What is AI visibility, and how do you get your business named in AI answers?

If you run a company that’s been doing good work for years and you’ve got a feeling buyers are asking ChatGPT who to hire and getting somebody else’s name, you’re probably wondering what AI visibility really is and whether you can do anything about it. You can. Here’s the short answer, and then the long one, with the sources so you can check me.

What does AI visibility actually mean?

AI visibility means your business gets named, quoted or linked when someone asks an AI tool a question your company should be the answer to. It’s counted per question and per tool, so a company can be visible in Perplexity for one service and missing from ChatGPT for the same service on the same day.

The way I think about it is pretty simple. Somebody hurt in a car wreck looking for a lawyer, a family shopping for a custom home builder, an operations manager looking for a charter jet company, they all used to type three words into Google and scroll. Now a lot of them type the whole question into ChatGPT or read Google’s AI answer at the top, and they get back a few names with sources attached. If your name is in that short list, you start the call already trusted. If it isn’t, you never hear about the deal at all.

So AI visibility has two parts I track separately. One is the mention, where the tool says your company’s name in the answer. The other is the citation, where the tool links to your page as a source. Both count, but a citation is worth more because it sends the buyer to a page you control and tells the model your page is the one holding the facts.

How do AI tools decide which businesses to name?

AI tools mostly name the businesses they find in a live web search they run right before answering, and then they quote the pages that answer the question most cleanly. A smaller part comes from what the model already knows about brands from its training, and that part gives known brands a head start.

The search step comes first

The clearest study I’ve read on this is from Mahe Chen and Nick Koudas at the University of Toronto. They looked at how ChatGPT picks sources and found that 100% of the 22,315 URLs ChatGPT cited had been returned by its own searches first. Nothing got cited that the search step didn’t fetch. The same paper found ChatGPT splits a question into about 4 keyword-style searches and adds the current year to 23% of them, so a buyer’s question about “best dental implant practice near me” turns into several smaller searches you need to show up in.

Rank inside those searches matters a lot too. The Toronto study found ChatGPT cites 43.8% of the pages it ranks first and only 3.5% of pages ranked 11 to 20. And a page that comes back from several of the searches had 10.8 times the odds of being cited, which tells me the pages that win are the ones covering the whole question, not one narrow slice of it.

Then the model’s memory gives some brands a head start

The same researchers found that when a brand was already in ChatGPT’s prior knowledge, ChatGPT wrote that brand’s name into its searches 67% of the time, against 11% when it wasn’t. Websites the model already knew had 1.60 times the odds of being cited at equal rank. That’s the part that worries established companies, and it should, because a competitor who’s been talked about more online gets searched for by name before your page is even considered.

But I’d read that as a reason to get moving. The prior is built from what’s written about you across the web, and that’s something you can add to.

Is AI visibility the same as ranking on Google?

AI visibility overlaps with Google rankings but it isn’t the same thing, and the overlap has been shrinking. A page-one ranking still helps a lot, but plenty of AI citations go to pages that don’t rank well, and plenty of top-ranked pages never get quoted.

Search Engine Journal reported on an Ahrefs study of 863,000 keywords and 4 million AI Overview URLs that found only 38% of pages cited in Google’s AI Overviews also ranked in the top 10 for the same search. In Ahrefs’ July 2025 study that number was 76%. And 31% of the citations came from pages ranked beyond position 100.

ChatGPT is its own animal. A 2025 Seer Interactive study of about 100 questions found more than 87% of ChatGPT search citations matched Bing’s top 20 results, against 56% for Google. The Toronto paper, looking at ChatGPT’s own rewritten searches, found only 0.5% of the pages ChatGPT retrieved were in Bing’s top 10. Those two findings point different ways because they measured different things on different samples, and I’d rather tell you the studies disagree than pick the one that sounds better. The practical takeaway I use is to be indexed and healthy in both Google and Bing, and to stop assuming a Google rank report tells you where you stand in AI answers.

Why does it matter if AI cites my company?

It matters because AI answers now sit on top of a big share of the questions buyers ask, and the brands named in those answers appear to get more of the clicks that are left. The research shows direction, not proof, but the direction is consistent.

Washington University in St. Louis researchers found Google showed an AI Overview on 13.7% of all queries and on nearly 65% of question-form queries, with about eight cited sources per overview. Buyers who are close to hiring someone tend to ask in full questions, so that 65% figure is the one I watch.

Seer Interactive looked at more than 3,100 queries from June 2024 to September 2025 and found brands cited in AI Overviews got 35% higher organic click-through and 91% higher paid click-through than brands that weren’t cited. Seer’s own study lead said they can’t prove citation causes the higher click rates, and I’d say the same. Still, if your company is a $10M-plus firm where one client is worth thousands of dollars, I wouldn’t want to bet the other way.

What makes a page more likely to get cited by AI?

Pages get cited more often when they answer the buyer’s exact question near the top, carry specific facts a model can quote, stay updated, load their words without JavaScript, and get backed up by mentions on other sites. There’s no trick in any of that, it’s the plain work of making a page the most quotable answer on the subject.

Answer first, with specifics

Most service pages I look at open with a headline that could belong to any competitor and a paragraph of claims with no numbers in it. A model can’t quote “we deliver excellence.” It can quote your price range, your timeline, your credentials, the number of homes you built last year, if you put them on the page and you approve them. That’s why every page I rebuild leads with a direct answer and then backs it with facts the client supplies. I go deeper on this in how to get your business cited by ChatGPT and in how to rank in Google AI Overviews.

Keep pages fresh

A 2026 Seer Interactive study of 7,683 dated pages found 73% of pages ChatGPT cited and 78% of pages Gemini cited had been updated within the past year. AirOps found pages that went more than a year without updates saw their citation likelihood drop by over 50%. A page you wrote in 2021 and never touched is quietly aging out of the answers.

