The skeptical analyst's read
A transparent, descriptive read on a company's fundamentals, drawn straight from its filings and looked at through three independent lenses.
The Filter Lab
Three lenses
Three independent lenses, each anchored to the company's own words.
How the company's structure is changing over time.
The financial signals analysts already trust.
Where the reported numbers deserve a closer look.
From thousands of companies to the few worth understanding.
The workflow
What you actually get
Everything that matters about a company, organized into a single comparable research view.
Anchored to the company's own words. Verified directly from EDGAR, with no paraphrase and nothing invented.
Structured intelligence, ready before it ever reaches ChatGPT, Claude, or your own workflow. Every figure sourced, nothing to rebuild.
A skeptical, descriptive narrative of what the numbers show. What's worth questioning, never what to do.
Original research
Across 1,208,117 company-months of US listings from 1998 to 2024, including three crisis periods, companies with the weakest balance sheets went bankrupt within two years 5.467% of the time against 0.570% for the strongest — nine and a half times more often, and six times once company size is held fixed, with confidence intervals that do not overlap either way. But the effect only exists once the share price is already moving. In the calmest companies the balance sheet carries no bankruptcy information at all.
August 18, 2026
We tested twelve measures of company quality against how much a stock moves, on 2,662 US-listed companies. Ten looked like they worked. Then we ran the same test inside market-capitalisation deciles and every one of them moved toward no effect. This measures the range of companies' movement, not their direction.
August 13, 2026
Across 1,279,541 company-months of US listings, companies with both a turbulent share price and a weak balance sheet typically lost 31.0% over the following two years. The same group ran to five times its value more often than any other group in the study. This measures the range of companies' movement, not their direction.
August 11, 2026
For data publishers
If your company publishes fundamentals, whether as a data feed, an API or a screen your users read, the same method is available as a paid engagement. Send a sample of your published output and get back a written report naming each disagreement with the source filing, the arithmetic behind it, and the field-level pattern that produced it. Private diligence, delivered to you, never published.
See it
Illustrative preview. Open a live analysis →
No black box. Every number traces back to the public filing it came from.
The weekly Filter, plus the daily market read. Descriptive only — no calls, no targets, no price predictions. One click to leave, any time.
No spam, and the list is never sold or shared. Unsubscribe from any issue.