gibbert.uk

We gave an AI model the visible text of your page and asked it two questions: what does this page offer, and what is the price or main action? It could not answer from what your page serves.

This is the same task a shopping assistant performs before recommending you. If a model reading your page can’t name your offer and how to act on it, an assistant answering a customer can’t either, and it routes the recommendation somewhere it can.

1. State what you offer in plain text near the top of the page: a sentence a stranger could quote.
2. Put the price (with currency) or one clear primary action ("Start free trial", "Book a demo") in server-rendered text, not only inside images or scripts.
3. If the probe found your offer but no price or action, that is the half to add. The check’s finding above says which half was missing.

• Re-scan: "Primary action reachable" flips when the model can name both the offering and the price or call to action.
• Read your page’s first screen of text aloud. If it doesn’t say what you sell and what to do next, neither does the HTML.

• Don’t hide a text summary for bots that humans never see. The probe reads the same visible text extraction as every other check, and divergence between audiences reads as cloaking.
• Don’t bury the primary action under five equal-weight buttons; a model (and a customer) should be able to tell which action is THE one.

A sitemap is a machine-readable list (sitemap.xml) of every page on your site. It is how crawlers find pages that aren’t linked from your homepage.

Agents discovering your site crawl outward from the pages they know. Without a sitemap, anything not linked from a top-level page (docs, older articles, product variants) may simply never be read, so it can’t be quoted or recommended.

1. Most platforms generate one automatically: Next.js (app/sitemap.ts), WordPress (Yoast/Rank Math), Shopify and Wix (built in). Turn it on rather than hand-writing.
2. Add a "Sitemap: https://yoursite.com/sitemap.xml" line to robots.txt so crawlers find it.
3. Submit it once in Google Search Console and Bing Webmaster Tools for the search side.

• Open yoursite.com/sitemap.xml. It should list real page URLs, not error.
• Re-scan: "Sitemap present" flips to PASS.

• Don’t list URLs that redirect or 404. A sitemap full of dead links teaches crawlers to distrust it.
• Don’t include private or staging URLs; a sitemap is a public document.

HTTP responses can carry a Link header: the same canonical and alternate-language information your HTML declares, but delivered in the response envelope itself. Agents that only send a HEAD request, or that decide what to do while the page is still streaming, read the header without parsing any HTML. Yours is missing (or carries none of the relations agents use: canonical, alternate, describedby).

An agent triaging fifty URLs doesn’t want to download and parse fifty pages to learn which are duplicates of which. The Link header answers at the cheapest possible layer. Sites that provide it get correctly de-duplicated and correctly language-routed even by the most minimal fetchers.

1. Emit a Link header alongside each page, mirroring what your HTML head already declares. Example: Link: <https://yoursite.com/page>; rel="canonical".
2. Most stacks set this in one place: Next.js headers() config, an nginx add_header line, or a CDN response-header rule.
3. If you serve translations, add rel="alternate" entries with hreflang, matching your HTML.

• Run: curl -sI https://yoursite.com | grep -i "^link:" and confirm the relations appear.
• Hit Re-scan above: "Link response headers" flips to PASS.

• The header must AGREE with the HTML. A Link header pointing one place while the in-page canonical points another gives machines two contradictory answers, which is worse than one missing answer.
• Don’t inject the same site-wide canonical on every page via a blanket CDN rule; each page names its own clean URL, exactly as in the HTML tag.

The meta description is a one-sentence summary of the page that you control, embedded in the HTML head. Without it, machines compose their own summary from whatever text appears first, often cookie-notice or navigation boilerplate.

This is the sentence you get to write about yourself. Pages without one are summarised by autopilot, and the autopilot frequently leads with the least representative text on the page.

1. Write one honest sentence per page (120–160 characters) saying what the page offers.
2. Add it as <meta name="description" content="…"> per page via your framework or SEO plugin.

• Re-scan: "Meta description" flips to PASS.

• Don’t reuse one description site-wide. A shared summary describes none of the pages.
• Describe the page, not your aspirations. Agents compare the description against the content and discount pages where they diverge.

Of everything your server sends for this page, very little is actual readable text. The rest is code, styling, and markup wrapper. Agents fetched a lot of bytes and found few words.

Agents work with retrieval budgets. A page that is 2% text either gets skimmed (and mis-summarised) or skipped. More signal per byte means more of your actual message survives into the agent’s summary.

1. Confirm your real copy is server-rendered. See the JavaScript check; these two usually fail together.
2. Cut boilerplate wrappers: deeply nested divs, inline SVG logos repeated per section, base64 images inlined into HTML.
3. Move large inline styles and scripts into external files.
4. Check how much of the document is inline script: your report now says. If most of it is, that is your framework serializing state for hydration, often a second copy of text already in the HTML. Send only the props a component cannot recompute, and render statically where a page does not need to hydrate.

• Re-scan: "Text-to-markup ratio" improves, and "Page weight" often improves alongside.

• Don’t pad the page with keyword text to inflate the ratio. The ratio is a proxy for substance, and stuffing is the opposite of substance.