Data That Moves Before the Market Does

Research-grade data with full history, normalized and ticker-mapped across every dataset. Deep historical granularity for backtesting and model training, production-clean for live systems.

Congressional Trades

US House and Senate member trading activity. Disclosure-based signals from STOCK Act filings.

Corporate Lobbying

Federal lobbying disclosures. Track corporate influence and regulatory exposure.

Government Contracts

Federal contract awards mapped to tickers. Revenue signals from government spending.

Patent Filings

Track granted patents by ticker. A structured signal of innovation and R&D output.

Insider Transactions

Daily SEC Form 4 filings. Track executive purchases, sales, and option exercises.

News Sentiment

AI-powered sentiment scores from financial news. Gauge market mood and momentum.

LinkedIn Metrics

Track employee counts and follower growth. Early indicators of company trajectory.

App Store Ratings

Mobile app performance data from iOS and Android stores. User sentiment signals.

Reddit Mentions

Track ticker mentions across Reddit communities like WallStreetBets and r/stocks.

Price Forecasts

Daily and monthly price forecasts with confidence intervals from time-series models.

Analyst Ratings

Track analyst upgrades, downgrades, and price target changes from major institutions.

Put/Call Ratios

Options market sentiment data. Monitor put/call ratios and options flow signals.

12,000+

US Stocks & ETFs

20

Forecast Markets

12

Datasets

10

Years Avg. History

8

Years Delivering Data

99.9%

API Uptime SLA

Built for Institutional Research

The qualities that make alternative data actually usable: deep history, broad coverage, granular records, and clean, ticker-mapped delivery.

Deep History

An average of 10 years of history across the core datasets — enough to train models and test strategies through different market regimes, not just the latest one.

Full US Universe

Alternative data across 12,000+ US stocks and ETFs — the full listed universe, not just the S&P 500 — with price forecasts extending to 20 global markets.

Granular Detail

Not just daily aggregates: individual insider filings, congressional trades, lobbying disclosures, and contract awards — drill down to the single transaction.

Normalized and Comparable

Consistent, clean tabular formats across every dataset, with normalized scores — so values compare across tickers and over time.

Mapped to Tradable Tickers

News, lobbying filings, contracts, and app data are matched to the right ticker before you ever see them — the messy entity-matching is already done.

Clean and Research-Ready

Filing datasets are collected directly from the official disclosure systems, and every series is screened daily for errors, duplicates, and gaps — minutes from API key to analysis.

Evaluate before you commit. Institutional clients get a 30-day free evaluation with live daily data and a trailing 24 months of history. We’ll scope the trial to your use case.

See How It Works

Five Unlike Sources, One Schema

Scraped PDFs, Form 4 XML, a quarterly registry and a weekly bulk archive, each on its own clock and its own time anchor. Every dataset has its own pipeline, running the same five stages, landing in the same schema.

Sources

Congressional Trading

House Clerk · Senate eFD

PDF and HTML filings, scraped — no bulk API exists

Daily, ~1-day lagFiling date

Insider Transactions

SEC EDGAR

Form 4 XML

Daily, ~1-day lagForm 4 filing date

Corporate Lobbying

Senate LDA registry

Official REST API

Quarterly, ~30 days after quarter-endPosting timestamp

Government Contracts

USAspending.gov

Public API, records upsert by award ID

Weekly collectionCurrent-state panel

Patent Filings

USPTO Open Data

Weekly grant XML plus a deep-history archive

Weekly, grants issue TuesdaysGrant date

Pipelines

Each dataset runs its own pipeline, built for the source it collects from. Every one of them runs the same five stages.

1. 01

   Collect & archive

   First-party from each source. The original filing is kept as the provenance record.

2. 02

   Extract

   Free text, XML and scanned filings are parsed into one structured shape.

3. 03

   Map to a security

   Registrants, awardees and assignees are resolved to a US-listed ticker. SEC and government datasets also carry the issuer CIK.

