AI Overview SEO Steps That Google's Own Documentation Supports

AI Overview SEO is the work of getting your page picked as a supporting link in a Google AI Overview. Most of the guidance sold under that name comes from tool vendors, so this guide starts from Google's own documentation.

AI Overview SEO is ordinary Search SEO plus one eligibility step. Google says a page qualifies as a supporting link if it is indexed and eligible to show with a snippet. Its July 2026 guide adds that the site must be included in Search generative AI features in Search Console. Google says llms.txt files and content chunking are unnecessary.

What is AI Overview SEO according to Google?

Google treats AI Overview SEO as Search SEO. Its guide for site owners says the usual SEO best practices still apply, because AI Overviews and AI Mode run on Google's core ranking and quality systems. The features retrieve pages from the Search index with retrieval-augmented generation, and they send related searches for subtopics through query fan-out. The guide also says that terms such as Answer Engine Optimization and Generative Engine Optimization are common, and that many of the hacks sold under them do not work the way Search works. For the ChatGPT side of the same question, see How to get cited by ChatGPT with crawler setup and clean capsules.

How much do AI Overviews cut clicks?

Clicks on traditional results fall when an AI summary appears. Pew Research Center's browsing study, published 22 July 2025, found that users clicked a traditional result on 8% of visits where an AI summary appeared, against 15% of visits without one. Pew also found an AI summary on 18% of the Google searches in its sample, and 88% of those summaries cited three or more sources. Pew's study is the primary source for those figures.

Ahrefs re-ran its study on December 2025 data and published the update on 4 February 2026. Across 300,000 keywords, it found that an AI Overview correlates with a 58% lower average clickthrough rate for the top-ranking page, up from 34.5% in its April 2025 study. Ahrefs' update built the figure from aggregated Search Console data. It describes a pattern in one keyword sample, so it does not measure what will happen to your site.

Before you start

Verify the site in Search Console and confirm you can open the Performance report and the generative AI performance report. Google's documentation names verifying your site in Search Console as the fastest way to find and diagnose technical problems.

This post is on bunny.directory, a launch directory for startups, so weigh the next point with that in mind. A founder can submit a product to bunny.directory for free. Each listing gets a permanent product page with a link back to the product's site, dofollow for as long as the bunny badge stays on that site. Google's guide does not address directory listings, so treat that link as distribution, not as a ranking step.

Step 1: Confirm each target page is indexed and eligible for a snippet

The only technical requirement for AI Overviews and AI Mode is that the page is indexed and eligible to show in Search with a snippet. Google's AI features documentation, last updated 10 December 2025, states there are no additional technical requirements. Run each target URL through the URL Inspection tool, which shows the HTML Googlebot received. Then confirm that robots.txt, and any CDN or hosting layer, allows crawling.

Preview controls are the other common cause of exclusion. Google says nosnippet, data-nosnippet and max-snippet limit what appears from a page in Search, and noindex blocks indexing altogether. If you set any of them by mistake, a page can drop out of AI features.

Step 2: Include the site in Search generative AI features

Google's generative AI guide, updated 10 July 2026, says a site must be included in Search generative AI features in Search Console to be eligible for display in them. The December 2025 documentation listed no extra technical requirement, so this inclusion step appears only in the newer guide. Check the setting described in help article 16908024 if you last reviewed Search Console before 10 July 2026.

Google also began rolling out an exclusion toggle on 3 June 2026. It started with a subset of UK site owners. Google says sites that opt out receive no traffic or impressions from those features, and that the control is not a ranking signal for classic results, according to Search Engine Land's report on the rollout.

Step 3: Publish first-hand content that a summary cannot restate

Google's guide says that unique, non-commodity content will likely shape your presence in generative AI search more than any other suggestion in the guide. A first-hand review carries experience that a summary of existing sources does not. Google advises against recycling what others have already said, or what a generative AI model could produce with little effort.

Google also warns against volume. Creating separate pages for each query variant, mainly to manipulate rankings or AI responses, violates its scaled content abuse spam policy. The guide says a high number of pages does not make a site more relevant. Google's test is whether visitors would find the content satisfying. The creating helpful, reliable, people-first content guide explains that test in more detail.

