Voting records and legislative history
Roll calls, bill activity, committee records, and legislative position history.
OppIntell / Public-record intelligence
Structured, source-cited profiles covering voting records, financial disclosures, public statements, and campaign finance across federal, state, and local races. OppIntell structures public-record data into candidate profiles with source paths preserved for review. The result is an evidence chain that a researcher can inspect before citing.
How we source this: before a line leaves the research desk, it carries a source a second person can open. Research methodology and trust and controls.
Profiles are organized for review by researchers who need a defensible record, not a narrative summary.
Roll calls, bill activity, committee records, and legislative position history.
Assets, liabilities, business interests, and disclosure records tied to source documents.
FEC and state filing data for committees, donors, expenditures, and reporting history.
Official statements, on-record positions, public remarks, and archived source links.
Available public court indexes, case metadata, jurisdictional references, and record status.
Organizations, PACs, employers, boards, committees, and other public-record associations.
Vet political figures before partnerships, investments, or board appointments.
Use source-cited public records for cases involving political figures.
Track candidate records across races, offices, jurisdictions, and time.
OppIntell structures public-record data into candidate profiles with source paths preserved for review.
Records are normalized for search, comparison, and export while retaining citation context for verification.
The result is an evidence chain that a researcher can inspect before citing.
The sourcing rules are on how a claim is checked. The candidate-file method is opposition research. Record-level passes cover a roll call, a committee report, a court docket, and a quotation in context.
The structured database on this page and the public hubs use one candidate index. This is how a filing becomes a field, and where to open the file.
The candidate count on this page is the public candidate-research index: candidate profiles across federal, state, and local races, all parties, after deduplication and the index gate. It is not a raw count of every filing row stored in the pipeline, and it is not a count of one party. The bar above states 50 or more structured fields on a profile. Those fields hold the voting record, the financial disclosure, the public statement, the campaign-finance filing, the court-record match, and the affiliation. A reviewer can move from a field back to the public source that supplied it. The evidence chain on this page is the same three steps for every record: a public source is identified and indexed, the record is normalized into a structured profile field, and the citation is retained for review, export, and verification.
Federal records begin with Federal Election Commission filings. That set covers candidate committees, principal campaign committees, independent expenditures, and Super PAC reports. State records begin with Secretary of State candidate filings from 47 states. County registrations come from the four county adapters in production. The records are normalized into one candidate universe. Deduplication uses legal name, jurisdiction, office, and the filing identifiers on the source. Each upstream source carries a reliability score, so a later query can prefer the more authoritative record when two sources disagree. Ingestion runs on a schedule rather than as a one-time copy. Candidate filings change, committees terminate, and address records update. A scrape that is not repeated goes stale within days.
A research claim carries a source URL, a source type, and a citation-validation status. The source type names the record: an FEC filing, a Secretary of State portal, a court record, an official campaign site, or a government press release. Validation checks whether the cited document still exists at the cited URL and still contains the cited language. A claim that fails that check does not become a public post. Where the record supports a direct fact, the published page states the fact and cites the record. Where the record only opens a question, the page uses posture language such as "may", "could", "would examine", and "appears consistent with". That wording is how a reader tells a documented fact from an open thread.
The publish guard is the publication standard. It requires a minimum length, structured data, and the absence of a broken citation or a source whose reliability score has dropped below threshold. A post that clears the guard is published. There is no separate human review of each post. Blog briefings regenerate hourly. Candidate research records refresh daily. When a cited document is moved, removed, or amended, the validation status flips and the affected post is held back until the source is resolved or replaced. The methodology page is the full write-up of that pass, including how official government sites are listed for each state.
Open the hubs when the question is who is running. Race hubs group candidate research by office. State hubs group the same public index by state, across federal, statewide, legislative, judicial, and local races. Choosing a state and an office lists the people in that race. A profile is the file for one person. Opposition research is that sourced file: voting history, public statements, campaign-finance filings, financial disclosures, and court records, with a source on each claim.
Four method pages each take one record. A roll call is read from the chamber that recorded it, including the question that was before the body. A committee report starts from the schedule the committee filed. A court docket is matched to one person before it is described as that person's case. A quotation is checked against the page that published it. State and local races are the jurisdiction pass for legislative, county, and municipal contests. Candidate vetting is the checklist a campaign runs on its own candidate.
The methodology page also lists the official government sites a researcher can open for each state: candidate filing lists, campaign-finance disclosure, personal financial disclosures, lobbying registrations, court records, election results, and legislative bills and votes. A category on that list stays empty when the verification pass did not confirm a live official URL. Federal offices that the pass resolved are the same block on every state page.
The certified 2026 nominees file is a separate snapshot. It is not a replacement for the live index. It lists governor, U.S. Senate, and U.S. House nominees copied from official election-office lists already on file. A state whose official list is not staged is marked not yet certified. Use that file when the question is who the official document named. Use the race hubs and the state hubs when the question is the live field. Field briefings are the public posts that have cleared the same publication quality gate.
The same sourced records are what due diligence teams, litigation support, and watchdog organizations open on this page. The use-cases page describes that evidence layer for litigation, corporate due diligence, reputation risk, watchdog research, and partner vetting. The public product still starts with politics. OppIntell does not run private investigations, and it does not publish rumors, anonymous tips, or unsourced social posts as claims. The same profile is built whether the candidate is a Democrat, a Republican, or another party.
Profile access, enterprise access, and a custom data export are requested with the form on this page or on the contact page.
For individual researchers who need profile-level access.
For teams that need exports, integrations, or custom coverage.
Contact OppIntell for profile access, enterprise access, or custom data exports.