- Why Google says AI search is still SEO
- The technical gate, according to Google
- What Google means by non-commodity content
- What Google says you can skip
- RAG and query fan-out in plain words
- How Google says to measure it
- What Google says about AEO and GEO services
- Seven checks to run against the guide
- My verdict: treat Google's guide as the floor
- Frequently asked questions
Google's AI optimization guide says that optimising for AI Overviews and AI Mode is still SEO, and that there are no additional requirements to appear in them. As of October 2026, the one technical gate Google names is that a page must be indexed and eligible to be shown in Google Search with a snippet.
The guide is called 'Optimizing your website for generative AI features on Google Search'. It sits on Google Search Central, it's written for website owners, and the page shows 'Last updated 2026-07-10 UTC'. Here's the guide read for a business owner, with the parts you can skip.
Why Google says AI search is still SEO
Google's answer to whether SEO still matters is yes. Its reasoning is short: the generative AI features are rooted in Google's core Search ranking and quality systems, so the best practices that work for classic results stay relevant. Google's separate page, 'AI features and your website', says the same for AI Overviews and AI Mode and adds that no other special optimisations are necessary.
Google also explains when these features show up. It says AI Overviews only appear when its systems decide they add something to classic Search, so they often don't trigger. AI Mode is described as most helpful for queries that need further exploration, reasoning or complex comparisons. And because the two may use different models and techniques, Google says the responses and links they show will vary.
The technical gate, according to Google
To be eligible for generative AI features, Google says a page must be indexed and eligible to be shown in Google Search with a snippet. It also says the site must be included in Search generative AI features in Search Console. Meeting every requirement doesn't guarantee crawling, indexing or serving, and Google says so plainly.
The crawling advice is the same advice you'd get for any site. Google says generative AI models use publicly accessible, crawlable content, so it asks owners to:
- Follow crawling best practices so Google can reach the pages.
- Use semantic HTML, which Google calls a good idea, though it says perfect code isn't required.
- Follow JavaScript SEO best practices if the site relies on JavaScript.
- Reduce duplicate content.
- Verify the site in Search Console.
What Google means by non-commodity content
This is the part I'd read twice. Google says creating valuable, non-commodity content will likely influence your presence in generative AI search in the long run more than any other suggestion in the guide. It describes that as a unique point of view, such as a first-hand review rather than a summary of what's already out there, organised so it helps the reader.
Google's own examples make the point. A post called '7 Tips for First-Time Homebuyers', built on common knowledge anyone could write, is its example of commodity content. A first-hand account of waiving an inspection, and what that saved, is its example of the non-commodity kind.
Google also says to add high-quality images and video where they make sense, because the AI features can bring them in. If you use generative AI tools to help create content, Google says the work must still meet its Search Essentials standards. And for local businesses and shops, Google says AI responses can include product listings, product information and local business details, with Merchant Center and your Business Profile helping your products and services be visible there.
What Google says you can skip
The guide has a mythbusting section, and it's the most useful part for anyone being sold an AI search package. Each item comes with what Google says to do instead.
| What you're told to do | What Google says | What Google says to keep doing |
|---|---|---|
| Write an llms.txt file for Google | Google Search doesn't use llms.txt or other 'special' markup | Have one for other services if you like: it won't harm or help your visibility in Google Search |
| Chunk content into tiny pieces | There's no requirement to chunk and no ideal page length | Organise content so it helps readers |
| Rewrite pages just for AI | Not needed, because AI systems understand synonyms and general meaning | Put the effort into non-commodity content |
| Add special schema for AI | Structured data isn't required and there's no special schema.org markup | Keep using structured data as part of overall SEO |
| Chase mentions across the web | Seeking inauthentic mentions isn't as helpful as it might seem | Google gives no reason to keep doing it |
| Make a page for every search variation | Doing it to manipulate rankings or AI answers breaks the scaled content abuse spam policy | Remember that more pages don't make a site higher quality or more relevant |
Google also calls a page for every fan-out query an ineffective long-term strategy. For llms.txt and other AI tools, I've covered whether your website needs an llms.txt file separately.
RAG and query fan-out in plain words
Retrieval-augmented generation, or RAG, also called grounding, is how the answer gets its sources. Google says it uses its core ranking systems to retrieve relevant, up-to-date pages, then generates a response with clickable links.
Query fan-out is the model running a set of related searches at the same time to fetch more results. Google's example is 'how to fix a lawn that's full of weeds' fanning out to queries like 'best herbicides for lawns'. On the AI features page, Google adds that fan-out works across subtopics and data sources, which is why AI Overviews and AI Mode can show a wider and more diverse set of links than a classic search.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
How Google says to measure it
Google points to one place: the Generative AI performance report in Search Console. As of October 2026, the guide says it covers generative AI features on Google Search and Discover. I'd open it, note where you stand today, and treat that as your baseline before changing anything.
What Google says about AEO and GEO services
Google says AEO and GEO are terms for work focused on visibility in AI search experiences. From Google Search's point of view, though, optimising for generative AI search is optimising for the search experience, and so it's still SEO. Google tells owners weighing up third-party AEO or GEO advice or services to review its own guidance.
On tools, it's blunter. Google says to be wary of any third-party tool that promises ranking success or claims to use 'internal' Google metrics, because no third-party tool has access to Google's internal ranking or AI systems. Its advice is to evaluate their recommendations against the official guidance.
Seven checks to run against the guide
- Check key pages are indexed and snippet-eligible in Search Console.
- Make sure Google can crawl them.
- Ask whether each page says something only you could say.
- Drop llms.txt, chunking and special schema work done for Google's sake.
- Keep Business Profile and Merchant Center details current.
- Record a baseline in the Generative AI performance report.
- Hold any AEO or GEO agency's claims up against this guide.
My verdict: treat Google's guide as the floor
I'd treat this guide as the floor for any AI search work on Google. If an agency or consultant pitches you AEO or GEO, me included, test every claim against it. If they're selling llms.txt for Google, chunked rewrites or special schema, Google has already told you those aren't needed. I wouldn't pay for any tool that claims to see inside Google's systems, because Google says none can.
Where I'd put the effort is the thing Google says will likely matter more than any other suggestion in the guide: content only you can write, from your own experience and your own point of view. For the practical side of getting seen, my step-by-step guide to appearing in Google AI Overviews picks up where this leaves off.
If you'd like help turning Google's guidance into a plan for your own site, that's the kind of thing I help with. I run AI training and workshops for teams, and help businesses with AI enablement, AI automation and GEO, which is about getting your business seen and cited in AI answers. Most of that work is with SaaS marketing teams, telecom companies and small businesses. Here's how we can work together.
Frequently asked questions
Does Google use llms.txt for AI Overviews?
No. Google says Google Search doesn't use llms.txt files or other 'special' markup, and creating one for other services won't harm or help your visibility in Google Search.
Is GEO different from SEO, according to Google?
Google says that from Google Search's point of view, optimising for generative AI search is optimising for the search experience, so it's still SEO.
What does a page need to appear in AI Overviews or AI Mode?
Google says it must be indexed and eligible to be shown in Google Search with a snippet, and the site must be included in Search generative AI features in Search Console. Google says there are no additional requirements.
How do I measure AI Overviews performance?
Google points to the Generative AI performance report in Search Console, which covers generative AI features on Google Search and Discover as of October 2026.