Step 4: Make the answer plain text that matches your markup

Google's guide says important content should be available as text, organized into paragraphs and sections with clear headings. Images and video can support the text, and Google says they create more places where a page can appear. Structured data is not required for generative AI features, and Google says there is no special schema.org markup to add. If you use structured data, the AI features documentation says it must match the visible text on the page.

JavaScript needs one check. Google processes content inside JavaScript as long as it is not blocked. If the answer only appears after client-side rendering, test it in the URL Inspection tool before you assume Google sees it.

Step 5: Read the generative AI performance report and the Performance report together

The generative AI performance report shows impressions, not clicks. It breaks impressions down by pages, countries, devices and dates, and you can filter by Web: text-based or Web: multimodal. Google said on 31 August 2026 that it had rolled these insights out to all websites worldwide. The newest data may be preliminary, shown as a dotted line.

Clicks from AI features are counted in the overall search traffic in the Performance report, under the Web search type, as Google's AI features documentation describes. To check whether a change worked, export your top queries from the Performance report. Compare two date ranges of equal length, one before and one after the change. Then run the same comparison on queries that show no AI Overview. This check is my own method, not a Google recommendation, but it separates a change in your page from a change in demand.

Which report tells you each step worked?

Common mistakes

• Adding llms.txt or Markdown files. Google says Search does not use them, so you do not need them to appear in its generative AI features.
• Chunking pages into small pieces. Google says there is no requirement to do this, and its systems can find the relevant part of a longer page.
• Rewriting copy for AI systems. Google says its systems understand synonyms and meaning, so you do not need to chase every long-tail variant.
• Seeking inauthentic mentions. Google says they are less helpful than they appear, because its AI features depend on both quality ranking and spam blocking.
• Adding special schema markup. Google says there is no special schema.org markup for AI features.
• Reading the Ahrefs figure as your own loss. The 58% comes from a large keyword sample, so measure your own exposed queries in Search Console.

What's new (as of 6 October 2026)

• 31 August 2026: The generative AI performance report reached all websites worldwide. It reports impressions, pages, countries, devices and dates, but no click data, according to Google's help article.
• 10 July 2026: Google's generative AI optimization guide listed the tactics to ignore and added the Search Console inclusion requirement.
• 3 June 2026: Google launched dedicated Search and Discover generative AI reports to a subset of sites, as described in the Search Central blog. The same day it began rolling out the exclusion toggle to a subset of UK owners.
• 4 February 2026: Ahrefs reported the 58% lower average clickthrough rate for top-ranking pages, based on December 2025 data.
• 10 December 2025: Google last updated its AI features documentation, which lists no additional technical requirements beyond indexing and snippet eligibility.

Key takeaways

• Google's documentation treats AI Overview SEO as Search SEO, plus a Search Console inclusion requirement for generative AI features.
• Google says llms.txt files, content chunking, AI-specific rewriting, inauthentic mentions and special schema are not needed.
• Pew measured clicks on traditional results at 8% of visits with an AI summary and 15% without one.
• The generative AI performance report in Search Console shows impressions, not clicks.
• Clicks from AI features are counted in the Performance report under the Web search type.

FAQ

Does Google require llms.txt for AI Overviews?

No. Google's guide says you do not need to create llms.txt files, AI text files, markup or Markdown to appear in Google Search, including its generative AI features, because Search does not use them. The llms.txt setup and file specs guide covers what the file is and what it does not do for search.

Can I see clicks from AI Overviews in Search Console?

Not in the generative AI performance report, which shows impressions only. The Performance report counts sites appearing in AI features within the overall Web search traffic, and Google's documentation does not describe a separate click figure for those features.

Can I opt my site out of AI Overviews?

For some owners, yes. Google began rolling out a Search Console toggle on 3 June 2026 to a subset of UK owners and has said it will expand access later. Sites that opt out lose traffic and impressions from those features. Preview controls such as nosnippet limit what appears, and Google-Extended covers AI training and grounding in some of Google's other systems.

If you are still setting up the technical basics before you submit anywhere, the startup launch checklist runs in your browser.