3spread

Fundamental datasets

Twelve data families, extracted from the SEC and standardized into consistent schemas regardless of who filed or when. Amendments linked, restatements preserved, and every row validated.

Insider transactions

Who bought, who sold, and when they told the SEC.

GET /v1/insiders

·Screen for clusters of insider buying before earnings

6 endpointsForms 3, 4, 5
Ownership & insiders

Institutional holdings

What every institutional manager owned, quarter by quarter.

GET /v1/institutional-holdings

·Diff a fund's portfolio quarter over quarter to see what they bought and sold

3 endpointsForm 13F
Ownership & insiders

Beneficial ownership

The 5 percent stakes, and whether the holder means to act on them.

GET /v1/beneficial-ownership

·Catch activist stake-building as it is disclosed, by watching 13D intent

2 endpointsSchedule 13D / 13G
Ownership & insiders

Proposed sales

Insiders signalling a sale before it settles.

GET /v1/proposed-sales

·See intended insider selling ahead of the Form 4 that confirms it

2 endpointsForm 144
Ownership & insiders

Fund portfolios

What mutual funds and ETFs actually hold, month by month.

GET /v1/fund-portfolios

·Look through an ETF or mutual fund to its real underlying positions

3 endpointsForm N-PORT
Funds

Fund census

The annual census: series, share classes, service providers.

GET /v1/fund-census

·Resolve a fund complex's full series and share-class structure

2 endpointsForm N-CEN
Funds

Money market funds

Money-market holdings, NAV, flows and liquidity.

GET /v1/money-market-funds

·Watch liquidity and flows across the money-market complex

7 endpointsForm N-MFP2
Funds

Proxy votes

How funds actually voted, vote by vote.

GET /v1/proxy-votes

·Check whether a fund's stewardship claims match how it voted

2 endpointsForm N-PX
Funds

Private offerings

The public record of private fundraising.

GET /v1/private-offerings

·Track private fundraising in a sector before it reaches the press

2 endpointsForm D
Offerings & registrations

Reg A+ offerings

Small-cap and crowdfunded raises, from the primary filings.

GET /v1/reg-a-offerings

·Track small-cap and crowdfunded raises from the primary filings

2 endpointsForms 1-A / 1-K / 1-U / 1-Z
Offerings & registrations

Registration statements

Prospectus text, split into labeled sections.

GET /v1/registration-statements

·Pull the risk factors of every IPO in a sector and diff them

4 endpointsS-1, S-3, S-4, S-11, F-1, F-3
Offerings & registrations

The 3spread extraction process

Textual Data Extraction

Many SEC filings contain labeled text sections: risk factors, MD&A, legal proceedings, business descriptions. Different entities name these sections differently. 3spread maps them all to consistent standard labels, with narrative text and embedded tables captured separately.

Consistent Labels Across Entities

Things like “MD&A,” “Management Discussion,” and “Item 7” all get mapped to a single standard label, regardless of how the entity titles the section.

Text and Tables Separated

Narrative content and embedded data tables are captured independently, so you can analyze each without untangling the other.

Full Section Coverage

All labeled sections are extracted: not just financials, but risk factors, legal proceedings, exhibits, and more.

Structured Data Extraction

Financial statements, ownership tables, compensation data, and other structured elements are extracted and delivered as clean JSON, with standardized schemas that make cross-entity comparisons work out of the box.

Standardized Schemas

Every filing type maps to a standardized schema built around regulatory and GAAP standards. Different industries have fundamentally different financial structures, so they get different schemas, but each one is consistent across every entity that files under that framework.

Values Preserved, Schemas Standardized

Reported values are preserved but spread into compliant schemas, so line items land in consistent categories across entities. The reshaping is what makes cross-entity comparison work out of the box.

Ready for Analysis

Normalized data types, consistent units, and clean JSON structure mean you can load responses directly into your pipeline without a transformation layer.

Get Started

Choose the path that fits your needs.

Read the docs

Every endpoint, parameter, and response schema.

Get started

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

See a guide

Runnable recipes against each dataset.