Free tool
Has something an engine can lift verbatim
Engines quote sentences that stand on their own: a definition, a list, a table, a number. Prose that only makes sense in context does not get lifted.
Article schema with a real author and dates
Article JSON-LD is how an engine knows this is a dated, attributed piece rather than a landing page. A named person as author is part of why it gets quoted.
Engines can tell it is current
Engines prefer the newest page that answers the question. A post with no date loses to a worse post that says it was updated last month.
Answers before the first subheading
An engine reads the opening paragraph to decide whether the page answers the question. A post that opens on a heading or an image gives it nothing to decide with.
Reachable from another owned page
A post only the sitemap knows about is a post crawlers rarely re-read. Internal links are how they find it and how they weight it.
Title length, meta description length, alt text, readability scores, word count, keyword density, broken links and page speed. Every one of them is something Ahrefs, Screaming Frog or a fifty-page agency audit will already tell you, and none of them is why an engine passed over a page. Keeping them out is what leaves room for the finding that matters.
Next
One blog post, guide or article, not your homepage. The five checks are about whether a piece of writing can be quoted; the homepage is graded by the AI Readiness Checker instead, which asks whether the site can be read at all.
Because title length, alt text, readability scores and word counts are a commodity every crawler sells, and none of them is why an engine failed to quote a page. The five here are the ones that change whether a sentence from the page can be lifted into an answer.
The tool classifies the page from its own markup: Article or BlogPosting JSON-LD, og:type, an article element with a date, or a byline with a date. If none of those are present, engines have the same trouble telling it is a dated piece of writing. The evidence line names which signal decided it.
The fixes here are static instructions, written for the platform we detect where that changes the answer. Inside CueScout, "Draft the fix" also writes the block for you, an opener or three FAQ pairs, using only facts that appear on the page. Nothing is invented and nothing is posted.
Not here. That needs the citation corpus and Search Console rows an account holds. The full Site audit runs these five checks on every article on your site and adds the line no other audit has: how much Google demand a page has and whether any engine has ever quoted it.