How to Rank in AI Overviews and Get Cited by ChatGPT (2026)

More of your customers now ask a machine before they ask a person. Someone types “who does furnace tune-ups near Dayton” into Google and gets an AI Overview, or asks ChatGPT for a bookkeeper who works with trucking companies, and the answer comes with a handful of linked sources. You want to be one of those links.

Search for how to rank in AI Overviews and you’ll find a lot of confident advice: chunk your pages into tiny blocks, add an llms.txt file, use special schema, get mentioned everywhere. This year Google published its own guide to generative AI features, and it tells site owners to skip most of those tricks.

This article separates what Google, OpenAI, Perplexity and Microsoft actually say from the guesswork. Then it covers the steps that do help, and how to measure whether AI tools are sending you anyone.

Is ranking in AI Overviews different from regular SEO?

Short answer: Mostly no. Google says there are “no additional requirements to appear in AI Overviews or AI Mode, nor other special optimizations necessary.” A page has to be indexed and eligible to show with a snippet, and from there the same ranking and quality systems decide what gets cited.

Google’s AI features page explains that AI Overviews and AI Mode can use “query fan-out,” running several related searches across subtopics to build one answer. That helps explain why a page can be cited for a question it doesn’t rank first for: it may answer one piece of the question well.

Google’s 2026 optimization guide goes further and lists what you don’t need to do:

• create llms.txt or other new machine-readable files;
• break content into tiny chunks (“There’s no requirement to break your content into tiny pieces”);
• rewrite pages in a special style for AI;
• chase inauthentic mentions, which “isn’t as helpful as it might seem”;
• add special structured data, since “Structured data isn’t required for generative AI search.”

What it asks for instead is familiar: crawlable pages, a good page experience, clear headings, original content and accurate Business Profile and Merchant Center details.

Do answer engine optimization and GEO actually work?

Short answer: The useful parts are good SEO and clear writing under new names. Answer engine optimization (AEO) and generative engine optimization (GEO) describe making content easy for AI systems to find, understand and cite. The tricks sold under those labels are what Google now tells you to skip.

Google’s guide sums up its advice as: apply foundational SEO, create helpful content, and skip “AEO/GEO hacks” such as chunking, llms.txt and inauthentic mentions. It also warns that “No third-party tool has access to our internal ranking or AI systems,” so treat any tool claiming to see inside Google’s AI with suspicion.

That doesn’t make the work pointless. Other AI tools crawl and cite the web differently from Google, and a page that answers a question plainly is easier for every one of them to quote. The rest of this article covers those differences.

How do ChatGPT, Perplexity and Copilot find the pages they cite?

Short answer: Each uses its own crawler or index. ChatGPT search relies on OpenAI’s OAI-SearchBot, Perplexity on PerplexityBot, and Microsoft Copilot on Bing’s index. If those crawlers can’t reach your site, you’re unlikely to be cited there, however good the page is.

OpenAI’s crawler documentation is clear: “Sites that are opted out of OAI-SearchBot will not be shown in ChatGPT search answers.” OpenAI hasn’t published how much ChatGPT search also draws on outside providers; independent testing since its 2024 launch, such as Yoast’s write-up, has pointed to Bing’s index as a major source. Being well indexed in Bing is how you reach Copilot, and it probably helps with ChatGPT too.

Perplexity says PerplexityBot is “designed to surface and link websites in search results on Perplexity” and isn’t used to train AI models. Microsoft’s Bing Webmaster Tools now reports how often your pages are cited in Copilot, Bing’s AI summaries and some partner products, and Bing says it respects robots.txt preferences.

Which AI crawlers should you allow in robots.txt?

Short answer: Allow the search crawlers (Googlebot, Bing’s crawler, OAI-SearchBot and PerplexityBot) if you want to be cited. Training crawlers such as GPTBot and Google-Extended are a separate decision, and blocking them doesn’t remove you from Google Search or ChatGPT search.

Google says Google-Extended controls whether your content is used to train future Gemini models and for grounding in some of Google’s other AI systems, and that it “does not impact a site’s inclusion in Google Search nor is it used as a ranking signal.” It isn’t a separate crawler, only a robots.txt token. OpenAI’s GPTBot is its training crawler; disallowing it “indicates a site’s content should not be used in training.”

A robots.txt that stays visible in AI search but opts out of model training looks like this:

User-agent: OAI-SearchBot
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: GPTBot
Disallow: /

User-agent: Google-Extended
Disallow: /

The last two blocks are optional; leave them out if you’re happy for your content to be used in training. OpenAI says changes can take about 24 hours to take effect for search. Two crawlers behave differently: ChatGPT-User and Perplexity-User fetch pages when a person asks, and both companies say robots.txt rules may not apply to those user-initiated visits.

Watch out: Check your firewall, CDN and security plugin as well as robots.txt. A “block AI bots” setting can shut out the search crawlers too, and Perplexity notes that sites behind a web application firewall may need to allow its bots explicitly.

If you’d rather stay out of Google’s AI answers altogether, Search Console now has a Search generative AI control under Settings. Excluding your site removes it from AI Overviews, AI Mode and generative AI features in Discover without affecting regular Search. For finer control, the nosnippet, data-nosnippet and max-snippet settings limit what Google can show from a page.

How do you write content that AI answers cite?

