Company Intelligence

The whole company.
Structured, from the filings.

Most tools search documents. AnalystBook builds a record: every SEC filing a company makes is parsed into structured facts — segments, financials, forensic scores, pay, insiders, subsidiaries — each one linked to the filing paragraph it came from. Readable by you. Reasoned over by AI.

14-day free trial — no credit card required. Cancel anytime.

The dossier · one company, structuredEvery figure → its filing paragraph
SectionWhat's extractedSource
BusinessSegments · geography · customers ≥10% · competitors · subsidiaries10-K · Exhibit 21
FinancialsStatements in the industry's own template · debt maturity wallXBRL · 10-K/10-Q
Forensic signalsAltman Z (sector-calibrated) · Beneish M · earnings qualityComputed from filings
GovernanceExecutive pay vs peers · board · CEO history · exec changesDEF 14A · 8-K 5.02
OwnershipInsider clusters · beneficial owners · activist positionsForm 4 · 13D/13G

30+ sections per company, 7,600+ companies — each fact filed into its own place the moment the filing lands. No AI model decides the numbers.

The dossier's structure: five dimensions, 30+ sections, every extraction traceable to its official source.

A record, not a search index.

01

Read by rules, not by AI. Never generated.

A dedicated extractor handles each disclosure — the machine-readable financials (XBRL), proxy pay tables, Exhibit 21 subsidiaries, Form 4s, 13D/G stakes, comment letters. The same filing always yields the same figures.

  • 25+ extractors
  • Organized data, not loose text
  • Same filing, same answer
02

Read the way the industry reports.

A bank's revenue is net interest, an insurer's is premiums, a REIT reports FFO. Eight industry templates structure each company on its own terms instead of forcing one shape onto every business.

  • Banks · net interest
  • Pharma · drug segments
  • SaaS · recurring revenue
  • REITs · FFO
03

Forensic signals on every filing.

Sector-calibrated Altman Z-Score, Beneish M-Score and earnings quality are recomputed as filings land — plus auditor and going-concern flags, SEC comment letters, and insider-cluster detection.

  • Z / Z' / Z'' by sector
  • M-Score
  • Going-concern flags
  • Comment letters

FAQ

Common questions.

Where does the data come from?

Exclusively from official filings on SEC EDGAR — 10-K, 10-Q, 8-K, DEF 14A, Forms 3/4/5, 13D/13G, 20-F for foreign filers, and more. No third-party data vendors, no scraped estimates, no paid feeds. That's why every figure can link back to its source paragraph.

Does AI generate the numbers?

No. Every number, score and change flag is computed from the filing by fixed rules — the same filing always yields the same figures. AI sits on top of the record (Ana and the Intelligence Brief), grounded in it and checked against it, but it never decides the data.

How current is the record?

AnalystBook watches EDGAR continuously. When a company files, the affected sections of its record are re-extracted and its forensic scores recomputed — and if you follow the company, the change shows up in your feed the day it happens.

Why does structured data matter for AI research?

Chat tools retrieve text passages and hope the model reads them right. AnalystBook's AI reasons over organized, industry-aware facts — segment tables, period-matched financials, computed scores — which is what lets it cite exactly, compare correctly across companies, and decline what the record doesn't support.

One workspace

It works with the rest.

Start with one company.
Research it from the source.

14 days free, no credit card required. Cancel any time before the trial ends.