▸ free proof · no signup · cc0
See the leak before installing anything.
Join the same SEC fundamentals by fiscal period and by the date they actually became public. The difference is information your backtest never earned.
3,280 rows · 40 tickers · 7 concepts · methodology public
naive period-end join makes the value visible here
Investors still could not read Apple's FY2024 10-K.
Two clocks, one value. Point-in-time research follows the public filing clock.
▸ choose your proof
Start with the artifact, not the pitch.
Every route below uses the same public 40-ticker sample. Nothing is uploaded or gated.
Run the same fundamentals join two ways and reproduce the 47/413 result in about three minutes. No account, API key, or local install.
Run the notebook ↗▸ honest scope
What this proof does not claim.
- Annual coverage. Quarterly 10-Q history is not shipped.
- No delisted companies. This is not survivorship-bias-free.
- Seven concepts. Not the complete SEC XBRL concept space.
- Forty unreliable dates. Excluded from filing-gap claims.
▸ shape the roadmap
What coverage would unblock your research?
Pick the missing piece that matters. You'll get the free sample link and one email when that coverage changes.
▸ after the proof
Query the broader annual dataset for $49/month.
5,216 companies and 314,436 point-in-time rows, with server-side as-of semantics.
View coverage and pricing