Executive Summary: Court records commercial insurance underwriting turns public litigation history, judgments, and judgment liens into risk-selection and pricing signals before a commercial policy is bound. A judgment is a court's final decision on a dispute, and a judgment lien is a claim attached to a debtor's property when a money judgment goes unpaid.[1][2] This guide shows commercial insurance and trade-credit underwriters how to read those records as forward-looking loss indicators, how the signal differs from lending use, and where a court-records data source like Cobalt fits inside a verification stack.
Why does court records commercial insurance underwriting matter at risk selection?
What does litigation history reveal about a risk?
Underwriting profit depends on selecting risks whose expected losses sit below the premium charged. Risk selection criteria are often unanswered or unverifiable on an application, forcing underwriters to research the file, decline it, or accept it at face value and absorb the loss-ratio consequence.[6] Court records give the underwriter an independent control. A business with active litigation, unsatisfied judgments, or a pattern of disputes carries financial stress a clean application will not show.
The highest-value pattern is not a single old lawsuit. It is a dense docket: recent filings, unpaid judgments, and the business appearing repeatedly as a defendant. That history correlates with the instability that drives claims, disputes, and non-payment.
How is this different from a lender's use of court records?
A lender reads court records to estimate recovery and repayment risk before funding. An insurer reads the same records to estimate future claim frequency, severity, and the chance of a contested or fraudulent claim. The data is the same; the decision is different. For trade-credit underwriters the logic sits closer to lending, since trade-credit insurance protects a seller against a buyer's non-payment, and an unsatisfied judgment against that buyer is a direct default signal.[9]
The application tells the underwriter what the applicant chose to disclose. The public docket tells the underwriter what a court already decided.
What court signals map to underwriting loss?
Which records carry the most pricing weight?
Not every filing deserves the same weight. The underwriter wants records that predict loss, not noise from routine commercial activity. Public court systems in major jurisdictions expose case search by party name, so these records are obtainable.[10]
• Unsatisfied money judgments. An open judgment against the business is the strongest single signal of financial distress and unpaid obligations.[1]
• Judgment liens on business property. A lien ties a judgment to assets and outranks claims filed after it, which signals encumbered collateral and prior creditor pressure.[2]
• Frequent defendant status. Repeated lawsuits naming the business as defendant point to contractual, employment, or quality disputes.
• Recent filing velocity. Several new cases inside a short window suggest a deteriorating situation that premium history will not yet reflect.
• Plaintiff-side litigation pattern. A business that sues often can still signal contentious operations and higher dispute cost on a policy.
These signals feed pricing and selection, not a single automated decline. The underwriter combines them with exposure, industry, and prior loss experience.
How does this connect to loss ratio and adverse selection?
Loss ratio is incurred losses plus loss-adjustment expense over earned premium, and a book that selects high-litigation risks at standard rates pushes that ratio up.[7] Court records help counter adverse selection. When the insurer cannot see what the applicant omitted, the worst risks self-select into mispriced policies, and an independent litigation check narrows that gap before the policy binds.
Which commercial lines use these signals most?
What lines benefit from a litigation check?
Commercial lines span workers' compensation, commercial auto, commercial multi-peril, and other liability coverages, each with its own loss drivers.[5] Court records add the most value where financial stability and dispute history predict loss.
| Coverage line | Court signal of interest | Underwriting use |
|---|---|---|
| Trade credit | Unsatisfied judgments against buyer | Default and non-payment probability |
| Management liability | Litigation pattern, regulatory cases | Claim frequency and governance risk |
| Commercial general liability | Defendant history, dispute volume | Contested-claim and severity risk |
| Surety and bonds | Judgment liens, financial distress | Principal solvency and recovery |
| Commercial property | Distress signals, ownership disputes | Moral hazard and claim integrity |
Where do credit-style signals already fit?
Insurers already use credit-based insurance scores to estimate claim likelihood, though state law limits using them as the sole reason to deny, cancel, or non-renew.[4] Court records belong to the same evidentiary family: a financial-stability signal used alongside, not in place of, the rating plan. The underwriter treats a judgment as one input that can move price or trigger review.
