3spread | Fundamental data built for traders

Any workflow. One API.

From deep historical backfill to live signals, one consistent API runs the whole research-to-production stack.

Point-in-time history, restatement-aware

500K+ filers, 30+ filing types, back to 2012

Train on what was actually known at each point in time. Amendments and restatements are tracked, so as-reported figures stay separate from later revisions and your models never learn from data that didn't exist yet, across every filer, back to 2012.

GET /v1/coverage

{
  "filers":       500000,
  "filing_types":     30,
  "earliest":  "2012-01-01",
  "latest":         "live"
}

Cross-section the whole universe

Same line item, same label, every filer

Filers tag the same data with different XBRL elements, contexts, and labels, domestic and foreign alike. 3spread maps them to one schema, so screens, factor work, and cross-sectional regressions run against consistent panels instead of a per-filer cleanup project.

Raw XBRL

<us-gaap:Revenues
  contextRef="P10-1Q4_ABCD"
  unitRef="USD" decimals="-6">
  1000000000
</us-gaap:Revenues>

<us-gaap:RevenueFromContract
  WithCustomerExcludingAssessedTax
  contextRef="P10-1Q4_EFGH"
  unitRef="USD" decimals="-6">
  500000000
</us-gaap:RevenueFromContract...>

3spread, standardized

[
  { "ticker": "ABCD", "revenue": 1000000000 },
  { "ticker": "EFGH", "revenue":  500000000 }
]

NLP-ready, section-labeled

Every section, separately addressable

Filings arrive split into their labeled sections (risk factors, MD&A, legal proceedings, business description, and every other defined section of the document), each a discrete field. Your NLP pipeline pulls exactly the sections it needs instead of scraping the whole filing.

GET /v1/companies/ABCD/10-K

{
  "filing": "10-K",
  "ticker": "ABCD",
  "sections": {
    "mdna":           "...",
    "risk_factors":   "...",
    "legal":          "...",
    "business":       "...",
    "...":            "..."
  }
}

Live, seconds after EDGAR

Production signals from new filings

No overnight jobs, no review queue. The moment the SEC accepts a filing, 3spread parses it and serves it through the same API within seconds, ready to drive an event-driven strategy.

event log

// New 8-K arrives
> 2024-01-01T00:00:00Z  ABCD  8-K filed
> 2024-01-01T00:00:07Z  parsed, standardized
> 2024-01-01T00:00:07Z  available via API

People, companies, funds: linked

Follow the connections across filers

The same person or fund appears across filing after filing, identified differently each time. 3spread resolves them into a single entity and links people, companies, and funds together, so ownership chains and insider networks are queryable, not a manual cross-reference.

GET /v1/people/0000123

{
  "name": "Jane Doe",
  "roles": [
    { "entity": "ABCD",         "form": "4",       "title": "CFO" },
    { "entity": "EFGH",         "form": "DEF 14A", "title": "Director" },
    { "entity": "Acme Capital", "form": "ADV",     "title": "Adviser" }
  ]
}

500K+

Filers tracked

From companies to the people behind them

30+

Filing types

Across companies, funds, and people

2012

History back to

Retroactively standardized

From Filing to Insight in Three Steps

An Entity Files with the SEC

Annual reports, quarterly earnings, insider transactions, institutional holdings, and more. Over 30 filing types across 500K+ tracked filers: companies, funds, advisors, and the people behind them.

3spread Processes and Standardizes

We parse each filing, extract the data and text, and map everything to a consistent schema, so the same data point gets the same label regardless of which entity reported it.

You Get Clean, Comparable Data

Structured JSON delivered through a standard REST API. Ready for your models, dashboards, compliance tools, or research. No cleaning or transformation required.

Designed to Perform

Cross-Filer Comparisons, No Cleanup Required

Different filers use different labels and formatting for the same data. 3spread transforms them to consistent specialized schemas so you can run comparisons and benchmarks without manual cleanup.

Built for Time Series, NLP, and Statistical Modeling

Panel data ready for longitudinal analysis: consistent periodization across filers, restatement-aware. Textual sections pre-segmented for NLP pipelines: MD&A, risk factors, CD&A, and others surfaced as discrete fields. Numeric schemas normalized for cross-sectional regressions and factor research without per-filer cleanup.

500K+ Filers Back to 2012

500,000+ filers, from companies, funds, and advisors to the insiders and individuals behind them, across 30+ filing types: 10-Ks, 10-Qs, 8-Ks, proxy statements, insider filings, and more. Retroactive standardization means the same schema applies whether you're looking at last quarter or ten years ago.

A Clean API That Gets Out of Your Way

RESTful endpoints, JSON responses, straightforward authentication. Comprehensive documentation with request examples so you can start pulling data in minutes.

Purpose Built

Quantitative Research

Pull 10 years of standardized income statements across thousands of entities in a single API call. Build factor models on data you can trust without spending weeks on cleanup.

Compliance & Monitoring

Track insider trading disclosures, ownership changes, and material events across your entire coverage universe: structured and searchable, not buried in paperwork.

Fintech & Product Development

Build financial analysis tools, screening platforms, or research products on top of reliable, standardized data. Enterprise licensing available for redistribution.

Get Started

Choose the path that fits your needs.

Community

Full API access and every dataset, for personal research and analysis.

Startup

3spread works with qualifying early-stage teams to help them get off the ground.

Apply Now

Business

3spread data at scale for your organization.

Contact Sales

Fundamental Data built for traders