Method / 01

One record. Every source attached.

Orbcrawl is built around a simple constraint: a conclusion is only useful when the source and date survive with it.

Ingest

We read public registers such as the Food Standards Agency, OpenStreetMap and Companies House. Every source row is stored before resolution.

Resolve

Names, postcodes and provenance are used to collapse records into businesses without pretending weak matches are certain.

Inspect

Domains are discovered and crawled for deterministic signals: HTTPS, mobile readiness, stale technology, dead hosts and other observable facts.

Refuse or rank

Site decay stays separate from commercial fit. A modern site is refused; a blocked crawl gets no verdict; a genuine problem may be ranked.

Two scores, deliberately

“Bad website” and “worth contacting” are different questions.

A dead site may belong to a closed business. A dated site may still do its job. Orbcrawl keeps technical decay separate from trading liveness and reachability so one score cannot disguise the other.

Blocked does not mean broken. Unknown does not mean absent. Those distinctions are enforced in the system rather than added as caveats later.

Known limits

What the current system does not claim.

Coverage is seven districts, not the whole UK.
Rank weights are judgement, not calibration against customer outcomes.
No automated dispatch or reply loop exists.
Longitudinal change detection is the next release, not a current feature.