Open government data, monitored
Your product depends on open datasets that change, move, break, and get relicensed without notice. GovDataHub watches them daily and tells you within 24 hours — with the license basis for every dataset you ship, traced to its source record.
Curated, license-verified listings in CSV and JSON, harvested from data.gov, data.gov.uk, Socrata, Zenodo, Hugging Face, and national open-data portals. Only allow-listed licenses are included — US public domain, public domain, CC0, CC BY, OGL, OGL 3.0. Everything else is quarantined and labelled, never silently cleared. The catalog is free and stays free; it is the proof of the monitoring work, not the product.
| Collection | Datasets | Formats |
|---|---|---|
| Climate & energy | 704 | CSV · JSON |
| Economy & small business | 807 | CSV · JSON |
| General | 1,402 | CSV · JSON |
github.com/GovDataHub/open-data-index · Zenodo archive (DOI 10.5281/zenodo.23308617)
Send us up to 20 datasets your product depends on. Within 5 days you receive a written report on each one: current license and the basis for it, freshness against its stated cadence, and whether the source links still resolve. No cost, no obligation — it is the check we run on our own catalog, applied to yours.
Include the dataset names or URLs in your email. If the report shows exposure, we will say so precisely; if it does not, you keep the report and owe nothing.
Continuous monitoring of the datasets your team ships on, run as a concierge service during the pilot phase. Daily checks against a timestamped change archive that has been accumulating since October 2026.
Pilot customers convert at the launch price only after pilot precision figures are published. A public launch on unmeasured precision is not something we will do.
Quoted before work starts. A data-provenance audit — not legal advice.
Zero license false positives is our operating standard, and the corrections log is how we are held to it. Errors are corrected fast, in public, with the cause stated. Methodology principles are published; the log is part of the product.
October 2026 — license mislabel. 205 datasets from Canadian provincial portals were published labeled as UK OGL 3.0. A generic text-matching rule in our license normalizer treated any "open government licence" wording as the UK license, without checking jurisdiction.
The error was caught by our own verification, every affected record was re-checked against its source and relabeled, and the rule was replaced with jurisdiction-specific matching. The full entry, including scope and current state, is in the public corrections log maintained with the catalog on GitHub.
The free weekly digest reports what changed across the monitored catalogs — new, updated, and broken datasets — every Monday. Pro adds analyst notes on top of the digest for individuals: $12/month, $120/year, $240 founding.