Can a machine
actually read your site?
Paste a URL. It gets fetched the way an assistant fetches it - no browser, no JavaScript - and scored against the same rubric the paid audit uses. Including the parts it can't check, which it will tell you about.
Do these, in this order
Every check, and what it measured
Six a fetch can settle outright, scored full, half or zero exactly as the rubric scores them. Six more that can prove something is missing but can't judge whether what's there is any good - those report and score nothing, rather than inventing a number.
What this can't see at all
Thirty points of the rubric need judgment or off-site data. Named here rather than quietly dropped from the denominator.
This measures mechanics. It cannot tell you whether your positioning is right, whether your comparison pages are honest, or whether anyone independent has ever written about you - and those decide as much as anything here. A fetch produces evidence. A person produces the conclusion.
Why the first check is the only one that matters
If your content isn't in the HTML your server returns, nothing else on this page is worth reading. That's not a weighting - it's a gate.
A retrieval layer fetches your URL and reads what comes back. It doesn't run your JavaScript, doesn't wait for hydration, and doesn't tell you it found nothing. So a site that looks finished in a browser can be an empty div to everything that would otherwise quote it, and the failure is completely invisible from the inside: no error, no warning, no missing-page report. It's the single most common fatal problem on sites built fast, which is why it's worth 12 points and stops the scoring.
Three gates, in order
- Retrievable. Can a machine fetch and read you at all? 40 points, and two of them are hard stops.
- Attributable. Can it tell what you are and who you're for? 35 points.
- Quotable. Did you publish anything worth extracting? 25 points.
Sequential, not weighted. A page nothing can fetch cannot be attributed; an entity nobody can identify cannot be quoted usefully. Most advice in this space starts at gate three, which is why most advice in this space doesn't work.
Every number here comes from one fetch of your page, and the tool reports the raw measurement next to the verdict - the word count, the response time, the exact crawlers named in your robots.txt - so you can disagree with a threshold and still use the evidence. The full rubric is published, including the 58 points this tool refuses to guess at.