Short answer: Answer the question directly at the start of the section, then back it up with specifics: prices, steps, examples, dates and first-hand detail. Use clear headings and tables, and keep facts current. That’s what makes a passage easy to quote.

Google’s guide asks for “non-commodity” content: a real point of view, first-hand experience and expert detail rather than a rewrite of what’s already ranking. Microsoft’s advice for Copilot citations is similar and more concrete: clear headings, tables and FAQ sections, claims supported by examples and data, current information, and IndexNow to tell participating search engines when pages change.

In practice, for a service page that means:

1. Lead with the answer. “We service gas and electric furnaces in Dayton and Kettering, usually within two days” beats three paragraphs about comfort.
2. Name the specifics. Service areas, what’s included, what isn’t, how booking works.
3. Show it’s yours. Photos of real jobs, the names and roles of the people who do the work, real examples.
4. Date what changes. Prices, rules and product features with “as of” dates.

Our take: Write every important section the way you’d answer a customer on the phone: the answer first, then the detail. We structure our own service pages answer-first. It helps people skimming on a phone as much as it helps any AI system, and it isn’t a trick; it’s just clearer.

How do you make your business clear to AI systems?

Short answer: Describe your business the same way everywhere: the same name, services, service areas, phone and address on your website, Google Business Profile, Bing Places, Apple Business and the directories that matter. Then earn genuine mentions from reputable sites that describe what you do.

AI tools stitch answers together from several sources, so contradictions hurt. If your site says you serve “all of Ohio,” your profile lists three counties and an old directory shows a different phone number, no system (and no customer) can be confident which is right. Google’s AI features page lists keeping Business Profile and Merchant Center information current among its best practices. Microsoft asks businesses to keep text, images and video consistent about the same entities, and local businesses to keep Bing Places current.

Structured data helps here even though Google doesn’t require it for AI features. Organization or LocalBusiness markup that matches the visible page states your name, address, phone and hours unambiguously. Our local SEO checklist has a minimal LocalBusiness example and the citation list, and our Google Business Profile guide covers the profile itself.

Mentions are where the “GEO” crowd and Google disagree most. Google says inauthentic mentions don’t help. Real ones still matter for the same reasons they always have: a local news story, a supplier’s dealer page or an industry association listing tells people, and the systems that read the web, that you’re a real business doing real work.

Do you need an llms.txt file?

Short answer: Not for Google. Google’s guide says an llms.txt file “will neither harm nor help your site’s visibility” in Search. Some AI agents and tools may read one, Chrome’s Lighthouse now checks for it, and it costs little to add, so it’s a reasonable extra rather than a priority.

An llms.txt file is a plain Markdown file at your site’s root (for example, example.com/llms.txt) that summarizes what the site is and links to its key pages. Lighthouse’s new, experimental Agentic Browsing category includes an llms.txt audit alongside checks on whether agents can use your accessibility tree and whether the layout stays stable. Chrome’s documentation says that without the file, “agents may spend more time crawling the site to understand its high-level structure.”

Our own site has an llms.txt file and scores 3 out of 3 in Lighthouse 13’s Agentic Browsing category. That’s a readiness check, not a ranking: it says agents can read and use the site, not that any AI tool will cite it. If your developer has an hour, add one. If they don’t, fix your crawler access and your content first. Our write-up on how we got our site to 100 in Lighthouse covers the rest of that work.

How do you measure traffic from AI Overviews and ChatGPT?

Short answer: In Search Console, AI Overviews and AI Mode are counted in the Performance report under the Web search type, and a separate Generative AI performance report shows impressions in those features. In Google Analytics, a new AI Assistant channel groups visits from chatbots such as ChatGPT, Gemini and Claude.

Here’s where each number lives as of October 2026:

• Search Console Performance report: clicks and impressions from AI Overviews and AI Mode are included in Web search totals.
• Search Console Generative AI performance report: impressions of your links inside AI Overviews and AI Mode, by page, country, device and date. Google rolled it out to all sites on August 31, 2026. If you don’t see it, you may not have enough AI impressions yet.
• Google Analytics: since May 13, 2026, visits whose referrer matches a recognized AI assistant land in an “AI Assistant” channel with the medium “ai-assistant.”
• Bing Webmaster Tools AI Performance: citations in Copilot and Bing’s AI summaries, with the grounding queries behind them. It counts citations, not clicks.
• Referral sources: look for chatgpt.com and perplexity.ai in your traffic sources. Google’s help pages include an example “AI assistants” custom channel group covering ChatGPT, Gemini, Copilot and Perplexity, placed above Referrals.

Add “ChatGPT or another AI tool” as an option on your “How did you hear about us?” question. Analytics can only label visits that arrive with a recognizable referrer, so asking catches the rest.

How to rank in AI Overviews: what to do this month

1. Check robots.txt, your CDN and your security plugin, and make sure Googlebot, Bing’s crawler, OAI-SearchBot and PerplexityBot can reach your site.
2. Verify your site in Bing Webmaster Tools as well as Search Console, and look at both AI reports.
3. Rewrite the opening of your five most important service pages so the answer comes first.
4. Make your name, services, service areas and phone number match across your site, Business Profile, Bing Places and Apple Business.
5. Add an AI option to your lead-source question and review the numbers monthly.

If you’d like a hand with the content or the technical side, our SEO and web development teams can help.