You rank well on Google. Maybe you sit in the local pack for your main service, your reviews are solid, and that listing has kept the phone busy for years. Then one afternoon you open ChatGPT, type “best [your service] in [your town],” and watch it confidently recommend three businesses. None of them are you.
That is not a bug, and it does not mean your SEO is broken. It means the thing you optimized for and the thing answering that question are two different machines, built on two different ideas of what a good answer looks like. Here is what is actually happening, and why it lands hardest on local service businesses.
They are not playing the same game
Google crawls the web, indexes individual pages, and ranks them for a query using signals like keywords, links, and technical health. When you do local SEO, you are feeding that machine: your site, your Google Business Profile, your citations, your reviews. The output is a ranked list, and you can climb it.
ChatGPT does not work that way. It is not crawling and ranking pages in real time. It answers from two places: the static snapshot of the web it was trained on, which is already months old by the time you use it, and whatever it can retrieve in the moment when search is switched on. By OpenAI’s own description, ChatGPT search rewrites your question into targeted queries and pulls back results to build an answer. Then it does the part that feels like magic. It synthesizes one answer, a short list of names it is confident about, instead of handing you ten links to sort through yourself.
The local piece matters here too. When you ask for something “near me,” ChatGPT uses your approximate location to localize the search, turning a vague request into something closer to “best plumber in [your city].” So it is absolutely answering local queries. The only question is whether it has enough trustworthy information to name you when it does.
That word, confident, is the whole story. Google can rank you without trusting you very much. ChatGPT will only name you if it is sure enough to put your business in a sentence and stand behind it.
It runs on consensus, not your homepage
Here is the part that trips people up. Google can know your business from your website alone. ChatGPT builds its confidence from agreement across many sources.
It is checking whether your website, your Google Business Profile, your directory listings, your reviews, and any mentions of you around the web all describe the same business in the same way. When those sources agree, you become a clear, real, recommendable entity. When you live mostly on your own site and nowhere else, or your sources contradict each other, there is no consensus to draw on, and the model quietly moves to a competitor it feels surer about.
A great website is necessary. On its own it is not enough, because one source is not a consensus.
Why local businesses go invisible
When I look at why a perfectly legitimate local business is missing from AI answers, it is almost always some mix of these:
- Inconsistent business details. Your name, address, and phone appear in slightly different forms across your listings. To you, “123 Main St.” and “123 Main Street, Suite B” are obviously the same place. To a model trying to verify you, they read like noise, and noise lowers confidence.
- No structured data. Your pages are plain text with no schema markup telling machines what you are, what you do, and where you do it. Google’s own documentation explains that it uses structured data to understand a page and to gather information about the companies described in it. Without that markup the model has to guess, and it would rather recommend a business it does not have to guess about.
- Copy written for vibes, not extraction. “Proudly serving the community with a commitment to quality” reads fine to a human and gives a machine nothing to lift. “We are a licensed plumber in [town] handling emergency repairs, water heaters, and repipes” can be quoted directly.
- A thin footprint and generic reviews. Few directory listings, few mentions, and a wall of “great service, would recommend” reviews leave the model very little to build on. Reviews that name the actual service, the neighborhood, or the tech who showed up are far more useful to it.
- Stale or missing knowledge. If you opened recently, changed your name, or moved, the training snapshot may not know you yet, or may have you wrong. I have watched ChatGPT shave years off a business’s history and list services it dropped ages ago.
None of these are about whether you are good at the actual job. They are about whether the data around your business is clear, consistent, and verifiable.
The good news, especially if you are small
It is tempting to assume the big franchise always wins here. It does not, at least not by default. These systems have a signal bias, not a size bias. They recommend the business with clear, consistent, well-structured information, and a focused local business can build exactly that on purpose.
In some ways this is a more level field than classic SEO. You are not trying to outspend a national chain on backlinks. You are trying to be the clearest, most consistent, most verifiable answer for your town, and that is a fight a local business can actually win. Because almost nobody in most local markets has done this work yet, the ones who show up early tend to stay there.
How to check where you stand
You can run the test yourself in about five minutes. Open ChatGPT, then Perplexity, then Claude, and ask each one the way a customer would: “Who is a good [your service] near [your area]?” or “best [your service] in [your town].”
Note three things: whether you appear at all, whether the businesses it does name are really your competitors, and whether anything it says about you is wrong. If you are absent, that is a visibility gap. If you are present but the details are off, that is a data accuracy problem. Both are fixable, and both trace back to the same root cause.
It is all the same foundation
The reassuring part is that none of this is a separate, exotic project with its own budget. The work that makes you visible to ChatGPT, consistent business information everywhere, real structured data, specific and answer-ready content, reviews with substance, is the same work that strengthens you in Google’s AI Overviews, in Gemini, in Perplexity, and in plain Google too.
This is what answer engine optimization actually is once you strip the jargon off it. You are not gaming anything. You are making sure a real business is described accurately and consistently enough that a machine can recommend it without guessing. Give the engines a clear, trustworthy, true picture of who you are and what you do, and the recommendations tend to follow. Leave them guessing, and they will hand the job to the business that did not.
Frequently asked questions
Why does my business rank on Google but not on ChatGPT?
Google ranks individual pages and can surface you from your website alone. ChatGPT only names a business when many independent sources agree about it, so if your details are inconsistent across listings or you have little presence beyond your own site, it has no consensus to draw on and recommends a competitor it is more sure about.
Can a small local business show up in ChatGPT?
Yes. These systems have a signal bias, not a size bias. They recommend whichever business has the clearest, most consistent, most verifiable information, and a focused local business can build that on purpose without outspending a national chain on backlinks.
Does ChatGPT use Google to find businesses?
Not directly. When search is enabled, ChatGPT rewrites your question into its own queries, uses your approximate location to localize them, and synthesizes one answer from the sources it retrieves. Either way the fix is the same: give the engines a clear, consistent, accurate picture of your business.
Sources
- OpenAI, How ChatGPT search works (query rewriting, approximate location, cited sources)
- OpenAI, Introducing ChatGPT search
- Google Search Central, Local Business (LocalBusiness) structured data
- Google Search Central, How structured data markup works
- Schema.org, LocalBusiness