4. 04

   Normalize

   Amounts to their statutory brackets with the as-filed string preserved, dates to ISO, and anything unreadable delivered flagged rather than dropped.

5. 05

   Dedupe & verify

   Amendments and restatements collapse to one row per real event, checked against the collection archive.

Delivery

• REST APIJSON, date and limit filtering
• Python SDKpandas DataFrames
• MCP serverfor LLM research workflows
• FinBrain Terminalscreeners and ticker pages
• CSV exportstraight from the Terminal

One schema. The same ticker keys throughout.

Straight to Your Pipeline
via REST API & SDKs

One integration for your engineering team. Build trading systems, research platforms, and client-facing applications in days.

Deploy in Hours, Not Months

A single REST API with full historical data and consistent schema across all datasets. One integration covers everything — no vendor patchwork, no missing history.

• Unified schema, all datasets - Consistent endpoint structure eliminates per-vendor integration work
• Python SDK - pip install finbrain-python for rapid prototyping and production use
• High-throughput access - Built for backtesting, live trading, and platform integrations
• Deep data history - Years of granular historical data for backtesting, model training, and research — full depth with Enterprise

Contact UsPython SDK Docs

Python

from finbrain import FinBrainClient

fb = FinBrainClient(api_key="YOUR_API_KEY")

# Get price forecasts
predictions = fb.predictions.ticker("AAPL",
    as_dataframe=True)

# Get insider trading data
insider = fb.insider_transactions.ticker("AAPL",
    as_dataframe=True)

# Get news sentiment
sentiment = fb.sentiments.ticker("NVDA",
    as_dataframe=True)

cURL

# Get price forecasts
curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.finbrain.tech/v2/predictions/daily/AAPL"

# Get insider transactions
curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.finbrain.tech/v2/insider-trading/AAPL"

# Get sentiment data
curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.finbrain.tech/v2/sentiment/NVDA"

JavaScript

const headers = { "Authorization": "Bearer YOUR_API_KEY" };
const BASE = "https://api.finbrain.tech/v2";

const predictions = await fetch(
  `${BASE}/predictions/daily/AAPL`, { headers }
).then(res => res.json());

const insider = await fetch(
  `${BASE}/insider-trading/AAPL`, { headers }
).then(res => res.json());

C++

#include <curl/curl.h>
#include <nlohmann/json.hpp>

json get_predictions(const std::string& symbol,
                     const std::string& api_key) {
    CURL* curl = curl_easy_init();
    std::string response, url =
      "https://api.finbrain.tech/v2/predictions/daily/" + symbol;
    std::string auth = "Authorization: Bearer " + api_key;
    struct curl_slist* hdrs = curl_slist_append(nullptr, auth.c_str());

    curl_easy_setopt(curl, CURLOPT_URL, url.c_str());
    curl_easy_setopt(curl, CURLOPT_HTTPHEADER, hdrs);
    curl_easy_setopt(curl, CURLOPT_WRITEDATA, &response);
    curl_easy_perform(curl);
    curl_slist_free_all(hdrs);
    curl_easy_cleanup(curl);
    return json::parse(response);
}

Rust

use reqwest::blocking::Client;
use reqwest::header::{AUTHORIZATION, HeaderValue};

fn get_predictions(symbol: &str, api_key: &str)
    -> Result<serde_json::Value, reqwest::Error> {
    let url = format!(
        "https://api.finbrain.tech/v2/predictions/daily/{}",
        symbol
    );
    Client::new().get(&url)
        .header(AUTHORIZATION, HeaderValue::from_str(
            &format!("Bearer {}", api_key)).unwrap())
        .send()?.json()
}

From Data to Insights
FinBrain Terminal

Give analysts and portfolio managers the same datasets through a visual interface — no engineering required.

Real-Time Dashboard

Market command center with geopolitical monitor, AI signals, earnings calendar, treasury yields, and activity wire — everything you need in one view.

