Payment history. How the company pays its trade vendors: on time, slow-to-30, slow-to-60, and so on. The most useful data point is trend, not the snapshot number.
• Trade lines. The specific vendors reporting payment behavior. A company with three trade lines is not the same risk as one with 30. Thin file coverage is the norm for private mid-market companies.
• Public records. UCC filings, liens, judgments, bankruptcies. These are highly actionable and should be checked before anything else.
• Financial data. Only for companies that have voluntarily disclosed financials to D&B or similar. Most small and mid-sized businesses have not.

The gap that most credit managers underestimate: the majority of private companies have sparse trade line coverage. If your prospective customer is a $40M distributor that does not share payment data with bureaus, their PAYDEX score is thin or nonexistent. A "no file" result is common. It tells you nothing about actual creditworthiness — and assuming low risk because of a missing file is how bad debt happens.

The sources worth using

Dun & Bradstreet (D&B) is the largest commercial credit database by coverage. The DUNS number is the de facto business identifier. If you're running one report, D&B is the starting point. D&B has the data. What D&B does not have is a workflow layer that tells your credit team what to do with that data, or a monitor that watches accounts between pulls. It is a data provider. It is not a credit decision platform, and treating it as one is how teams end up with a warehouse full of PDFs and no action items.

Experian Business has competitive trade line coverage and its Intelliscore is useful for small business credit. A solid complement to D&B for lower-volume accounts or when D&B coverage is thin.

Equifax Business is thinner on commercial trade line data than D&B or Experian but has strong consumer crossover data, which matters for sole proprietors and small LLCs where personal and business credit are intertwined.

Cortera pulls trade payment data directly from businesses, bypassing the bureau aggregation model. Better signal for mid-market companies that have not opted into bureau reporting. Worth running alongside a bureau pull for accounts above your significant-exposure threshold.

NACM trade groups — the National Association of Credit Management runs credit groups by industry where members share payment data on mutual customers. For specific verticals like food distribution, building materials, or manufacturing, this is more accurate than any bureau because the data is direct, recent, and from vendors extending similar terms to yours. If you are a credit manager in one of these industries and you are not in an NACM group, the bureau pulls you're running are the second-best source of data, not the first.

When to run a credit check

New customer onboarding. Every new credit account should get a bureau pull as part of the credit application process. This is the minimum. It does not replace trade references, but it surfaces public record flags faster than any reference call will.

Credit limit increases. Before approving a significant increase, run a fresh pull. A customer's bureau profile from 18 months ago may not reflect the current state of their business. Lean on the trend data, not just the current score.

After a payment behavior change. If an account that paid net 30 starts slipping to 45 or 60, that is a signal. Pull a new bureau report and compare it to the prior one. More importantly, check for new UCC filings or public record activity — these often appear before payment deterioration shows up in your AR.

Annual review cycle. Standard practice is to re-pull credit on accounts above a certain limit threshold once a year. This is better than never. It is not the same as monitoring.

What to look for in the results

Public records first, always. A judgment or lien filed in the last 90 days is more actionable than a PAYDEX of 75. A recent UCC filing on accounts receivable or inventory means the company has pledged those assets as collateral — which changes your recovery position in a default scenario. Check public records before you even look at the score.

Trade line count and recency matter more than the score headline. Three trade lines from vendors who last reported 18 months ago is not meaningful coverage. You want recent data from vendors extending similar payment terms to yours.

Score trend, not just score level. A company that scored 72 a year ago and scores 68 today is a different risk than a company stable at 68 for three years. Most bureau reports show historical trend data. Use it.

The thin file problem. When a company has sparse bureau data, do not default to assuming low risk. Assume you have incomplete information. Supplement with trade references, financial statements if available, and industry-specific data from NACM groups. Research agents can also surface UCC filings, lien activity, news, and financial distress markers that do not require the company to have opted into bureau reporting.

The limitation most programs ignore

A business credit check is a snapshot. It reflects the company's credit profile at the moment the report was pulled. A customer can pass a bureau check in February, slip behind on five vendor accounts in March, and have a lien filed against them in April. If your next scheduled review is in 12 months, you will not see it until it is in your bad debt column.

This is not a flaw in credit bureaus specifically. It is a structural problem with any point-in-time process. The analyst who approved Harvest Sherwood Food Distributors had reasonable information on the day they approved the account. What they did not have was a monitor that would have flagged the supply chain concentration risk and cash flow deterioration that preceded the bankruptcy.

The answer is not to run bureau checks more frequently — that gets expensive and creates noise without improving signal quality. The answer is to monitor the indicators that move faster than bureau data: payment behavior in your own AR, UCC activity, news, and financial stress markers. B2B credit risk monitoring is the layer that goes on top of the point-in-time bureau check. One without the other is an incomplete program.

How credit checks fit into a full credit process

A bureau pull is one input among several for a new account decision:

1. Credit application — collects company information, references, and consent to pull credit
2. Bureau report — D&B or Experian, for public records and payment history
3. Trade references — direct contact with existing vendors (treat as supplemental; vendors rarely give negative references on accounts they still want to keep)
4. Financial statements — for larger accounts or borderline decisions where the bureau data is thin
5. Internal data — if the prospect has an existing relationship with your business or a related entity, your own payment history is the most reliable signal you have

Your credit management software should pull bureau data automatically, surface public records, and score the account against your credit policy without requiring an analyst to manually assemble five browser tabs. If your team is running bureau checks by hand, copying results into a spreadsheet, and emailing decisions back to sales, that process is the bottleneck — not the data quality.

Once the account is open, the bureau check recedes in importance. What matters is whether the customer's payment behavior in your AR matches the risk profile you approved them at, and whether external signals are moving in the wrong direction. That is trade credit management done properly — not just at the point of approval, but between every review cycle you will ever run.

The goal is not a better bureau report. The goal is an account portfolio where deteriorating customers surface before they default, not after. Bureau checks get you to the starting line. Continuous monitoring wins the race.