Google’s AI Overviews now sit at the top of many search results, and a whole industry has grown up around “optimizing” for them. Much of the advice is confusing, and some of it is simply wrong. If you own or manage a website, you want to know two things: what can you actually do to appear in AI Overviews, and which tactics are a waste of time?
This guide answers both, and it is built mainly on Google’s own published guidance rather than on guesses about how the system works. You will learn what AI Overviews are, how they choose pages, what is required to be eligible, what kind of content tends to earn a place, which popular tactics Google says you can ignore, and how to measure your results. No one can guarantee that a page will appear in an AI Overview, and this guide does not.
The quick answer
Google says there is no special optimization required for AI Overviews and AI Mode. The same fundamentals that make a site successful in Google Search apply. In practice, that means:
- Make sure your pages are indexed and eligible to show a snippet, and that your site has not been excluded from generative AI features in Search Console.
- Publish unique, non-commodity content based on real experience and expertise, not a rewrite of what is already everywhere.
- Organize pages clearly, with sensible headings, short sections and the answer early.
- Cover the topic and its likely follow-up questions on a helpful page, without creating thin pages for every query variation.
- Add high-quality images and video where they help.
- Keep local and product information accurate in Google Business Profile and Merchant Center.
- Provide a good page experience on every device.
- Measure with Search Console’s generative AI report and judge success by business outcomes, not by screenshots.
What are AI Overviews and AI Mode?
AI Overviews are AI-generated summaries that Google can show at the top of search results when its systems decide a summary would help the searcher. They draw on pages from Google’s search index and show links to supporting sources. AI Mode is a separate, more conversational search experience where people can ask longer, more complex questions and follow up. Google notes that the two may use different models and techniques, so the responses and the links they show can differ.
Google has said that AI Overviews have more than 2.5 billion monthly active users and AI Mode has passed one billion monthly users. These are Google’s own figures, but they show why many website owners care about visibility in these features. Google also says it has increased the number of inline links in AI responses and added website previews to encourage visits, and has introduced Preferred Sources, which lets people choose websites they want to see more prominently.
How AI Overviews choose sources
Google describes two ideas that help explain what happens behind the scenes.
- Grounding with retrieval. The AI does not write purely from memory. Google’s systems use core search ranking to retrieve relevant, up-to-date pages from the search index, review information on those pages and then generate a response that links to supporting pages. This approach is meant to improve accuracy and freshness.
- Query fan-out. A single question can trigger several related searches at once. For example, someone asking how to fix a patchy lawn might also cause searches about causes of brown patches, how much to water, which seed to choose and when to reseed. The system gathers results for these sub-questions to build a fuller answer.
The practical lesson is that AI Overviews are built on top of ordinary search. A page that is good at being found, understood and trusted in Google Search is the same page that can be retrieved and used here.
What Google officially says about optimizing
Google’s guidance for site owners makes several points that are worth knowing before you spend any money on AI-focused SEO services.
- SEO best practices remain relevant, because generative AI features are rooted in Google’s core ranking and quality systems.
- There are no additional technical requirements and no special markup needed to appear in AI Overviews or AI Mode beyond the normal requirements for appearing in Search.
- The terms AEO (answer engine optimization) and GEO (generative engine optimization) describe work aimed at AI search experiences, but from Google Search’s perspective, optimizing for generative AI is still SEO.
- Site owners should be careful with third-party advice. Google says no third-party tool has access to its internal ranking or AI systems, so claims to use internal Google metrics should be treated with caution.
- Plenty of content performs well in Google Search, including in AI features, without any deliberate SEO, and you do not need to do everything in the guide to succeed.
This does not mean there is nothing to do. It means the work is the same fundamental work: being eligible, being useful and being clear.
