Public markets, read as they happened.
XS1 Intelligence builds models that read filings, disclosures and news about listed companies, detect the events in them, and measure how disclosure changes, all on point-in-time data. The output is research signals and decision support for professional teams, not investment advice.
Point-in-time filing tape
What was known, and when.
A fictional issuer’s quarter, replayed. Filings become structured events the day they are public, and a change in disclosure language is measured against the issuer’s own history rather than a market-wide average.
Drag across the tape, or use the arrow keys
- Filing event
- News event
- Issuer baseline
- As known at
Point-in-time: the day 9 filing stays as it was known on day 9, even after the amendment, so nothing learned later leaks backwards. Research signals, not investment advice.
As known at
Events are extracted and filed into lanes as they become public. Wording changes so far sit inside this issuer's usual range.
Latest events
What it covers
Open-source intelligence for public markets.
The same discipline as every other area of the division, applied to issuers, sectors and the events that move them. These are the parts XS1 Intelligence builds models for.
Filings analysis
Machine reading of regulatory disclosures: current reports such as 8-K items, annual and quarterly reports, ownership filings (Form 4, 13D/G, 13F) and their equivalents in other jurisdictions' registries.
- 8-K
- 10-K / 10-Q
- Ownership
Disclosure language change
Tone and change-in-wording analysis across risk factors, management discussion and earnings-call transcripts, compared with each issuer's own past language.
- Risk factors
- Transcripts
Event detection
Discrete corporate events found in filings and news (leadership changes, agreements, ownership moves, outages, guidance changes) and structured as timestamped records.
- Timestamped
- Deduplicated
Alternative data and nowcasting
Models over non-traditional public and licensed datasets, such as job postings, web activity and commercial imagery, to estimate current conditions before official figures arrive.
- Provenance reviewed
Supply-chain and exposure mapping
Issuers linked to suppliers, customers, facilities and jurisdictions, so a disruption or a sanctions action can be traced to the companies it touches.
- Suppliers
- Facilities
Manipulation-pattern analytics
Market-abuse pattern detection for compliance teams: models that help find patterns such as spoofing, wash trading and pump-and-dump schemes.
- For compliance teams
Guardrails
Research signals, built to be defensible.
Market work is only useful if its inputs can be explained to a compliance team. These controls apply to every markets engagement.
- 01
Point-in-time data
Every record keeps what was known and when, revisions included, so models and backtests never see information from after the moment they describe.
- 02
Documented provenance
Each dataset's source, terms of use and collection method are reviewed and recorded before it is used.
- 03
No material non-public information
We do not seek, accept or use material non-public information, and datasets are checked for it before use.
- 04
Decision support, not advice
Outputs are research signals for professional teams. We do not make investment recommendations or predict prices.
Markets intelligence
A research question about public markets?
Tell us what you want to measure and how the signal will be used. Every request is reviewed before work is scoped.