AI Visibility pillar / check ai-answer-snippet-controls

Snippet and archive controls that limit AI answers

We read every robots meta tag (robots, googlebot, bingbot) and X-Robots-Tag directive plus data-nosnippet coverage, and show the effect Google and Bing document for each one on AI answers.

By Shimon Carroll, Founder, SEO for AI Agents · Last updated

What this check measures

Every robots directive on the page: meta name="robots", "googlebot" and "bingbot" (or "msnbot"), and the X-Robots-Tag response header including bot prefixes such as "googlebot: nosnippet". We look for noarchive, nocache, nosnippet and max-snippet, and measure how much of the main text sits inside elements marked data-nosnippet. Each directive is listed with the engine it applies to and that engine's documented effect.

Why it matters

Google's robots meta specification says nosnippet also prevents the content from being used as a direct input for AI Overviews and AI Mode, and that max-snippet also limits how much content may be used as a direct input for them. Google's AI features page says to use nosnippet, data-nosnippet, max-snippet or noindex to limit what is shown. Bing's Webmaster Guidelines say NOARCHIVE prevents content from being used in Copilot responses and grounding results, NOCACHE limits Copilot to the URL, title and snippet, and NOSNIPPET and DATA-NOSNIPPET may limit Copilot citation quality. These controls are legitimate; we only make sure you know what they do. We do not measure Copilot citations; we check only whether your page allows them.

How we score it

Measured. noarchive for all bots or for Bing, and nosnippet or max-snippet:0 for all bots or for Google, are MEDIUM. nocache, a small max-snippet value, a Bing-only nosnippet, and data-nosnippet over most of the main text are LOW. max-snippet:-1 means no limit and passes. Plain noindex is reported by the indexability checks, not here. Every failing finding says: if you set this on purpose, ignore this item.

Confidence-flag rules

HIGH. The directives are read verbatim from the HTML and the response header. We cannot see controls a CDN adds only for some user agents.

Common mistakes

  • Leaving a noarchive tag from an old caching concern, which now keeps the page out of the AI answers Bing grounds.
  • Setting max-snippet:0 to stop snippet scraping, which also removes the page from AI Overviews input.
  • Wrapping a whole page template in data-nosnippet instead of one sensitive passage.

How to fix it

Remove noarchive, nocache, nosnippet and small max-snippet values from pages you want quoted. To keep one passage out of snippets, wrap only that passage in data-nosnippet.

Primary sources

Changelog

  • · Initial publication. Measured check. max-snippet:-1 means no limit and passes. Sources re-read in a real browser on 2026-10-07.
← Back to the methodology indexRun an audit