AI & TechnologyHow to Show Up in Google AI Overviews (2026)

Google says there is no special trick to being cited in AI Overviews or AI Mode. A page needs to be indexed and eligible to show a snippet in Google Search, and after that it competes on the same quality and relevance signals as any other result. There is no extra markup, no AI text file and no separate sign-up. What you can control is whether you are accidentally blocking yourself, and whether the page is worth citing.

What Google actually requires

Google's AI features documentation is short and clear on this point: "There are no additional requirements to appear in AI Overviews or AI Mode." The usual Search technical requirements and policies still apply. Indexing and serving are never guaranteed, even for pages that follow the rules.

Its newer guide on optimizing for generative AI features says the same thing from another angle. These features rely on Google's core Search ranking and quality systems. Google treats "AEO" and "GEO" as names for SEO applied to AI search.

That is useful for a business owner, because it means the work is not a separate project. If you want the general Google ranking basics, I cover them in how to rank on Google in the age of AI search. This guide stays on the AI Overviews specifics.

Indexed and eligible for a snippet

Google says that to be shown as a supporting link in these features, a page must be indexed and eligible to appear in Search with a snippet. Two things follow from that:

1. If the page is not indexed, nothing else on this list matters.
2. If something stops Google showing a snippet from the page, you may be removing yourself from AI features without knowing.

The second one is the quiet failure. Which brings us to snippet controls.

Snippet controls can remove you from AI features

Google's robots meta tag documentation lists several controls. Here is what each does, in Google's words where it matters:

Google's AI features page lists nosnippet, data-nosnippet, max-snippet and noindex together as the way to limit what is shown from your pages.

Why this matters in practice: a directive set once for a legal or copy-protection reason can sit in a template or header and be forgotten. Check your page source and response headers for nosnippet and a low max-snippet value. If you do use data-nosnippet, keep it for the parts you really want kept out, such as a price table you do not want quoted.

If you change a control, Google says to confirm Googlebot can see it with the URL Inspection tool. Recrawling can take days to months, so a fix will not show up overnight.

Google-Extended is not an AI Overviews switch

It is easy to confuse the two. Google-Extended is a standalone token you can put in robots.txt. Google's crawler documentation says it manages whether crawled content may be used to train later Gemini models, and for grounding in Gemini Apps and Grounding with Google Search on Vertex AI. It has no separate user agent, since crawling uses existing Google user agents.

Google also states that Google-Extended does not impact a site's inclusion in Google Search and is not a ranking signal. That page does not mention AI Overviews at all. So I do not tell clients that blocking it removes them from AI Overviews, and I do not tell them it keeps them in. If you want to limit how Search features use a page, the documented tools are the snippet controls above.

For the crawlers behind ChatGPT and Perplexity, which are separate, see how to get your SaaS cited in ChatGPT and Perplexity.

What you do not need to do

Google's guide has a list of myths. The ones that matter for a small business:

• llms.txt and special files. Google says you do not need new machine-readable files, AI text files or Markdown to appear in Search, and that Google Search does not use them. Creating one neither helps nor harms your Search visibility. Other services may use them. More in what llms.txt does.
• Special schema. Not required for generative AI search. Google still recommends structured data for rich result eligibility.
• Chunking. No need to split pages into tiny pieces. Google says its systems can handle multiple topics on one page and there is no ideal page length.
• Rewriting for AI. No special phrasing is needed, because the systems understand synonyms and intent.
• Fake brand mentions. Seeking artificial mentions across the web does not help, since ranking and spam systems filter for quality.

What does help: being worth citing

Google's guide favors original, first-hand content over summaries or generic material that a generative tool could write. It also warns that mass-producing pages for query variations can violate its scaled content abuse policy.

AI features also use what Google calls query fan-out: they issue several related searches across subtopics and data sources, and can find extra supporting pages while writing the answer. Google says this can show a wider and more diverse set of links than classic search. For you, that means a narrow, specific page can be cited for a sub-question even if it would never win the head term.

For a service business, I would put effort here:

• Answer the specific questions your buyers ask, with details only you have: your process, your real prices or ranges, your service area, examples from your own work.
• Keep pages focused on one job, so the useful answer is easy to find.
• For local and commerce businesses, Google says Merchant Center and Google Business Profile can help products and businesses appear in AI responses.

Google also says AI Overviews appear only when the system decides they add value beyond classic results, so they often do not trigger. And AI Mode and AI Overviews may use different models, so the links can differ between them. Do not expect a stable, guaranteed slot.

How to measure it

Google says AI Overviews and AI Mode traffic is included in overall Search traffic and appears in the Performance report under the Web search type. It is not split out into its own search type there. Google's optimization guide also points to a Generative AI performance report in Search Console for how your content performs in AI features. I have not tested that report against a live property for this article, so check what it shows for your own site.

For a view Search Console does not give you, you can run your real buyer questions yourself, logged out, and note which sources get cited. Google also notes that no third-party tool has access to its internal ranking or AI systems, so treat any tool's "AI Overviews score" as an estimate.

Checklist

1. Confirm your key pages are indexed (URL Inspection in Search Console).
2. Check page source and headers for nosnippet, a low max-snippet value, and noindex on pages you want cited.
3. Review any data-nosnippet attributes and remove ones added by accident.
4. Do not rely on Google-Extended as a way in or out of AI Overviews.
5. Skip llms.txt and extra schema as an AI Overviews tactic. Keep schema accurate for rich results.
6. Write pages that answer one buyer question with first-hand detail.
7. Claim and complete your Google Business Profile, or Merchant Center if you sell products.
8. Watch the Web search type in the Performance report and the Generative AI performance report.
9. Re-check after changes, allowing days to months for recrawling.

Where I can help

If you want to see where you stand rather than guess, I offer a fixed-price AI search visibility audit at $750. I take 30 buyer questions, run each three times per engine in ChatGPT, Perplexity, Google AI Overviews and Gemini while logged out, and report citation share. Gemini results are indicative. The wider SEO, GEO and AEO service covers the fixes that follow. I do not promise a citation, because nobody outside Google can.