← All checks

Text to HTML ratio checker for AI crawlers

We divide the length of visible text in your homepage HTML by the total HTML length. A ratio of 0.15 or more scores full points, 0.08 or more scores one.

Why it matters

Crawlers and agents work with a limited amount of text per page. When 95% of the response is inline CSS, hydration JSON and nested wrapper divs, the real content is a small part of what is fetched, and it may fall past the cut-off. Script and style contents are excluded from the text count, so they lower the ratio.

How to fix it

1. Move large inline <style> and <script> blocks into external files.
2. Remove serialized state dumps such as __NEXT_DATA__ from pages where it is not needed.
3. Cut wrapper divs and utility-class bloat in repeated components.
4. Add real text content where the page is only images or icons.

How to verify

Compare curl -s https://example.com/ | wc -c with the same output piped through sed "s/<[^>]*>//g", or re-scan this page.

Check your site free

Scan a site

Related checks

• Markdown negotiation
• llms.txt
• llms-full.txt / .md of key pages
• No walls
• Content without JS

All checks