How does a data source like Cobalt return court records?
What does the API actually do?
Most teams know the manual path: searching state court portals and county clerk systems one business at a time. Public systems such as the Miami-Dade Clerk and the New York court system expose civil case search, much as federal case files are reached through the courts' own record systems, but checking each by hand does not scale across an underwriting queue.[3][8][10] A court-records API automates that lookup. Cobalt's Court Records API is a data source, not a decisioning engine. It returns case information, filing dates, parties, and judgment amounts where the court makes them available, and the underwriter's policy decides what the signal means.
The endpoint runs asynchronously. A request is submitted with the business name and jurisdiction, and results return to a callback URL.
curl --location 'https://apigateway.cobaltintelligence.com/courtCases?businessName=Acme%20Holdings&jurisdiction=newYork&callbackUrl=https://yoursite.com/callback' \
--header 'x-api-key: YOUR_API_KEY' \
--header 'Accept: application/json'What are the honest coverage limits?
Cobalt's Court Records API covers two jurisdictions only: New York State and Miami-Dade County, Florida. It is not a nationwide litigation screen, and a business may have cases in courts this product does not reach. It is also asynchronous and requires a callback URL, and judgment amounts appear only when the court record includes them. The right framing is precise: a deep check in two markets that matter for many commercial risks, not a national clearance.
What workflow turns records into decisions?
How should an underwriting team route the signal?
The first policy pass can stay simple. No relevant records means the file continues on the standard rating path. One older, satisfied judgment means continue with a note. Open judgments, judgment liens, or several recent filings should trigger referral to a senior underwriter for pricing, conditions, or decline. A jurisdiction the data source does not cover routes to a supplemental search, never to silent clearance.
What should the file store for audit?
Store the raw API response, the parsed signals, and the decision reason. For a compliance officer, court-record checks support a reasonable-diligence standard, and a stored response log is the audit trail that proves the litigation review happened. The decision reason tells a future reviewer why the underwriter priced, conditioned, or declined the risk.
How does this fit a full verification stack?
What should run alongside court records?
Court data is strongest paired with other checks. Each layer answers a different question, and no single layer is sufficient on its own.
• Entity verification. Confirm legal name, status, and registration before any other lookup.
• UCC lien data. Detect existing secured creditors and overlapping collateral claims.
• Court records. Surface judgments, judgment liens, and litigation history.
• TIN and identity matching. Confirm the business is who the application says it is.
• Sanctions screening. Check the business and principals against watchlists.
For deeper court-records mechanics, Cobalt's guide to judgment lien searches walks through the data behind these signals, and the equipment-finance underwriting guide shows the same logic applied to a lending file.[11][12]
Where does the check belong in the timeline?
Place the litigation check before binding, while the result can still change price, terms, or the decision to write the risk. A judgment discovered after the policy is bound is a claims problem, not an underwriting control. Run it early enough to act on, late enough that the legal name and entity are confirmed.
What should teams build first?
What does a minimum production version include?
Build version one around intake, entity-name confirmation, court-records lookup, referral routing, and coverage-gap handling. Keep the policy clear enough that underwriting, actuarial, compliance, and engineering describe it the same way.
1. Confirm the legal entity name and operating jurisdiction.
2. Run the court-records lookup against supported jurisdictions.
3. Parse judgments, liens, filing dates, and party roles.
4. Apply referral rules before binding.
5. Store the raw response, parsed signals, and decision reason.
6. Route unsupported jurisdictions to a supplemental search.
7. Review referral outcomes against loss experience and adjust thresholds with actuarial sign-off.
How do teams measure whether the signal works?
Treat court records as a testable input, not a permanent assumption. Tag each bound policy with its court signal at underwriting, then review loss experience by signal tier once enough policies mature. If files flagged for open judgments produce worse loss ratios than clean files, the referral rule earns its place. If they do not, actuarial and underwriting can retune the thresholds. That feedback loop keeps the litigation check tied to real loss outcomes rather than intuition.
.jpg){kind=small}
.png)