Geopolitical Intelligence

Interactive globe with global conflict and event data. Defense and OSINT analysis feeds. Futures positioning across all asset classes.

Ticker Deep Dive

Every dataset for any ticker — price forecasts with confidence intervals, sentiment charts, insider transactions, analyst ratings, lobbying filings, and Reddit activity.

Commodities & Energy

Crude oil inventories, natural gas storage, US production data, and futures positioning across energy, metals, and agriculture.

15 Alternative Data Screeners

Filter 12,000+ US tickers by price forecasts, insider buying, congressional trades, lobbying, government contracts, Reddit mentions, and more.

Transaction-Level Analysis

Drill down to individual filings — analyze every insider purchase, congressional trade, lobbying disclosure, and government contract by transaction value, date, and entity.

Crypto Markets

Top coins by market cap, sentiment index, Bitcoin network stats, trending coins, and institutional futures positioning.

Try TerminalTerminal Docs

Plug Alternative Data
Into Your LLMs

Give your team's AI assistants direct access to FinBrain data. Accelerate research workflows across the organization.

AI-Native Data Access

Use the Model Context Protocol (MCP) to connect FinBrain data to Claude, ChatGPT, and custom LLM applications. Multiply analyst productivity.

• Natural language queries - Analysts query predictions, insider trades, and sentiment in plain English
• Claude Desktop integration - Just pip install finbrain-mcp and add your config
• Custom LLM apps - Build internal research copilots and client-facing AI tools
• Current to the last collection run - Each dataset on its own documented cadence, daily for filings and market data

Contact UsMCP Docs

Analyst asks

Any congressional buying in AAPL before April 2024?

the model picks a tool

Tool call

house_trades_by_ticker(symbol="AAPL")

the server returns rows

8 rows returned

date · disclosureDate · politician · owner · transactionType · amount · amountFlag

answered from those rows

Answer

4 purchases to 4 sales across 6 members. Half in spouse or dependent-child accounts. Disclosure lag 0 to 42 days, median 26.

Example exchange. Every figure is computed from the returned rows, not written by the model.

Who FinBrain Is For

Teams across the investment landscape build on the same rows — each through the access that fits how they work. Pick a team to see what they touch.

Quant fundsPlatformsCorporate strategy

Quant funds & systematic traders

Add alternative signals to your research pipeline in days, not months. Full history for backtesting, clean daily delivery for production strategies.

Python SDK

client.house_trades.ticker("AAPL")

A DataFrame, ready to join to your factor set.

	date	disclosureDate	politician	owner	transactionType	amount	lag
0	2024-01-10	2024-02-05	Rohit Khanna	SP	Purchase	$1,001 - $15,000	26d
1	2024-01-10	2024-02-05	Rohit Khanna	DC	Purchase	$1,001 - $15,000	26d
2	2024-01-18	2024-02-15	Josh Gottheimer	JT	Sale (Partial)	$1,001 - $15,000	28d
3	2024-01-19	2024-01-24	Blake Moore	SELF	Sale	$15,001 - $50,000	5d
4	2024-02-29	2024-02-29	Pete Sessions	SP	Purchase	$360review	0d
5	2024-03-13	2024-04-24	Jonathan Jackson	JT	Sale	$15,001 - $50,000	42d
6	2024-04-23	2024-05-06	Rohit Khanna	SP	Purchase	$1,001 - $15,000	13d
7	2024-05-09	2024-06-13	Michael T. McCaul	SP	Sale	$15,001 - $50,000	35d

8 of 275 AAPL rows. SP spouse, DC dependent child, JT joint — rows 0 and 1 are the same trade in two accounts, not a duplicate.

Trading platforms & aggregators

Offer your users 12 new datasets through a single integration. Redistribution and client-facing display rights come with the enterprise agreement.

REST API

GET /v2/congress/house/AAPL

The same field names across all 12 datasets.

