U.S. EDGAR and global corporate disclosures parsed into clean, multi-level JSON and read through the same sentiment engine our quant clients trade on. History from 2006 to today.
2006History starts, for backtesting and validation
2 layersU.S. EDGAR and global, each with its own taxonomy
24/7Parsed as filings publish
Why it matters
Most vendors deliver documents. We deliver processed, risk-scored content across global disclosures, so oversight stops depending on who had time to read.
Coverage
Two layers, one pipeline.
MRFMachine Readable Filings · U.S. SEC EDGAR[CONFIRM: "In partnership with S&P Global Market Intelligence" for public use]
01 / StripNoise removedPage numbers, images, watermarks and HTML tags taken out. Scanned documents are OCR'd first.
02 / StructureMulti-level JSONRebuilt by the document's own heading hierarchy, section by section.
03 / ScoreRisk language surfacedOur sentiment engine scans for risk language and disclosures. Native-language filings run the same pipeline.
04 / LinkTied to SNL institutionIdDirect integration with Capital IQ and Xpressfeed, with no separate entity resolution.
How teams use it
One output, two desks.
Risk managersEvery holding, every disclosure.Systematic filings monitoring as a longer-term signal, with risk-relevant content surfaced as companies publish.
Portfolio managersThesis-relevant flags.The same output, used to surface disclosures and risk flags on the names in the book.
Custom projectsYour documents, our engine.Custom sentiment engine projects on document repositories from other exchanges, scoped with a CA analyst.
FAQ
Frequently Asked Questions
Most vendors deliver documents. We deliver processed, risk-scored content across global disclosures, reconciled with the sentiment signal you may already be evaluating.
8 core SEC filing types — 10-K, 10-Q, 8-K, 20-F, 6-K, 40-F, S-1 and DEF 14A — from 2006 through present day.
Non-U.S. corporate disclosures under a separate taxonomy: Annual Reports, Annual Reports to Shareholders, Financial Supplements, Interim/Semi-Annual Reports, Metals and Mining Annual Reports, Quarterly Reports, Japanese earnings releases, and an other-financials catch-all for documents not yet mapped to an existing type, from 2006 through present day.
Both the EDGAR and Global layers cover filings and documents from 2006 through the present, over 20 years of history for backtesting and strategy validation.
Each filing is stripped of extraneous elements (page numbers, images, watermarks, HTML tags), then rebuilt into clean, multi-level JSON structured by the document's own heading hierarchy. Scanned documents are OCR'd first, and every filing is tied to the SNL institutionId for direct integration with Capital IQ and Xpressfeed. Native-language documents run through the same pipeline as English.
Filings data is delivered via RESTful API in JSON and Snowflake, with real-time alert delivery for high-impact events.
Common uses include compliance monitoring, continuous portfolio risk tracking, and due diligence research across both domestic and international holdings.
Stop reading line by line.
See your holdings' filings parsed and risk-scored, with history back to 2006.