Straight answer: ChatGPT, Perplexity, Gemini and Google’s AI Overviews are already answering questions about your business, your prices, your opening hours and whether you’re even still trading, and most small business owners have never once checked what these tools are saying. You can influence it, you cannot fully control it, and the fix has almost nothing to do with the SEO advice you were given three years ago.
I asked ChatGPT who I was, and it got it wrong
Back in early 2025 I typed my own name into ChatGPT out of curiosity, the way you’d Google yourself after a few glasses of wine. It told a stranger asking about UK marketing consultants that I was based in Manchester. I’ve never lived in Manchester. It also referenced a book I haven’t written, mashed together with a real one I have. Nothing malicious, no scandal, just quietly, confidently wrong, the way a colleague who half remembers a story tells it back to you as fact.
That’s the bit that should worry you more than it does. This wasn’t a troll leaving a bad review you could reply to. There was no comment box, no way to flag it in the moment, no customer service number for “the AI invented something about my business.” It just sits there, feeding slightly wrong information to anyone who asks, and you’d never know unless you went looking.
Multiply that by every small business owner who has never typed their own company name into an AI chat tool, and you start to see the scale of the problem nobody’s talking about in the usual “10 AI tools for small business” listicles.
Why this matters more than your ranking on page one of Google
For twenty years the game was: rank on Google, get the click, convert the visitor. That game still exists, but a growing slice of searches never produce a click at all. Adobe’s own analytics data showed AI-referred traffic to US retail sites jumped by well over 1,000% year on year during the 2024 holiday shopping season, and that curve has only kept climbing through 2025 into 2026. People are asking Perplexity “who’s the best accountant near me” or asking ChatGPT to compare three local web designers, and getting an answer, a recommendation, sometimes a full comparison table, without ever visiting a single website.
Your customer’s first impression of your business might now be a paragraph an AI model generated about you, built from whatever scraps of information it could find, and you had no say in the wording.
This is the part that separates the businesses that will quietly win the next few years from the ones that won’t. The winners aren’t necessarily the ones with the flashiest website. They’re the ones who are correctly, clearly, boringly described everywhere an AI model might look.
The uncomfortable bit most marketing blogs skip
Here’s what nobody selling you an “AI SEO package” wants to say out loud: you cannot buy your way to being cited correctly. Backlinks, the old currency, still matter for traditional search, but AI answer engines lean far more heavily on consistency and clarity across the open web, on structured, plain-language content, and on trusted third-party sources, than on how many sites link to you. An agency charging you £800 a month for “50 high authority backlinks” is selling you 2019’s answer to a 2026 question.
And there’s a second uncomfortable truth: even if you do everything right, these models are trained on snapshots of the internet from months ago. Fix a wrong address on your site today, and some AI tool might still be telling people you closed down for another six months, because it hasn’t been retrained or hasn’t recrawled that page yet. There is no complaints department. You just have to keep at it, check regularly, and accept a bit of lag as the cost of doing business in 2026.
A simple audit you can do this afternoon
I run this check on my own name and business roughly every six weeks now. It takes about twenty minutes.
- Open ChatGPT, Perplexity, and Google Gemini in separate tabs
- Ask each one: “What does [your business name] do?” and “Who is [your name]?”
- Ask each one: “Who are the best [your industry] in [your town/city]?” and see if you’re even mentioned
- Ask: “Is [your business name] still trading?” if you’ve moved, rebranded, or been quiet online for a stretch
- Write down every factual error: wrong address, wrong prices, wrong services, wrong dates, wrong people
The first time I did this, I found my old office address still floating around, a service I’d stopped offering two years earlier listed as current, and a rounding error on a stat I’d once quoted that had been repeated by three separate AI answers as if it were gospel. None of it was dramatic. All of it was fixable.
What moves the needle
Once you know what’s wrong, the fix isn’t a redesign. It’s discipline.
- Fix your Google Business Profile first. Address, hours, phone number, services. A shocking number of AI answers still lean on this data, and it’s the single fastest correction available to you.
