About The Read

Systematic stock intelligence. Primary source analysis. Public accountability.

What we do

The Read is an AI-native intelligence platform that maintains live entity-state reads on 40+ stocks. We read what most investors skip: 10-K filing diffs, earnings call linguistics, insider trade clusters, and activist accumulations. All synthesized into quantitative state updates every two hours.

Every direction shift is backed by a 9-factor quantitative model and timestamped. Shifts are scored against real prices at 5, 10, and 20 trading days. Every miss is published alongside the wins. The track record is the product.

The edge

Most financial data products give you raw data or other people's opinions. We give you systematic intelligence: primary source analysis that would take a team of analysts weeks, delivered in seconds.

Primary Source Analysis
10-K filing diffs, earnings call linguistics (hedging, avoidance, confidence shifts), 13D activist detection, 8-K material events. All read at scale from SEC EDGAR.
9-Factor Quantitative Model
Estimate revisions, insider activity, price momentum, analyst revisions, narrative gap, options flow, earnings history, sell-the-news probability, filing risk changes. Scored continuously.
Cross-Entity Intelligence
Supply chain propagation, sector contagion detection, co-movement pattern analysis. When TSM moves, we know which stocks follow and how often.
Public Accountability
Every direction shift timestamped and scored against real prices. Direction change history with 5/10/20-day outcomes. No cherry-picking, no quiet deletions.

How it works

1

Ingest

50+ sources scanned every 2 hours. RSS feeds, SEC filings (10-K, 10-Q, 8-K, 13D/G), earnings transcripts, insider trades, analyst ratings, estimate revisions, options flow. Full-text extraction, not headlines.

2

Analyze

Each entity's position is computed from its full signal environment: 9 quantitative factors, primary source analysis, and cross-entity patterns. Direction shifts are detected automatically when the evidence changes.

3

Score

Direction changes are scored against actual price movement at 5, 10, and 20 trading days after the shift. The system calibrates its own weights based on scored outcomes.

Track record

1,000+
stocks tracked continuously
9
quantitative factors per entity
Every shift scored
wins and misses published

Full details on the track record page: per-entity breakdowns and scored transition outcomes.

What we read that others don't

■
10-K/10-Q filing diffs. We compare this year's SEC filing to last year's, word by word. New risk factors, changed forward-looking language, and removed disclosures are surfaced automatically.
■
Earnings call linguistics. Hedging score, confidence level, question avoidance, prepared vs Q&A tone gap. Quantified signals from how management speaks, not just what they say.
■
Insider trade clusters. Not individual transactions, but patterns. Cluster detection across multiple insiders, officer vs director splits, dollar-weighted sentiment.
■
13D/13G activist positions. Activist investors filing with the SEC before the market knows. We detect accumulation from EDGAR filings in real-time.

See for yourself

Every factor score, every filing diff, every scored direction shift is public.

Browse all stocks → Track record → Earnings calendar →

Source Reliability

We track which news sources produce signals that correctly predict price direction. Sources are ranked by hit rate against our scored transition outcomes.

Source Hit Rate Correct Wrong
benzinga 50% 80 81
cnbc finance 56% 23 18
Google News aapl 67% 2 1
Google News ai infra 56% 10 8
Google News amzn 40% 2 3
Google News asts 58% 14 10
Google News defense 67% 2 1
Google News googl 67% 2 1
Google News market 46% 16 19
Google News meta 50% 2 2
Google News nbis 55% 16 13
Google News nvda 40% 2 3
Google News rklb 67% 6 3
Google News semiconductor 55% 40 33
Google News tsla 67% 2 1
reddit investing 42% 14 19
reddit raceto10m 75% 3 1
reddit stocks 43% 26 34
reddit wsb 50% 3 3
techcrunch ai 58% 7 5
Yahoo Finance 50% 223 225