Site owners sometimes reach for a single tag when they want to change how a page appears in Search. But indexing a page, using text in a snippet, and using content in generative AI features are different decisions. Google refreshed its control what you share with Google Search guide on October 21, 2026, bringing several existing options together.
Start with the outcome you need
Before editing a template, write down what should change: should the URL be excluded from Search, should a specific passage stay out of a text snippet, or should content be excluded from a supported generative AI feature? These goals map to different mechanisms. A sitewide change can also affect important pages, so scope it to the pages or content that need it.
Use page or file directives for indexing and snippets
The robots meta tag can provide directives in a page’s HTML. An X-Robots-Tag response header can apply directives to non-HTML files such as PDFs. Directives such as noindex address whether a page should be indexed; snippet controls can limit how much text is shown in a result. They are not interchangeable.
Search crawlers need access to fetch a page to see its meta tag or response header. Blocking the URL in robots.txt can prevent a crawler from seeing the directive. If the goal is to keep a page out of results, follow Google’s implementation guidance and confirm the page can be recrawled.
Exclude just a section with data-nosnippet
If the page should remain eligible for Search but a specific block should not appear in a text snippet, Google supports the data-nosnippet attribute on span, div, and section elements. Apply it narrowly and check the rendered HTML. It is intended for snippets; it does not hide that content from visitors or make it private.
Understand the Search generative AI control
Google’s updated guide also lists its Search generative AI control. Read the current documentation for exactly which features it affects and how it is implemented. Do not assume that this control governs every AI service, every crawler, or ordinary Search snippets. Those are separate systems and settings.
Test the change on a small set of URLs
- Pick a representative page and decide whether you want to affect indexing, snippets, or a supported generative AI feature.
- Add the documented directive at the correct level: HTML meta tag, HTTP header, or marked page section.
- Inspect the response or rendered source to make sure the directive is present and valid.
- Check that robots.txt does not prevent the relevant crawler from seeing it.
- Allow time for recrawling, then review Search Console and actual search results. A valid directive guides Google; it does not guarantee an immediate or exact display.
Keep private information behind access controls
Search directives are not security. A noindex instruction can be missed if a crawler cannot fetch the page, and it does not prevent someone with the URL from opening public content. Put confidential information behind authentication and authorization. Use Search controls to manage public content’s appearance, not to protect secrets.
Google’s October update is a documentation refresh describing existing options, not an announcement that each control is new. The useful takeaway is to match the control to the specific Search behavior you want to affect, apply it narrowly, and verify what crawlers can see.
Primary sources: Google: control what you share with Google Search · Google Search documentation updates, October 2026 · Google: robots meta tag and X-Robots-Tag.