Make sure the bots can read it

Vercel reported that none of the major AI crawlers currently render JavaScript, naming GPTBot, ClaudeBot and PerplexityBot. If your service descriptions only show up after a script runs, or they’re tucked into tabs and accordions that load late, those crawlers may be reading an empty page. Vercel also found ChatGPT spent 34.82% of its fetches on 404 pages, so broken links and dead URLs waste the little attention you get.

Get talked about off your own site

Searchable, summarizing an Ahrefs study of about 75,000 companies, reported that branded web mentions correlated with AI visibility at 0.664 against 0.218 for backlinks. A digest of AI search citation research from Metavert puts the share of citations going to sites the brand doesn’t own at 85.7%. Those are correlations and digests, so I treat them as a direction. The direction is that directories, review profiles, industry listings and press about you matter a lot, which is why the bigger packages I offer include cleaning those up.

What doesn’t move AI visibility?

Special AI files, extra schema and keyword tricks don’t show a reliable effect in the studies I’ve read. Google says outright you don’t need any of them, and the independent data mostly agrees.

Google’s own guide to AI features says there are no additional requirements to appear in AI Overviews or AI Mode, and that you don’t need new machine-readable files, AI text files or markup. SE Ranking looked at nearly 300,000 domains and found no correlation between having an llms.txt file and how often a domain got cited. Ahrefs tested 1,885 pages that added schema against 4,000 that didn’t and found a small, statistically significant 4.6% decline in AI Overview citations, with AI Mode and ChatGPT changes too small to separate from zero.

I still add schema to every page I build, because it helps search engines understand who you are and what each page is about. I just don’t sell it as a citation lever, because the data doesn’t back that up. And the Metavert digest cites a July 2026 survey of 45 GEO studies that concluded no reviewed technique shows a stable, cross-platform causal effect. Anyone selling you one magic tactic for AI visibility is ahead of the evidence. What holds up is the boring stuff, good pages that answer real questions, kept current, on a site the bots can read.

How do you measure AI visibility?

You measure AI visibility by picking a fixed list of real buyer questions, running each one several times in each AI tool on a set schedule, and counting how many times your company is named or cited. Then you compare against your own starting point, rather than some vanity score a tool made up.

Running each question more than once matters more than people think. The Metavert digest cites an April 2026 study finding roughly 65% of the sources an engine cites for a query change from one day to the next. One screenshot proves almost nothing. So the method I put in my contracts runs every question three times in ChatGPT, Claude, Perplexity, Gemini and Google AI Overviews, and a question counts as a win when the client is named or cited in at least 2 of the 3 runs in at least one tool.

For Microsoft’s side there’s a free first-party report. Bing Webmaster Tools added an AI Performance report that shows total citations, average cited pages, the grounding queries the AI used to find your content, and citation counts per URL. If you haven’t set up Bing Webmaster Tools, do it this week, it costs nothing and it’s the closest thing to real AI citation data you can get for your own site.

The same churn is also the hopeful part. If the cited sources shift that much day to day, the list isn’t locked, and a company that publishes the clearest answer can work its way in.

What does it cost to wait on AI visibility?

I’m not going to tell you AI will wipe out your business next quarter. Plenty of established companies still get most of their leads the way they always have. What waiting costs is slower and quieter than that.

Every month a competitor’s pages get cited and yours don’t, the model sees more about them. The Toronto study shows brands already in ChatGPT’s memory get searched for by name 67% of the time against 11% for the rest, so that gap can widen while you sit still. Your older pages keep aging, and AirOps found citation likelihood drops by over 50% once a page goes more than a year without updates. And the questions that trigger AI Overviews most are the full-sentence ones buyers ask when they’re close to deciding.

The worst case, a firm that drops out of every answer in its market, is rare. The common case is a firm that’s still fine on Google and slowly stops being mentioned where the shortlist gets made. Either way the fix is the same, and it starts with knowing where you stand. Run five of your buyers’ questions through ChatGPT and Perplexity today and write down who gets named.

How I know this works

I’ve watched this happen on a business I own. My own ecommerce brand’s Bing AI citations went from 9 a month in April 2026 to 1,275 a month in August 2026, measured in the Bing Webmaster Tools AI performance report, with 63 of its pages cited by AI. That’s one site in one market, and it also runs paid ads, so I don’t claim every bit of its growth came from the pages.

On the Google side, when I checked rankings live in July 2026, the brand ranked #1 on Google for the “best” search for its product and Google’s AI showed that product first. 18 of its 22 target search terms were in Google’s top 3, and Search Console’s 3-month average up to June 2026 showed 400+ search terms on page one. You can see the results and how each one was measured on my homepage. Hailder House is new and has no finished client builds yet, so these come from my own earlier work, they aren’t typical results, and yours will be different.

What are your options for improving AI visibility?

There are four realistic ways to go at this, and the right one depends on your budget, your team and how big your site is. Here’s how I’d lay them out honestly.

If you’re under 500 organic clicks a month, I’d start in-house or with a tracking tool, because my guarantee can’t be measured on a base that small and I’d rather tell you that now.

Where does AI visibility fit with the rest of your marketing?

AI visibility sits on top of three other things, and it falls apart without them. The pages have to be written for people first, which is what helpful content really means. They still need a search foundation, because the AI tools search before they answer, and I lay that out in whether SEO is dead because of AI. And once a buyer clicks through from an AI answer, the page has to turn them into a call, which is what conversion rate optimization covers.

Inside this AI visibility guide, start with getting cited by ChatGPT if most of your buyers use ChatGPT, or ranking in Google AI Overviews if your traffic still comes mostly from Google.