Step 1: Confirm you are eligible
Before improving content, check that nothing is blocking your pages. Google says a page must be indexed and eligible to appear in Google Search with a snippet to be shown in generative AI features. Check these items:
| Check | What to do |
| Page is indexed | Use the URL Inspection tool in Search Console to confirm Google has indexed the page. Indexing is never guaranteed, even for good pages. |
| Crawling is allowed | Make sure robots.txt, your hosting and any security or CDN settings do not block Google’s crawlers from important pages. |
| No accidental noindex | Check for noindex tags left on pages from development or from plugin settings. |
| Snippets are allowed | Pages that use nosnippet or very restrictive snippet limits can limit how your content is shown. Review robots meta tags and data-nosnippet attributes on key pages. |
| Site is included in generative AI features | Google has introduced a control in Search Console that lets site owners choose whether their site appears in and helps ground generative AI features. Sites that opt out receive no traffic or impressions from those features. Check that your setting matches your intent. |
| Content is in crawlable text | Important information should exist as text in the page, not only inside images or content that fails to render. |
| Mobile and page experience | The site should work well on phones and load quickly. See the Core Web Vitals targets in the on-page checklist. |
| No heavy duplication | Reduce duplicate URLs and near-identical pages that waste crawling and confuse indexing. |
Even when everything is in order, Google does not guarantee that a page will be crawled, indexed or shown. Eligibility is the starting point, not a promise.
Step 2: Create unique, non-commodity content
Google says that creating content people find unique, compelling and useful will probably influence your presence in generative AI search over the long run more than any other suggestion. The key idea is the difference between commodity and non-commodity content.
| Commodity content | Non-commodity content | |
| What it is | Common knowledge that could come from anyone | Insight, experience or data that goes beyond the obvious |
| Typical example | “Seven tips for first-time buyers” that repeat widely known points | A first-hand account of a specific decision, with real numbers and what you would do differently |
| Who could write it | Almost anyone, including an AI tool with no real experience | Someone with direct experience or expertise |
| Value to AI systems | Easily replaced by other sources | Offers something distinctive to cite |
| Value to readers | Little they could not find elsewhere | A reason to trust and return |
How to make your content non-commodity
- Bring first-hand experience. Describe what you did, tested or saw, with specifics.
- Add original information: your own data, pricing patterns you have observed, photos of real jobs, results from real customers (with permission), or lessons from mistakes.
- Take a clear point of view and explain your reasoning, instead of summarizing other pages.
- Include expert review for topics where accuracy matters, such as health, legal and financial subjects.
- Show who wrote the content, why they are qualified and when it was last updated.
- Be honest about limits, trade-offs and uncertainty.
- Avoid recycling what competitors already say, or what a generative AI tool could easily produce without any real knowledge.
A note on using AI to help write
Google says content created with generative AI tools must still meet its Search Essentials and spam policies. AI can help with outlining, editing and drafting, but it cannot supply experience you do not have. A knowledgeable person should check every fact and add the real examples that make a page worth citing. Do not present AI-assisted work as something it is not.
Step 3: Organize content so it is easy to follow
Google recommends writing for your human audience and structuring pages with paragraphs, sections and headings that make navigation easy. Practical steps include:
- Open with a clear answer or summary, then add detail.
- Use descriptive H2 and H3 headings, ideally phrased as the topics or questions readers have.
- Keep paragraphs short and use lists and tables where they make information easier to scan.
- Define terms in plain language the first time you use them.
- Keep facts, figures and dates accurate, and update them when they change.
- Link to credible sources for claims that need support.
Google also says there is no requirement to break content into tiny pieces for AI, and no ideal page length. Make pages as long or as short as your audience needs.
Step 4: Cover the topic and its related questions without overdoing it
Because AI features may search for several related subtopics at once, a page that thoughtfully covers the main question and the follow-ups a reader would naturally have is more useful. Consider what someone would want to know next, such as cost, timing, risks, alternatives and how to get started, and answer those where they fit naturally.
