Our data

What we have, where it comes from, and what's missing

Tirion builds its own datasets from primary sources — county assessors, the SEC, the IRS, the FEC, state corporate registries. We know exactly where every data point comes from because we collected it ourselves. And when we don't know something, we say so.

The dataset

What we cover

Every record links back to its original source. Every confidence level is visible. Provenance is part of the culture.

Real estate & property
100M+ parcels
Ownership, assessed values, transaction history, and property characteristics from county assessor records across all 50 states.
Sources: County assessors, recorder offices
Update: Quarterly to annually by county
SEC & corporate filings
EDGAR full index
Officer and director roles, executive compensation, insider transactions, and beneficial ownership from public SEC filings.
Sources: SEC EDGAR
Update: As filed
Political donations
FEC bulk data
Federal campaign contributions with employer, occupation, and amount. Linked to individuals via entity resolution.
Sources: FEC
Update: Quarterly
Nonprofit & foundation data
IRS 990 filings
Board membership, officer compensation, grant history, and revenue for tax-exempt organizations.
Sources: IRS 990, 990-PF, 990-EZ
Update: As filed
Corporate officers & directors
State SOS filings
Company affiliations, registered agents, and officer/director roles from Secretary of State business registrations.
Sources: State SOS offices
Update: Varies by state
News & public records
28+ sources
Named entity extraction from regional and national news sources. Business journals, press releases, and public announcements.
Sources: Regional business journals, news wires
Update: Daily
Why it matters

Why we collect data directly

Direct from the source
We collect from county assessors, the SEC, the IRS, and the FEC directly. Going to primary sources means we control the pipeline end to end and can keep costs accessible for teams of every size.
Full provenance chain
Every record in Tirion carries a source URL, collection date, and processing lineage. You can trace any data point back to its origin and judge its quality for yourself.
We fix what we find
When we discover data quality issues — inconsistent address formats, missing fields, broken records — we fix them in our pipeline. Data quality is a continuous process, not a one-time import.
Freshness on our terms
We re-collect sources on our own schedule. When a county updates its records, we pick it up in our next collection window. Every record carries a "last verified" timestamp.
Honest gaps

What we don't have

Research is about decreasing uncertainty — not pretending it doesn't exist. Here's where our uncertainty is highest.

California property owner names
California does not publish property owner names in bulk assessment data. We have parcel records, assessed values, and transaction history — but owner names must be resolved through recorded deeds, which we are adding county by county. If your prospect is in California, property ownership may be incomplete.
Bank accounts and brokerage balances
Liquid asset data isn't part of the public record. We show what is: real estate equity, SEC-disclosed holdings, and compensation from 990 filings. Each carries a confidence tier so you know exactly how much weight to give it.
Real-time data
County records update on their own schedules — some monthly, some annually. SEC filings appear when companies file them. We don't pretend data is fresher than it is. Every record carries a "last verified" timestamp so you know exactly what you're looking at.
Complete geographic coverage
Property data quality varies by state and county. Some counties have excellent digital records; others are still working from scanned PDFs. We are transparent about where coverage is strong and where it's thin. Our coverage map shows exactly what's available in any area.
Giving history to your specific organization
We surface public giving data — IRS 990 contributions, political donations, foundation grants. But we can't see gifts to your organization unless they're in a public record. Your internal CRM data is yours. Tirion enriches it; it doesn't replace it.
Confidence, not scores

How we rate what we know

Research is messy. Sometimes you find a primary source. Sometimes the best you have is an AI-assisted match. Both are useful — but they're not the same. Every data point in Tirion carries a confidence tier so you always know what you're working with.

Tier Meaning Example
Primary Direct from the authoritative source SEC filing, county deed
Extracted Parsed from a structured public document 990 compensation, FEC occupation
Inferred Derived from multiple corroborating records Equity from assessed value + sale price
AI-assisted Identified by automated entity resolution Name match across two datasets
Unverified Present but not yet corroborated Single-source news mention

You always know why we think what we think. Click any data point in the app and see the source.

See it yourself

Prospect research is only as good as the data behind it. If you want a platform that shows its work and tells you when it doesn't know something — that's what we're building.

Get early access