Methodology
What RoaringSEO measures, where evidence comes from, and how limitations are reported.
Evidence sources
Findings come from direct HTTP crawls, isolated rendered pages, user uploads, server logs, explicit user input, or deterministic derivations. Each released rule is versioned.
Scoring
Site health counts each related rule family once per affected page. Coverage and confidence are disclosed. Machine-readability rubric v2 publishes separate crawl-access, semantic-structure, structured-data, content-clarity, provenance/contact, canonical-consistency, and machine-delivery components; it is not a prediction of external-engine inclusion.
Unavailable intelligence
Search index state, global backlink coverage, rankings, traffic, and advertising approval are unknown unless users import their own evidence. RoaringSEO does not scrape results or manufacture proxy metrics.