But there is a trap. Google warns that creating separate pages for every possible variation of how people might search, or for each fan-out query, primarily to manipulate rankings or AI responses, violates its scaled content abuse policy. Pumping out many thin pages does not make a site higher quality, and Google says its systems can understand relevance even when the page does not exactly match the query wording.
| Situation | Better approach | Avoid |
| Several closely related questions on one topic | One thorough page that answers them in clear sections | A separate thin page for each question |
| A truly different intent, such as pricing vs. troubleshooting | Separate pages if each can be genuinely useful on its own | Duplicating the same text with small changes |
| Many locations or services | Pages with real, different local or service details | Template pages with only the name swapped |
| Different phrasings of the same query | Use natural language and synonyms on one page | Creating pages for every keyword variation |
A worked example
Imagine a small lawn-care company wants to be a useful source for people with patchy lawns. The details here are invented to show the approach.
Weak approach: publish twenty short pages with titles such as “fix patchy lawn,””patchy lawn repair,””how to repair patchy lawn” and “patchy lawn solutions,” each with similar generic tips.
Stronger approach: publish one in-depth guide that opens with a short summary of the most common causes of patchy lawns and what fixes each. Then add sections on how to diagnose the cause, when to reseed in your region, how much water is right, what the work costs when done professionally, common mistakes the company has seen, and before-and-after photos from real jobs. Add the author’s name and experience, a last-updated date and a link to the service page.
The stronger page contains information that is specific, experience-based and organized. It also naturally answers many of the sub-questions an AI system might ask, without needing a separate page for each.
Step 5: Use images and video well
Google says generative AI features can bring in relevant images and video, which creates more ways for a site to appear beyond ordinary links. Support your text with high-quality visuals where they genuinely help.
- Use your own photos and short videos of real work, products, processes and people where possible.
- Follow image and video SEO basics: descriptive file names, accurate alt text, sensible compression and fast loading.
- Place visuals near the text they illustrate and add short captions where useful.
- Make sure video pages have text that describes the content, such as a summary or transcript.
- Do not use misleading or irrelevant images to attract clicks.
Step 6: Keep business, local and product details accurate
Google notes that AI responses can include local business and product information. If you are a local business or a seller, the details Google holds about you outside your website matter too.
- Claim and complete your Google Business Profile, and keep your name, address, phone number, hours, categories and photos accurate.
- Respond to reviews and keep service descriptions current.
- If you sell products, keep your Merchant Center feeds accurate, including price, availability and shipping details.
- Make sure the information on your website matches what appears in these profiles.
- Review Google’s documentation on business details for the specifics of what you can provide.
Google also mentions other merchant experiences such as Business Agent, a conversational feature that helps customers chat with a brand. Consider these if they fit your business and you have time.
Step 7: Get the technical foundations right
Technical SEO is still the backbone, since AI features rely on the same index. In addition to the eligibility checks above:
- Provide a good page experience: fast loading, stable layout and a design that works on all devices, with the main content easy to tell apart from ads and other elements.
- If your site depends on JavaScript, follow JavaScript SEO best practices and test that key content is visible to Google.
- Use semantic HTML where practical. Google does not require perfect code, but clear structure helps people using screen readers and other tools.
- Use internal links so important pages are easy to find.
- Reduce duplicate content and keep your sitemap clean.
- Use structured data where it accurately describes the page and helps with standard rich results, remembering that it is not required for AI features.
A note on FAQ markup: Google stopped showing FAQ rich results in Search on 7 May 2026. FAQ content is still valuable for readers, but the markup is not a path to AI Overviews.
Myths to ignore: what you do not need to do
Google’s guidance specifically lists several popular tactics as unnecessary for Google Search. Knowing them can save you time and money.
| Myth or tactic | What Google says |
| Creating llms.txt files | Google Search does not use them. You can keep them for other services if you wish, but they will neither help nor harm your visibility in Google Search. |
| Chopping content into tiny “chunks” for AI | No such requirement exists. Google’s systems can understand pages with several topics and show the relevant part. |
| Rewriting content in a special way for AI | You do not need to write differently. The systems understand synonyms and meaning, so you do not have to cover every long-tail variation. |
| Seeking inauthentic mentions across the web | It is not as helpful as it seems. Ranking systems focus on high-quality content, and other systems block spam. |
| Obsessing over structured data | It is not required for generative AI search and there is no special schema for it, though it remains useful for standard rich results. |
| Trusting tools that claim insider metrics | No third-party tool has access to Google’s internal ranking or AI systems. |
| Paying for guaranteed AI Overview placement | No one can guarantee placement. Treat such promises as a red flag. |
How to measure your performance
Google has been adding reporting for generative AI features. It states that pages appearing in AI features are included in your overall search traffic in Search Console, and it offers a Generative AI performance report that shows how your content appears in AI features on Search and Discover. Google has described insights such as impression metrics and which pages and countries appear in AI responses, and said in 2026 that it had rolled these features out to all websites worldwide. The exact metrics may expand, so check Search Console for what is available to you.
