Your customer's first impression is now written by AI. Does it have one story to tell?

Before a potential client ever lands on your website, there's a good chance an AI has already described you to them. Whether that description is accurate depends on something most brands haven't looked at yet.

My colleague Paton was the first to point it out. AI tools had started recommending some of our clients by name, and once he mentioned it, I saw it everywhere, including in the monthly reports I put together for clients. That answer is now the first touchpoint in a customer's journey, and the business doesn't write it.

In this article, I share what I've learnt about generative engine optimisation so far, how we've started using it for our clients, and five simple steps you can take with your own brand.

What is generative engine optimisation?

Generative engine optimisation, or GEO, is the process of making your business easy for AI tools to find, understand and recommend. Traditional search engine optimisation (SEO) is about earning a spot in the rankings, while GEO is about getting your name into the actual answer.

Let's use a homeowner with a burst pipe as an example. When they search for "plumber near me", Google returns a list of websites that offer plumbing, and ideally yours appears on the first page. SEO is the method you'd use to get to the top of that list. The same question in ChatGPT, however, returns no list at all. It returns an answer, something like: "Two plumbers with strong reviews in your area are Bob and Anthony. Bob offers call outs 24 hours a day." GEO is the work of becoming Bob, the business the AI names in its response.

Why now

People are already changing how they search. Pew Research Center tracked the browsing of 900 US adults and found that people who saw an AI summary on Google clicked through to a website in 8% of visits. People who didn't see one clicked through in 15% (Source: Pew Research Center, July 2025). For a lot of people, the answer has become the destination.

That doesn't make SEO redundant. While AI search is becoming far more popular, SEO still lays the groundwork that GEO needs to work well. In July 2025, Ahrefs found that 76% of the pages cited in Google's AI Overviews also ranked in Google's top 10 for the same search. When the team repeated the study in early 2026, that figure had dropped to 38%, although Ahrefs notes that better tracking explains some of the change (Source: Ahrefs, reported by Search Engine Journal, March 2026). So ranking well still helps AI tools cite you, but it no longer guarantees it.

The real insight: one story across the whole funnel

The biggest thing I've learnt is how much GEO depends on everything else. AI tools don't only read your website. They pull from your LinkedIn page, your Google Business Profile, directory listings, reviews, articles and anything else they can find about you. When all of those tell the same story, the AI has something solid to repeat. When they don't, it hedges, gets it wrong or leaves you out entirely.

I saw how much this changes the buying journey when I went looking for a new laptop recently. I used AI to help me make the final call, and I had a lot to weigh up: the specs, the price, whether to buy on contract or pay cash, and whether a second hand MacBook Pro made more sense than a new MacBook Air. The AI came back with considerations I hadn't thought of, along with reviews. Finding the options, understanding them and comparing them all happened in one conversation. In sales funnel terms, the first three stages collapsed into a single question and a single answer.

That doesn't make your other touchpoints any less important. A friend's recommendation, a LinkedIn post, your website, the first call with a salesperson and the welcome pack once someone signs all still shape the experience, and in many cases the AI feeds off them. This is how I see it playing out at each stage:

Stage	Where GEO plays a role	Other touchpoints	How they feed the end result
Awareness	The AI decides whether to name you at all	Word of mouth, events, LinkedIn posts, a billboard	Every mention gives the AI more to work with and builds recognition
Interest	The AI describes you in its own words	Your website, social profiles, a webinar	A consistent description keeps the AI's summary accurate
Consideration	The AI compares options and often recommends one	Brochures, case studies, a showroom visit	Your proof points become the reasons the AI recommends you
Intent	The AI acts as the trust check	Reviews, the first phone call, a reply to an enquiry	A good first conversation confirms what the AI promised
Purchase	An indirect role, since the AI has already set expectations	The quote, the contract, checkout or booking	A smooth close keeps the promise and earns loyalty
Loyal customer	Feeds the next person's AI answer	Onboarding, support, a follow up call	Reviews and referrals loop back to awareness

The last row is the one I find most interesting. Your loyal customers' reviews and referrals become the raw material for the next person's AI answer, so the end of one customer's journey shapes the start of the next.

What we've seen with our clients

Two things have changed in how I work. The first is FAQ sections, built properly. Every article I write for a client now ends with questions in the words a real person would use, each with a short, direct answer. I also add FAQ schema to the page. Schema is a small piece of code that labels those questions and answers, so search engines and AI tools can read them easily.

The second matters more to me. We care a lot about making sure a client's story is the same everywhere it appears. Their website, social profiles, directory listings and articles should all describe who they are and what they do in the same way. When someone asks an AI about them, we want the answer to be accurate.

The difference shows most with our medical clients. For a healthcare practice, patient education is both the most effective and the most compliant way to reach people.

Answering the questions patients already ask, clearly and honestly, builds trust without making promises. As more people turn to AI tools for the kind of questions they used to type into Google, we've noticed our medical clients showing up in AI responses more often. Clear answers to real patient questions seem to be exactly what these tools draw on.

Five simple steps to start with

You don't need a big project to start thinking about GEO in your brand messaging and content strategy. These are the steps I'd suggest:

• Ask the AI about yourself. Search for your business by name in ChatGPT, Perplexity and Google. Then ask for the service you offer without naming yourself, and compare the two answers

• Check that your story matches. Read your website, LinkedIn page, Google Business Profile and directory listings side by side. Do they describe you in the same way?

• Answer real questions. Write down the questions customers ask you most, and answer them clearly on your website, ideally in an FAQ section

• Show your sources. Back up your claims with real statistics and link to where they came from

• Keep your SEO going. Good rankings still give AI tools a reason to find you

Frequently asked questions

Naturally, this is a GEO article, so let's get into some FAQs.

What is generative engine optimisation?

Generative engine optimisation, or GEO, is the practice of making your business easy for AI tools like ChatGPT, Perplexity and Google AI Overviews to find, understand and recommend. SEO aims to rank a page in a list of links, while GEO aims to get your business named in the answer itself.

Does GEO replace SEO?

Not right now. Good SEO still helps AI tools find your content, although Ahrefs research shows that a high Google ranking is a weaker signal for AI citations than it was a year ago. The two work best together.

How do AI tools decide which businesses to recommend?

No AI company publishes its exact formula. Princeton University researchers found that adding citations, quotations and statistics to content improved its visibility in AI answers by up to 40%. AI tools also draw on information from across the web, so a consistent story across your channels helps.

Why does GEO matter for medical practices?

Patient education is one of the most effective and most compliant ways for a practice to be found. Clear, factual answers to the questions patients already ask build trust, and they are the kind of content AI tools tend to draw on.

(Source for the Princeton finding: Aggarwal et al., "GEO: Generative Engine Optimization", Princeton University, 2023)

Every touchpoint counts, including the ones AI writes for you

Think back to the last thing you bought after a bit of research. Chances are you trusted the option where everything you read lined up. GEO rewards the same thing. You can't write the AI's answer for it, but you can make sure it finds one clear, consistent story about you, with an experience behind it that lives up to that story.

At Kaleidoscope Consulting, we help teams see their brand the way a customer does. Our Customer Journey Mapping workshop maps every touchpoint, from the first answer an AI gives to the last handover, so you can spot where your story breaks before your customers do.

Explore our workshops or book a consultation to talk it through with our team.

By Nicole Crawford . Brand strategist