Methodology

How it was produced

By auditing a production codebase line by line: roughly 2,100 indexable URLs, 42 schema.org types in use, and a programmatic tier generating 1,689 of those URLs. The audit examined what was implemented, what was not, and what the gap cost.

It was not produced by reading blog posts and summarising them, which is the ordinary method and the reason most SEO writing is unfalsifiable.

What verified means

Three levels, distinguished throughout. Verified against code: the mechanism is implemented in the Code Kit and covered by a test that would fail if it regressed. Documented: the behaviour is stated in a primary source, meaning Google Search Central, Bing Webmaster documentation, or an operator's own crawler documentation. Opinion: a judgement call where no primary source exists, labelled as one.

Anything that is none of the three does not ship. The mapping from claim to evidence is maintained in a claims register in the repository, which is both an integrity guard and the thing that makes updating the document tractable when a platform changes.

How often it is reviewed

Two chapters are reviewed quarterly, because they age fastest: bot economics and AI answer engines. Crawler user agents change, engines are added, and citation behaviour shifts on a scale of months.

Everything else is reviewed annually. Every chapter carries both an updated date and a separately maintained last-reviewed date, because those are different claims and conflating them turns a trust signal into a false one.

Limitations, stated openly

The audit is of one codebase. Numbers described as coming from the reference implementation are one site's numbers, not an industry benchmark, and they are labelled that way wherever they appear.

The AI and answer engine chapters carry more opinion than the rest, because the operators publish very little about selection. Observations there come from a fixed set of 20 seed queries run monthly across three engines, which is a small sample and is described as one.

This site scores itself against its own checklist and publishes the failures. That is the strongest available evidence that the methodology is applied rather than described.

Verify it yourself

The proof page carries this site's current score against its own checklist, including every check it fails.