A simple measurement approach
- Record a baseline of impressions, clicks, and conversions for your key pages before you make changes.
- Use the Generative AI report to see which pages appear in AI features and how often.
- Track outcomes beyond clicks. Look at enquiries, calls, sign-ups and sales in your analytics tool. Some commentators claim visitors from AI features are more engaged, but that is not established, so check your own data.
- Spot-check results manually. Search a handful of your key questions and note whether an AI response appears and which sources it links to. Results vary by person, location and time, so treat these checks as clues, not proof.
- Change one thing at a time where possible and keep notes of dates.
- Review monthly, and expect fluctuations. AI features are changing quickly.
Be careful about drawing strong conclusions from small samples. Visibility in an AI Overview can come and go for the same query, and an impression is not the same as a visit or a customer.
Should you opt out of AI features?
Google has introduced a control in Search Console that lets site owners decide whether their site appears in and helps ground generative AI Search features such as AI Overviews, AI Mode and AI Overviews in Discover. Sites that opt out do not receive traffic or impressions from these features, and Google says the control is not used as a ranking signal for search results outside of them.
This is a business decision, and it depends on your goals. Here are the questions to weigh.
| Reasons you might stay in | Reasons you might opt out |
| You want the chance to be cited and linked in AI responses | You are concerned about your content being used to produce answers that reduce visits |
| Your business benefits from awareness even when people do not click | Your revenue depends heavily on clicks and you see AI features cutting them |
| Your Search Console data shows useful traffic or engagement from these features | Your content is licensed or paywalled in ways you want to protect |
| You want to be present as search habits shift | Legal, contractual or editorial policies require limiting reuse |
If you are unsure, review your own Search Console data before deciding, and revisit the choice as the features and reports develop. This article is general information and not legal advice, so consult a professional if contracts or rights are involved.
A 30-60-90 day plan
| Timeframe | What to do |
| Days 1 to 30: Foundation | Verify Search Console. Check indexing and snippet settings on your top 10 pages. Confirm your site’s generative AI setting. Fix crawling, duplication and speed problems. Update your Business Profile and Merchant Center details. |
| Days 31 to 60: Content | Choose three to five important topics. Rewrite or expand those pages with first-hand experience, original examples, clear structure, accurate dates and strong visuals. Add author information and sources. |
| Days 61 to 90: Measure and refine | Review the Generative AI report and your conversions. Note which pages gain impressions or engagement. Improve weak pages, merge thin overlapping ones and plan new, genuinely useful content based on customer questions. |
Page-level checklist for AI Overview readiness
| Check | Done? |
| The page is indexed and not blocked by robots, noindex or nosnippet | |
| Your site is set to appear in generative AI features, if that is your intent | |
| The page has a clear main topic and answers the key question early | |
| The content includes first-hand experience, original examples or data | |
| The page is not a rewrite of what competitors already say | |
| Headings and sections are clear and easy to navigate | |
| Likely follow-up questions are answered on the page | |
| Facts, prices, dates and policies are accurate and current | |
| Author, credentials and last-updated date are visible | |
| Credible sources are cited where claims need support | |
| Images and video are original, relevant and well optimized | |
| Business Profile and product feeds match the website | |
| The page loads quickly and works on mobile | |
| No thin, duplicate or near-identical pages target the same topic | |
| Performance is tracked in Search Console and analytics |
Preparing for AI agents
Google also points to a growing category of AI agents that act for people, such as comparing products or booking reservations. Some browser agents read a page by looking at its visual rendering, its structure and its accessibility information. Google suggests that, if it is relevant to your business and you have time, you review guidance on making sites easy for agents to use, and notes that emerging protocols such as the Universal Commerce Protocol may let Search agents do more in future.
