seoClarity's AI Search Readiness Checker marked our page down for a cookie banner
What seoClarity's AI Search Readiness Checker reported
We ran aiwebpageseo.com through the free seoClarity AI Search Readiness Checker. It scored the page 98/100, "Low Risk", and said: "Minor optimizations may still improve citation accuracy."
It found 8,898 characters of text without JavaScript and 9,196 with JavaScript: a difference of 298 characters, with 97% of the text the same.
So the 2 missing points come from 298 characters that AI crawlers supposedly can't see. That's all the free check gives you. It doesn't tell you what those characters are.
What the 298 characters actually are
We went through every script on the page. Only two of them add text.
The cookie banner, 250 characters. Our consent script adds it after the page loads: "We use cookies — essential cookies keep you signed in and process payments…", plus the Reject all, Settings and Accept all buttons.
Two live status lines, about 42 characters. The trust cards under our menu show the current Google Safe Browsing result ("No unsafe content found") and our Mozilla Observatory grade and score. They're filled in from our server's latest check.
That accounts for about 292 of the 298. The rest is spacing between the banner's pieces.
Everything else, including every audit description, every question and answer, and all the structured data, is in the HTML that AI crawlers receive.
Why the deduction is wrong
The checker counts characters, not content. It can't tell a cookie banner from a paragraph about your product, so every extra character counts against you equally.
Following its suggestion would mean putting the cookie banner into the HTML so AI crawlers read it. Nobody wants ChatGPT or Google's AI Overviews quoting a cookie notice. A consent banner is meant to appear only for the visitor in front of it.
A useful check would show which text differs, not just how much, and leave out consent banners and other interface text before scoring.
Check what AI crawlers really see
Our AI Agent Compatibility audit fetches your page as GPTBot, ClaudeBot and PerplexityBot, and shows what each one actually gets. It detects bot-blocking, JavaScript-only content, cloaking and missing AI signals.
Frequently asked questions
What does the seoClarity AI Search Readiness Checker measure?
The free check compares how much text a page has without JavaScript against how much it has with JavaScript, then scores the page on the difference. On the AI Web Page SEO home page it found 8,898 characters without JavaScript and 9,196 with it, a difference of 298 characters with 97% of the text the same. It scored the page 98 out of 100, rated it Low Risk, and said minor optimisations may still improve citation accuracy. The free check reports how many characters differ but does not tell you what those characters are, so you have to trace them yourself.
Should a cookie banner be in the HTML so AI crawlers can read it?
No. A consent banner is meant to appear only for the visitor in front of it, and nobody wants ChatGPT or Google's AI Overviews quoting a cookie notice. When AI Web Page SEO traced the 298 JavaScript-only characters flagged by seoClarity's free checker, 250 of them came from the cookie banner added by the consent script after the page loads, including its Reject all, Settings and Accept all buttons. Following the checker's suggestion would mean putting that banner into the HTML. A more useful check would show which text differs, not just how much, and exclude consent banners and other interface text before scoring.
Why can a JavaScript text difference score be misleading for AI search readiness?
A score based on character counts cannot tell a cookie banner from a paragraph about your product, so every extra character counts against the page equally. On the page tested, only two scripts added text: a 250-character cookie banner and two live status lines of about 42 characters showing a Google Safe Browsing result and a Mozilla Observatory grade, filled in from the server's latest check. Spacing accounted for the rest. Every audit description, every question and answer and all the structured data was already in the HTML that AI crawlers receive, so the deduction did not reflect missing content.