{ "trades": [
  {
    "date": "2024-01-10",
    "politician": "Rohit Khanna",
    "owner": "SP",
    "transactionType": "Purchase",
    "amount": "$1,001 - $15,000",
    "amountFlag": null,
    "disclosureDate": "2024-02-05"
  },
  {
    "date": "2024-02-29",
    "politician": "Pete Sessions",
    "owner": "SP",
    "transactionType": "Purchase",
    "amount": "$360",
    "amountFlag": "review",
    "disclosureDate": "2024-02-29"
  },
  // … 273 more AAPL rows
] }

amountFlag ships as "review" where a filed amount could not be read as a statutory bracket. Flagged, never silently dropped.

Corporate intelligence & strategy

Track competitor hiring, lobbying, government contracts, and market sentiment — through the Terminal or your team's AI assistants, no engineering needed.

MCP

ask your assistant

No engineering. The data reaches the analyst directly.

“What did Congress do in AAPL through H1 2024?”

Across 275 disclosed AAPL trades, the eight in this window split 4 purchases to 4 sales across 6 members. Half sat in spouse or dependent-child accounts. Disclosure lag ran from same day to 42 days, median 26.

Example exchange. Every figure is computed from the rows in the other two tabs.

Licensing & Plans

Enterprise data licensing for funds, platforms, and research teams — plus self-serve plans for professionals.

MonthlyAnnualSave 17%

Professional

Visual alternative data platform for traders and researchers.

$199/month

• Full access to the FinBrain Terminal for one user
• 16 data screeners across 12 alternative datasets
• 12,000+ US stocks & ETFs covered
• Per-ticker pages with charts & tables
• Real-time Dashboard & macro indicators
• Geopolitical monitoring & OSINT feeds
• Yield curves, COT positioning & more
• Portfolio tracking & analytics
• CSV export from every screener
• Email support

Get Started

For Teams

Enterprise

For funds, platforms, and teams with custom requirements.

Custom

• 12 ticker-mapped alternative datasets for US equities
• Full REST API, Python SDK and MCP access, with rate limits sized to your pipelines
• Complete history for backtesting, 10 to 20 years depending on the dataset
• Collected first-party from official US government sources, with no aggregator in between
• Filing dates delivered alongside transaction dates, so your backtests stay look-ahead-free
• A free 30-day evaluation on live data before you commit
• Rights to redistribute the data and display it to your own clients
• Rights to train internal models and build AI products on the data
• A dedicated account manager and a contractual SLA
• Due-diligence and compliance documentation for your vendor review
• FinBrain Terminal seats for your analysts, as an add-on

Contact Us

Professional

Visual alternative data platform for traders and researchers.

$1,990/year

• Full access to the FinBrain Terminal for one user
• 16 data screeners across 12 alternative datasets
• 12,000+ US stocks & ETFs covered
• Per-ticker pages with charts & tables
• Real-time Dashboard & macro indicators
• Geopolitical monitoring & OSINT feeds
• Yield curves, COT positioning & more
• Portfolio tracking & analytics
• CSV export from every screener
• Email support

Get Started

For Teams

Enterprise

For funds, platforms, and teams with custom requirements.

Custom

• 12 ticker-mapped alternative datasets for US equities
• Full REST API, Python SDK and MCP access, with rate limits sized to your pipelines
• Complete history for backtesting, 10 to 20 years depending on the dataset
• Collected first-party from official US government sources, with no aggregator in between
• Filing dates delivered alongside transaction dates, so your backtests stay look-ahead-free
• A free 30-day evaluation on live data before you commit
• Rights to redistribute the data and display it to your own clients
• Rights to train internal models and build AI products on the data
• A dedicated account manager and a contractual SLA
• Due-diligence and compliance documentation for your vendor review
• FinBrain Terminal seats for your analysts, as an add-on

Contact Us

Frequently Asked Questions

Common questions about integration, coverage, and enterprise plans.

