Methodology

How data is collected

Data is collected from official public record sources such as OSHA, state license boards, and Secretary of State registries.

How data is normalized

Company names are normalized (case/punctuation/legal suffix handling) to support soft matching across datasets.

How matching works

Matching is based on normalized company names and state-level context. Results are aggregated conservatively to avoid false precision.

What is not included

No subjective ratings, legal opinions, or private personal data are provided.

Update frequency and limitations

Refresh intervals vary by source and may introduce delays. Verify critical details directly with the source agency.

Editorial and update accountability

Each dataset has an assigned refresh cadence and validation checks before publication.

When source systems change schema or publication format, ingestion rules are reviewed and adjusted before the next release.

Quality control and correction handling

Quality checks include duplicate control, date validity checks, null-value handling, and normalization consistency.

Reported issues are triaged, source-verified, and resolved through a tracked correction process with data re-run when required.

Related trust documents: Sources · Editorial Policy