Data Sources & Methodology
Every page on Docket Nexus is generated deterministically from public records. No content is hand-written or AI-narrated; when the data is absent, the page says nothing rather than guessing.
Sources
- Federal court dockets — bulk docket data for U.S. district courts, as collected in the CourtListener / RECAP archive maintained by the Free Law Project. Federal court records are public domain.
- Judge biographies — the Federal Judicial Center biographical directory, via CourtListener's people database (appointing president, confirmation date and vote, service status, birth year).
- Company identity & industry — U.S. Securities and Exchange Commission EDGAR filings: company names are matched to SEC registrants, and industries derive from each registrant's SIC classification. Industry labels exist only for parties that match an SEC filer.
- Statute text — verbatim United States Code text from the U.S. Government Publishing Office (govinfo.gov), public domain.
Scope
The index covers civil cases in U.S. district courts, restricted to commercial case categories. Categories dominated by sensitive personal matters — immigration, prisoner petitions, Social Security, and similar — are excluded by design, as are administrative and placeholder dockets.
Update Cadence
The core docket index is refreshed from quarterly bulk releases of the court archive. Per-case filing details are added continuously on a rolling basis. Each case page notes its source docket; the official court record is always the authority.
Known Limitations
- Party grouping is by name. Records that share a name are grouped together and may describe different people or entities; corporate name variants may appear as separate pages.
- Quarterly lag. Recently filed cases appear after the next bulk refresh.
- No outcome data. Dockets record filings and termination dates, not rulings or "win rates" — so this site does not present any.
- Attorney information is not currently included.
Corrections
If a record appears wrong, see the contact page — corrections are verified against the official docket and applied to the underlying data, not patched cosmetically.