.jpg)
Every US public company files with the SEC on a schedule. Annual reports, quarterly reports, and disclosures of material events all land in EDGAR, and together they form the source of truth for company fundamentals. The data is public and it is free.
The problem is not access. The problem is shape. EDGAR hands you documents, not records. If you want to know which companies filed an 8-K yesterday, or pull the last ten years of income statements for a screening model, or build a research tool that reads disclosures, you need that filing history as something you can query. That is the job an SEC filings API does.
An SEC API sits between EDGAR and your application and turns filings into structured, queryable records.
At the reference layer, it returns a filing object for every submission: the filing date, the date the SEC accepted it, the period the filing covers, the report type, a unique SEC identifier, and URLs to the filing itself. Each record is tied to a company through consistent identifiers, so a ticker, a CIK, or an LEI all resolve to the same entity.
That structure is what makes the data usable. Instead of crawling an index and parsing HTML, you request filings filtered by company, report type, date range, or industry, and page through the results. You can ask for every 10-K a company has filed, or every filing across all companies since a given date, and get back a clean list you can write straight to a database.
Above the reference layer sits extracted data. The filings themselves contain XBRL financial statements, and a good SEC filings API parses those into financial statement records you can compare across companies and across time, rather than leaving you to interpret each company's tagging yourself.
Three filing types carry most of the weight for financial analysis.
10-K is the annual report. It contains audited financial statements, a full business description, risk factors, and management's discussion of results. It is the deepest disclosure a company makes each year and the anchor for most fundamental analysis.
10-Q is the quarterly report. It is lighter than a 10-K and unaudited, but it arrives three times a year between annual reports and is what keeps a model current.
8-K is the material events filing. Companies file an 8-K when something happens that shareholders should know about promptly: an acquisition, an executive departure, a bankruptcy, a change in auditor. Earnings releases usually arrive as an 8-K with the results attached as an exhibit, which makes 8-K the fastest structured signal that a company has reported.
Across all three, the fields that matter most are:
Beyond the filings themselves, XBRL financial statement notes are separately searchable. Those notes hold accounting policies, segment breakdowns, and contingencies, which is the detail that never appears in a summary financial statement.
The path from a ticker to a specific filing's financials runs in a few steps.
Start with the filings endpoint. Filter by company and report type to get the filings you care about:
GET /filings?company=AAPL&report_type=10-K
Add start_date and end_date to bound the range, or drop the company parameter entirely to get the latest filings across every company, which is the pattern for keeping a local database in sync.
Each filing in the response carries an ID. From there you retrieve the fundamentals associated with that filing, and then request the financial statements for the fundamental you want.
Statements are also addressable directly through a readable lookup format built from the ticker, the statement, the year, and the period, so AAPL-income_statement-2018-Q1 resolves without needing to walk the chain first.
For 8-K monitoring, the pattern is different. Rather than working backward from a company, you poll the filings endpoint by date and filter on report type, which surfaces material events across your whole universe as they are accepted.
An SEC filings API generally gives you two versions of the same filing, and choosing between them is the most consequential decision in your integration.
Raw documents are the filing exactly as submitted, available as text or HTML. This is what you
want if you are running natural language processing over disclosures, building retrieval for a research assistant, tracking how risk factor language changes year over year, or doing anything where the words themselves are the data.
Structured financials come in two forms, and the distinction is easy to miss.
As-reported financials preserve each company's own presentation. The line items, the ordering, and the labels are the company's. This is the right choice when accuracy to the source matters more than comparability, such as reconciliation work or audit trails.
Standardized financials map every company onto a consistent chart of accounts. The same concept lands in the same field regardless of how the company tagged it. This is what makes cross-company screening possible. Without standardization, comparing revenue across five hundred companies means reconciling five hundred different tagging conventions, and that reconciliation is the work that quietly consumes engineering teams that decide to parse EDGAR themselves.
Most production systems use both. Standardized data drives the screens and the models. As-reported data and raw documents back the detail view, so an analyst can always trace a number to the filing it came from.
Fundamental screening. Standardized statements across a wide universe let you rank and filter on any reported metric without normalizing the data yourself.
Earnings monitoring. Polling for 8-K filings tagged with an earnings disclosure gives you a structured trigger the moment results are accepted, which is what drives alerting and surprise calculations.
Valuation models. Ten or more years of consistent annual and quarterly history is the input for discounted cash flow work, multiples analysis, and any model that depends on trend.
Research and AI applications. Filing text and searchable statement notes are the corpus behind retrieval-augmented research tools. Pairing the raw disclosure with the structured financial record lets a model cite a specific filing rather than assert a number.
Compliance and audit. Stable filing identifiers and as-reported statements give you a defensible chain back to the original submission.
Portfolio analytics. Filing-driven fundamentals feed exposure analysis, factor models, and attribution.
Intrinio delivers SEC filing data as part of its US Fundamentals coverage, which is included on every plan: filing reference data, raw filing text and HTML, searchable XBRL statement notes, and financial statements in both as-reported and standardized form, all from one API.
Data is available over REST, WebSocket streaming, bulk download, and the Intrinio MCP server, so the same filing history can back a nightly batch job, a live application, or an AI agent without a second integration.
If you would rather evaluate it than read about it, you can start a free trial and work against real data from day one. Provisioning takes about two minutes.
Start Free Trial or talk to us about Enterprise access if you need custom coverage, higher limits, or a formal SLA.