- Write your website in plain question and answer language. If a customer would ask “how much does X cost” or “how long does Y take,” answer that exact question in a short, clear paragraph somewhere on your site. AI models are essentially looking for a clean, quotable answer, the same instinct behind a good interactive calculator that answers a specific question directly instead of making a visitor dig for it.
- Get named accurately in a handful of credible places. A press mention, a directory listing, an industry roundup, a Wikipedia reference if you’re eligible. Quality and correctness beat quantity every time.
- Keep your “about” information identical everywhere. Same business name, same address format, same founder bio, on your site, your LinkedIn, your Companies House listing, your directory profiles. Inconsistency is what causes an AI model to guess, and guessing is where the Manchester mix-up came from in my case.
- Re-check quarterly, not once. This isn’t a project with an end date. Treat it like checking your bank statement.
What this looks like when a brand gets it right
Look at how brands with consistent messaging behave everywhere, not just on their own site. Ahrefs built a marketing strategy around being repeatedly, boringly, correctly cited as the source on SEO data, and that consistency is exactly what makes them easy for any AI model to summarise accurately. Compare that with how Disney has kept its brand story identical across a hundred years of channels, so there’s never a mismatch for anything scraping the web to trip over. Small businesses can copy the principle even without the budget: say the same thing about yourself, everywhere, in plain language, and keep saying it.
Headspace’s marketing has always led with a single clear promise rather than a shifting list of features, which is precisely the kind of single, quotable idea an AI answer engine can lift cleanly. And GoDaddy’s positioning as the small business champion has stayed consistent enough for years that it shows up the same way no matter where you ask about them. None of these brands are leaving their AI reputation to chance, and neither should you.
The Richard Thaler angle nobody applies to this
There’s a behavioural point buried in all of this that ties back to Richard Thaler’s work on how people make decisions, not how we assume they do. Thaler’s research shows people take the path of least cognitive effort, they anchor on the first answer they’re given rather than digging for a second opinion. If an AI tool gives someone a confident, clean answer about which local plumber to call, most people won’t cross-check it against three websites. They’ll ring the number they were given. Being the accurate, well described default answer isn’t a nice-to-have anymore, it’s the whole ballgame for a chunk of your future customers.
When it’s worth getting help
If you’ve run the audit above and found a mess, or you simply don’t have the time to chase this every quarter on top of running your business, this is one of the areas where bringing in outside help pays for itself, because the cost of being wrongly described to a customer who never even lands on your site is invisible until you go looking for it. I cover exactly this kind of work with clients through hands-on AI consultant support for small businesses, walking through where your business currently sits in AI answers and fixing the gaps methodically rather than guessing.
Whether you do it yourself or bring someone in, the point stands. Your business already has an AI reputation whether you asked for one or not. The only question is whether you’re the one shaping it, or whether you’re leaving it to whatever half-remembered scraps the internet happened to serve up.
Frequently asked questions
What is GEO and is it different from SEO?
GEO stands for generative engine optimisation, the practice of making your business easy for AI tools like ChatGPT and Perplexity to describe accurately. It overlaps with SEO but leans more on clear, consistent, plain-language answers and correct third-party mentions than on backlinks and keyword density.
Do I need to rebuild my website to fix this?
No. Most fixes are additions and corrections, not a rebuild. Update your Google Business Profile, add clear question-and-answer style paragraphs to key pages, and make sure your business details match everywhere online.
How long does it take for AI tools to show corrected information?
There’s no fixed timeline and it varies by model. Some tools pull live data and update within weeks, others rely on training snapshots that can lag by several months, which is exactly why checking quarterly matters more than fixing it once.
Does this replace traditional SEO work?
No, it sits alongside it. Google search still sends real traffic, and good SEO fundamentals still help. Think of AI reputation checks as a new, additional layer, not a replacement for the work you’re already doing.
Related reading: I Asked AI to Audit My Own Blog. It Told Me to Delete Half of It. and I Write Client Proposals With AI in 20 Minutes. Here’s the Bit Nobody Tells You.
Want the complete version? Read where I break down AI marketing.