For most small businesses, the practical step is the same as good usability and accessibility: clear page structure, labelled buttons and forms, and information that is easy to find. This area is still developing, so treat it as something to monitor rather than a priority to rush.
Common mistakes to avoid
- Chasing AI hacks instead of fundamentals. Fix indexing, content quality and page experience first.
- Mass-producing pages for every query variation. This risks violating Google’s spam policy on scaled content abuse.
- Publishing unedited AI content. Content with no original value is easy to replace and may fall foul of quality policies.
- Blocking snippets or crawling by accident. Check robots and meta settings after site changes.
- Ignoring the Search Console setting. Make a deliberate choice about whether your site appears in AI features.
- Relying on third-party tools’ claims. Use them only as aids and compare their advice with Google’s official guidance.
- Letting information go stale. Out-of-date prices, hours and policies hurt both users and trust.
- Expecting guarantees. No tactic ensures an AI Overview citation, and results change over time.
- Measuring only rankings. Track engagement and conversions as well.
Frequently asked questions
What are Google AI Overviews?
They are AI-generated summaries that Google can show at the top of search results for some queries, with links to pages that support the information. They are built using Google’s search index and ranking systems.
Can I optimize for AI Overviews?
Yes, in the sense of making your site eligible and your content genuinely useful. Google says there are no special requirements beyond standard SEO best practices, and that optimizing for generative AI search is still SEO.
Is there special schema markup for AI Overviews?
No. Google says structured data is not required for generative AI search and there is no special schema.org markup for it. Structured data remains useful for eligibility for standard rich results.
Do I need an llms.txt file?
Not for Google Search. Google says Search does not use such files, so they will not help or harm your visibility there. You may create one for other systems that use it.
What is the difference between AI Overviews and AI Mode?
AI Overviews are summaries shown within the normal results page. AI Mode is a more conversational experience for longer and more complex questions. Google says they may use different models and techniques, so the answers and links can differ.
What is query fan-out?
It is a technique where the AI generates several related searches from a single question to gather more information. It explains why a page that covers related subtopics helpfully may be more useful, though you should not create separate thin pages for each one.
How do I see if my site appears in AI Overviews?
Use the Generative AI performance report in Search Console, if it is available for your property, and check your overall search performance. Manual searches can add context but vary by person and time.
Can I opt out of AI Overviews?
Google has introduced a Search Console control that lets site owners choose whether their site appears in generative AI Search features. Opting out means no traffic or impressions from them. Consider the trade-offs for your business first.
Will AI Overviews reduce my traffic?
It depends on your topic, audience and content. Some queries may lead to fewer clicks, and others bring more engaged visitors. Measure your own data in Search Console and analytics rather than relying on general claims.
Does using AI to write content hurt my chances?
Not by itself. Google says AI-assisted content is acceptable if it meets Search Essentials and spam policies. The risk is publishing generic, unchecked content that offers nothing original.
Should I hire an AEO or GEO specialist?
Only with care. Google treats these as part of SEO, and advises evaluating third-party advice against its official guidance. Ask for clear, evidence-based methods and avoid anyone who guarantees AI placement.
Conclusion
Optimizing for Google AI Overviews is less mysterious than it sounds. Google’s own guidance says the foundation is the same as for the rest of Search: be indexed and eligible, create unique content with real experience and expertise, organize it well, support it with good visuals, keep your business information accurate and give visitors a good experience.
Ignore the hacks, from llms.txt files to content chunking and manufactured mentions, and spend your effort on pages that people would genuinely find satisfying. Then measure what happens in Search Console and your own analytics, and make decisions, including whether to remain in AI features at all, based on evidence. The tools and reports will keep changing, but a useful, trustworthy site will remain a sound investment.
✏️ Found an outdated tool detail or an error?

💬 Join the Technical Discussion (0 Comments)
No comments yet. Start the discussion.