Why doesn't my business show up in Google AI Overviews?

If you have searched for your own kind of business and watched Google's AI name your competitors but not you, that stings. You paid for the website. You did the right things. And the answer box at the top of Google acts like you do not exist.

The rest of this post explains what is happening in plain terms and gives you a clear place to start this week. One honest note first: our dataset is young, enough to see patterns, not enough to declare rules. We will keep publishing as it grows.

What is a Google AI Overview, and why does it matter to you?

Search Google for something like "physio for back pain near me". On many searches, the first thing on the page is no longer a list of websites. It is a paragraph written by Google's AI that answers the question directly and names a few businesses and sources. Google calls this an AI Overview.

When that paragraph names a business, we say the business is cited. Cited just means mentioned, by name or by link.

Here is why it matters. That paragraph sits above everything else. Many customers read it, pick from the businesses it names, and never scroll further. Being missing from it means losing customers you never knew were looking.

This is no longer a niche feature. Google has folded its AI answers into the heart of search, and more than 2 billion people use them every month.

So why isn't my business in there?

We test this directly. For each business in our dataset, we ask around 10 questions a real customer would ask, in the major AI tools and in Google, and record who gets named.

Two things stood out.

Most businesses are either in or out. About a third of the businesses we tested were never named in a single Google AI answer, across all their questions. The businesses that were named tended to be named again and again. There was little middle ground. The other AI tools, like ChatGPT and Claude, behaved differently: almost every business got named somewhere, at least occasionally.

Google's door is the one worth working on. When Google's AI named a business, the chat tools almost always recommended that business too, on the very same question. It barely worked the other way around. Being known to ChatGPT did not get businesses into Google's answer.

In plain terms: there is one hard door and several easy ones, and the hard door is attached to the place your customers already search.

I know how this lands if you are in the "out" group. The good news is what comes next.

"But my SEO is fine." It probably is. That is the point.

SEO is the work of making your website easy for Google to find and understand. The basics look like this: your site is secure (the padlock next to your web address), it works well on a phone, and Google can find and read your pages.

Here is what surprised us: that work did not separate the mentioned businesses from the missing ones. Almost everyone in our testing had the basics done, in both groups. Their technical health scores were nearly identical.

So if your web person says your site is in good shape, believe them. The basics are the entry fee. They get you considered. Something else decides who gets mentioned.

One more myth while we are here: the missing businesses actually published more blog posts than the mentioned ones. Writing more content was not the answer either.

What did the mentioned businesses have?

Two things, and neither is technical.

A steady base of Google reviews. These are the star ratings on your Google business listing, the panel that appears when someone searches your business name. The mentioned businesses had two to three times more Google reviews than the ones left out. A separate 2026 study of local searches found the same pattern. Google has never said reviews are part of how its AI picks, so we treat this as a strong lead, not a law. But reviews are public proof that real customers chose you, and that is exactly the kind of evidence an AI answer can lean on.

Pages that answer questions in plain words. Most of the mentioned businesses had what we would call question-and-answer content: a page that asks "How much does a session cost?" or "Do you work weekends?" and answers it simply, the way you would on the phone. Most of the missing businesses did not.

Think about what an AI answer actually is: a direct reply to a customer's question. A website that already answers those questions, in the customer's own words, gives the AI something to quote. A website that only says "Welcome to our homepage" does not.

Who should you trust for advice on this?

This space moves fast, and that creates a real burden for any business that relies on Google for leads. Even the experts' favourite statistics have changed within months. So here is a simple filter for any advice you hear, including ours.

1. What Google says itself. Surprisingly little, and that is useful. Google's own guidance says: there are no special requirements, no magic file or code to buy, keep your Google business listing up to date, and publish things only you could write. If someone sells you a special "AI optimisation file", Google's own documents say you do not need it.
2. What the industry believes. Studies and expert claims are useful weather reports, but they change fast. Treat them as direction, never as guarantees.
3. What gets measured. Our numbers come from real testing, and we always tell you how many businesses were behind them. Any claim that arrives without a sample size deserves a polite raised eyebrow.

If you pay an agency for SEO, three fair questions: What changed in Google's AI guidance in the past six months? Which of your recommendations does Google confirm, and which are educated guesses? How are we measuring whether AI tools mention us? A good partner will enjoy answering. A stale one will wobble.

Where would I start this week?

In this order. The first three cost nothing but time.

1. See it for yourself. Search Google for your service and your town, the way a customer would. Read who the AI answer names. Ask ChatGPT the same question. Now you know which group you are in.
2. Update your Google business listing. Hours, services, photos, phone number. This is the one concrete thing Google itself recommends for AI visibility, and most listings are quietly out of date.
3. Start a review habit. Ask every happy customer for a Google review, right after the job, every time. Steady beats burst. If you are at a handful of reviews, this is the single biggest gap our data points to.
4. Put your phone questions on your website. Write down the five questions every customer asks you. Answer each one on your site in the same plain words you use out loud. This helps customers regardless of what Google does with it.
5. Spend nothing on AI magic. No special file, code, or paid trick gets you into AI answers. Google says so in its own guidance. Put that budget into reviews and answers instead.

Pick one and start. Small businesses do not lose this game by moving slowly. They lose it by guessing.

The bottom line

Google's AI answer is the new front door, many small businesses are not getting through it, and the fix our early data points to is refreshingly unglamorous: earn visible proof from your customers, and answer their questions in plain words on your site.

Now the part where we eat our own cooking. GetRecommended.io is in the missing group of its own dataset. We launched recently, our review base is thin, and Google's AI does not mention us yet. So we are running every step above on ourselves and with our beta group, in the open, and we will publish what moves and what does not.

If you want to know which group you are in without doing the manual checks, the free scan at getrecommended.io/scan shows you whether the major AI tools mention your business today. Run it, find your biggest gap, and fix one thing this week.

Related reading: why a business can rank on Google and still not be recommended by AI, how reviews affect AI recommendations, and how to check if your business shows up in AI search.