State court systems: a developer's guide

Federal courts get most of the attention in legal technology, but more than 95% of all litigation in the United States is filed in state courts — not federal courts. Debt collection judgments, evictions, personal injury cases, criminal prosecutions, family law matters, and most commercial disputes below the federal jurisdictional threshold all live in state court systems.

For AI agents operating in legal, financial, and compliance workflows, state court docket data is often more directly relevant than federal court data — and significantly harder to access programmatically. This guide maps the state court technology landscape, explains why it is so fragmented, and describes what that means for developers building automated monitoring systems.

How state court systems are organized

Each US state operates its own independent court system. While there is significant variation across states, most follow a three-tier structure similar to the federal system.

Trial courts of general jurisdiction handle the most serious civil and criminal matters. Depending on the state these are called Superior Courts, District Courts, Circuit Courts, or Courts of Common Pleas. These are the primary courts for commercial litigation, felony criminal cases, and major civil disputes.

Trial courts of limited jurisdiction handle smaller matters — misdemeanor criminal cases, small claims, traffic violations, and landlord-tenant disputes. These are called Municipal Courts, Justice Courts, Magistrate Courts, or County Courts depending on the jurisdiction.

Appellate courts hear appeals from trial courts. Most states have an intermediate appellate court and a supreme court. The total number of state court filings in the United States runs into the tens of millions annually — orders of magnitude larger than federal court filing volume.

Why state court technology is fragmented

Unlike the federal system, where the Administrative Office of the US Courts oversees technology for all 94 district courts, state courts are governed independently. Each state funds, procures, and operates its own court technology infrastructure. Within states, individual counties often make their own technology decisions independently of the state.

The result is a landscape of extraordinary fragmentation. Across 50 states and thousands of counties, dozens of different case management systems are in use — ranging from modern cloud-based platforms to legacy systems built in the 1980s that have never been replaced. Some jurisdictions have no electronic access at all; a courthouse visit is the only way to obtain case information.

There is no state-court equivalent of PACER. No single login, no unified data model, no common API.

The technology landscape: three tiers

Federal courts vs state courts: key differences for developers

Dimension	Federal (PACER)	State courts
Governing body	Administrative Office of US Courts	50 independent state systems
Unified login	Yes — PACER	No equivalent
Underlying platform	CM/ECF — one system	Dozens of different systems
Data consistency	Inconsistent but mappable	Highly inconsistent
Programmatic API	No — HTML scraping	No — portal scraping, more complex
Filing volume	~400,000 new cases/year	Tens of millions/year
Case types	Civil, criminal, bankruptcy	Civil, criminal, family, probate, landlord-tenant, small claims, juvenile, traffic
Integration complexity	High	Very high

What is actually accessible publicly

Across state court systems, public access follows a consistent pattern — available through web portals, not APIs, with wide variation in what is exposed.

Most state court portals provide basic case information (case number, party names, case type, filing date), docket entry listings (filing type and date, sometimes a brief description), and court calendar information. Document content — the actual text of filed documents — is frequently restricted, paywalled, or unavailable without attorney credentials.

Real-time updates are rare. Most portals are batch-updated rather than refreshed continuously. Historical depth varies significantly — some portals only show current open cases.

The normalization problem at scale

The fundamental technical challenge in aggregating state court data is normalization. Each platform uses different field names, different case type codes, different party role labels, and different filing category taxonomies. A "Motion for Summary Judgment" in a Tyler Odyssey court in Texas may be categorized differently from the identical filing in a C-Track court in California or a proprietary system in New York.

Manual mapping rules per jurisdiction are not feasible at scale — writing and maintaining court-specific parsers for thousands of counties is a permanent engineering burden with no end state. DocketLayer addresses this using an AI normalization layer that converts raw portal data into a consistent schema regardless of source system. This is what makes state court monitoring tractable as coverage expands.

DocketLayer's state court expansion

DocketLayer Phase 2 targets Tyler Odyssey courts in Texas and California — the largest Odyssey deployments in the country, covering a combined population exceeding 60 million people. Both states have relatively consistent county-level Odyssey implementations, making normalization more tractable than states with more fragmented technology landscapes.

Phase 3 targets high-volume non-Odyssey jurisdictions: New York, Florida, New Jersey, and Pennsylvania — major commercial litigation states with proprietary or fragmented systems that require more bespoke integration. The full expansion roadmap is at /coverage.