Getting Started

Questions about API access and setup.

Data & Coverage

What data is available and how often it updates.

Pricing & Support

Subscription plans and support options.

How do I get started?

Sign up at terminal.finbrain.tech for instant access. Most users are in production within a day. Enterprise clients can contact sales for guided onboarding.

What programming languages are supported?

We provide an official Python SDK with pip install finbrain-python. For other languages, use our REST API directly with any HTTP client.

What are the rate limits?

API access is part of an Enterprise agreement, with rate limits sized to your team's throughput requirements. Evaluation trials run at 2,000 requests per hour.

Can we redistribute or display the data to clients?

Redistribution and client-facing display rights are available with our Enterprise plan. Contact sales to discuss licensing for your use case.

Can we evaluate the data before subscribing?

Yes. Institutional clients get a 30-day free evaluation with live daily data and a trailing 24 months of history across the full universe. Book a call and we'll scope the trial to your use case and onboard your team.

What markets are covered?

Our alternative datasets focus on US-listed assets — 12,000+ stocks and ETFs across NYSE and NASDAQ. Price forecasts additionally cover 20 global markets, including international equities, forex, crypto, and commodities, and news sentiment extends to selected international markets.

How often is the data updated?

Price forecasts and sentiment scores update daily before US market open. Insider transactions and congressional trades are collected daily as new filings are published, so a filed transaction is typically available within a business day. Individual dataset pages document exact cadences.

How far back does historical data go?

Historical depth varies by dataset. Our longest-running datasets carry 10+ years of history, corporate lobbying goes back 15+ years, and newer series are being backfilled toward similar depth. Full historical depth is delivered under an Enterprise agreement. The Professional plan is the FinBrain Terminal browser platform, which shows recent activity per ticker rather than full history. Price forecasts are delivered point-in-time and represent forward-looking predictions only — historical forecasts are not exposed via the API.

How should we evaluate the price forecasts?

Our forecasts are generated using ARIMA time-series models that capture the statistical properties of historical price movements and produce out-of-sample predictions with calibrated confidence intervals. Forecasts are delivered point-in-time — to evaluate them, capture the daily output and compare against realized prices over time. See the Price Forecasts dataset page for methodology details.

Is the data suitable for backtesting?

Yes — the disclosure datasets are built for it. Filings carry both the event date and the public disclosure date, so you can anchor a backtest on the moment information actually became public and keep it free of look-ahead bias. Core datasets carry 10+ years of history, up to 15 for corporate lobbying. Each dataset page documents the correct point-in-time anchor to use.

Does the data contain inside information or personal data?

No. All datasets are built entirely from publicly available information. They contain no material non-public information and no non-public personal data — the only names that appear are those in mandatory public disclosures — which keeps vendor due diligence and compliance review simple.

What plans are available for teams?

Teams and institutions are served by the Enterprise plan: custom pricing with SLAs, dedicated support, custom rate limits, and redistribution licensing. Programmatic access, meaning the REST API, Python SDK and MCP server, is part of an Enterprise agreement. Individual professionals can self-serve with the Professional plan at $199/month for the FinBrain Terminal.

What support is available?

The Professional plan includes email support. Enterprise clients get a dedicated account manager, prioritized response, and SLA-backed response times.

What does the Enterprise plan include?

Data redistribution rights, client-facing display rights, AI/LLM training rights, dedicated account management, SLA guarantees, on-premise deployment options, and white-label solutions. Enterprise plans are tailored to your team's specific requirements.

Do you offer SLAs?

Enterprise agreements include a contractual SLA covering API availability with a 99.9% monthly target and prioritized support response, backed by service credits. Professional plans receive best-effort support.

Can we start with a professional plan and upgrade later?

Yes. Institutional teams evaluate through a scoped 30-day trial with live data and a trailing 24 months of history, then move to a full Enterprise agreement. Your integrations carry over unchanged.