What the check looks at
What Google shows: the title and description, whether they exist and whether they fit. Those two lines decide whether anyone clicks, and most pages get at least one of them wrong.
What Google reads: one main heading, enough text, and whether the page answers early or hides the answer under three paragraphs of introduction. The same thing decides whether AI answers cite the page.
What Google needs: structured data, canonical, language, mobile viewport, alt text on images and links to the rest of the site. Dull things nobody sees, which is why they are so often missing.
What is often invisible: prices that only exist in an image, and a noindex instruction left over from when the site was built. Both mean the page can never be shown, however good it is.
What the tool does and does not do
- We read one page at a time, exactly as your server delivers it. Whatever JavaScript builds afterwards we do not see, just as most AI crawlers do not.
- We measure against Google's documented limits and our own measurements of what gets cited. We do not guess the page's ranking; that needs Search Console.
- Nothing is stored. The address you enter is not logged, and the page content is discarded once the answer is sent.
Tedrix runs the same check across your whole site every week, sees which pages are getting fewer visitors, and rewrites what needs changing.