Executive Summary: This civil court docket lender guide walks an underwriting team through what a docket actually is, how to read one without legal training, and which fields matter for credit risk. A federal case file is a docket sheet plus the documents filed in the case, and the same basic structure carries over to the state courts where most lending-relevant litigation lives.[1] Since the large majority of cases are handled in state courts, a lender reading dockets is mostly reading thousands of separate state and county systems, not one national index.[2] This guide gives a practical reading method, then shows where structured data can spare your team from hand-reading every docket.
What is a civil court docket?
What does a docket contain?
A docket is the chronological record of a single case. It lists the parties, the case number, the court and division, the filing dates, the documents entered, and the disposition once the matter resolves. At the federal level a case file is the docket sheet plus the underlying documents, accessed through PACER, and state courts follow the same shape through their own portals.[1][3]
Why does a lender read it at all?
A docket is an independent record of how a business behaves under stress. It shows whether a borrower is being sued for unpaid debts, whether judgments have been entered, and how often disputes recur. Self-reported financials will not show any of that. A borrower who answers no to a litigation question at intake may still be a named defendant in three open collection matters, and the docket is where that gap surfaces before money moves rather than after a default.
How do you read a case caption and parties?
What does the caption tell you?
The caption names the parties and their roles. The plaintiff brings the action and the defendant responds. For credit risk, the role matters: a business that is repeatedly the defendant in collection or breach suits looks different from one that is occasionally a plaintiff enforcing its own contracts.
Why do exact-name matches matter?
Court systems index by name, and businesses share or resemble names constantly. A docket hit only counts if it ties to your borrower's verified legal entity, which is why entity verification precedes docket review. A near-match is a lead to confirm, not a finding to act on.
How do you read filing dates and disposition?
What do the dates signal?
Filing dates put the litigation on a timeline. A cluster of recent filings signals active pressure. A single matter filed and resolved years ago signals far less. Recency near the application date is the stronger credit signal because it may not have reached bank or credit data yet.
What does the disposition mean?
The disposition is the outcome: dismissed, settled, judgment entered, or still pending. A judgment entered against the borrower is the heaviest signal, because a recorded money judgment can attach to property as a lien and compete with your recovery.[4] A dismissal carries far less weight.
| Docket field | What to read | Credit-risk meaning |
|---|---|---|
| Caption and parties | Plaintiff vs defendant roles | Repeat-defendant pattern signals dispute pressure |
| Filing dates | Recency and clustering | Recent clusters may precede bank or bureau data |
| Disposition | Dismissed, settled, judgment, pending | Entered judgment is the heaviest recovery risk |
| Judgment amount | Dollar value where published | Amounts near deal size warrant a funding hold |
How does Cobalt return structured case data?
What does the API call look like?
Instead of opening each portal and reading dockets by hand, Cobalt's Court Records API returns structured case data through a callback-only endpoint. You send the business name, the jurisdiction, and a callback URL, and the result is POSTed back when the search completes.
curl --location 'https://apigateway.cobaltintelligence.com/courtCases?businessName=Acme%20Trucking%20Inc&jurisdiction=miamiDade&callbackUrl=https%3A%2F%2Fyour-app.com%2Fcobalt-callback' \
--header 'x-api-key: YOUR_API_KEY' \
--header 'Accept: application/json'What does it return, and what does it not?
The documented fields are judgment details, case information, filing dates, parties involved, and judgment amounts where available. That covers the caption, dates, and disposition reading above as parsed data your system can route on. The API does not stream live docket events such as new motions or hearing notices, so it is a structured case search, not a real-time docket monitor. Treat it as a point-in-time snapshot.
What is the coverage scope and fallback?
Which jurisdictions are supported?
The endpoint accepts `newYork` and `miamiDade`. Coverage is New York State and Miami-Dade County, Florida only, which is limited but targeted at venues dense with alternative-lending borrowers. It is not a nationwide litigation search and does not pull from PACER or federal courts.
What about everywhere else?
For any other state or county, the structured search does not apply, and the team falls back to the relevant court portal or a third-party vendor. California matters, for example, route to the California courts portal rather than this API.[5] Florida matters outside Miami-Dade route through the broader Florida courts system.[6]
A docket no one reads is not diligence. Structured case data lets a team review covered venues at volume and reserve hand-reading for the fallback jurisdictions and the genuine exceptions.
How should docket signals fit underwriting policy?
Who owns the decision?
Cobalt is a data source, not a decisioning engine. The lender sets the thresholds for when a judgment, a recent filing cluster, or a defendant pattern moves a file to manual review or a funding hold. This pairs naturally with the lien half of diligence in the pre-funding UCC lien search underwriter workflow and the sanctions half in BSA/AML for alternative lenders and OFAC's role.
What compliance limit applies to old records?
When docket-derived judgments inform a credit decision through a consumer reporting agency, the FCRA limits reporting of most civil judgments older than seven years or the governing statute of limitations, whichever is longer.[7] Counsel should set how stale dispositions are weighted.
How can a lender start using structured case data?
The simplest first step is to run a handful of current applicants in New York or Miami-Dade and read the parsed parties, dates, and dispositions next to the raw dockets your team already reviews. You can book a Cobalt demo to see the return fields, the callback flow, and how to route venues outside the two covered jurisdictions to a fallback.
.jpg){kind=small}
.png)