Last Updated on July 28, 2026 by Becky Halls
Here’s a quick thought experiment: type your best-selling product into ChatGPT and ask it to recommend the top five options in your category. Are you on the list? If not, you’ve just discovered the single biggest visibility gap in marketing right now – and no amount of ranking #1 on Google will fix it on its own.
Welcome to Generative Engine Optimization, or GEO: the practice of getting your brand quoted, cited, and recommended inside AI-generated answers rather than in the ten blue links underneath them. And it’s not a fringe concern anymore. AI search engines – ChatGPT Search, Gemini, Perplexity, Copilot and friends – now field an estimated 12–18% of English-language informational queries, up from under 2% a year earlier. That’s a land grab happening in real time.
GEO isn’t a new religion — it’s SEO growing up
Before you panic and rip up your strategy, take a breath. The people who’ve actually spent 2026 in the trenches are refreshingly calm about all this. As SEO Strategy consultant Becky Halls put it in one of her recent posts: “I don’t really care what optimizing for AI Search is called: SEO, AI Search Optimization, AEO, GEO, or whatever the new acronym ends up being. I just adapt to the terminology clients use if it helps them understand the opportunity and get internal buy-in.”
That’s the right energy. GEO is less a brand-new discipline and more the natural evolution of things you already know – good content, clear structure, trustworthy sources – pointed at a new kind of reader who happens to be a large language model.
The reason it matters now rather than “someday” is the click math. Ahrefs analysed 300,000 keywords and found that pages losing visibility to AI Overviews saw clickthrough rates fall by as much as 58% for top-ranking results. When the answer is served on the page, the click never happens. If you’re not in the answer, you’re increasingly invisible.
What actually gets you cited
Here’s the plot twist that surprised a lot of practitioners. In that same Ahrefs research, the strongest correlation with AI Overview visibility wasn’t backlinks or domain rating – it was branded web mentions, at r = 0.664, roughly three times stronger than the backlink signal. In other words, AI models seem to notice how often and how consistently the web talks about you, not just who links to you.
So what do you actually do? A few things that are working:
1. Answer the question in the first two sentences. LLMs love content that leads with a clear, self-contained answer and then explains. Bury the payoff in paragraph seven and you’ll get skimmed past. Front-load the takeaway, then support it.
2. Get mentioned on the sources AI already trusts. Reddit, industry roundups, comparison sites, credible news, well-known niche blogs — these are the pages models pull from constantly. A mention in a “best tools for X” listicle can do more for your AI visibility than a dozen quietly-acquired links.
3. Build genuine topical depth, not one hero page. Cover a subject thoroughly across multiple connected pieces. Models reward entities that clearly “own” a topic with consistent, corroborating information across the web.
4. Add structure a machine can parse. Clear H2s framed as questions, concise definitions, comparison tables, FAQs, and clean schema markup all make your content easier to lift and cite.
That last point comes with a warning, though. As Lily Ray noted after Google published its own guidance, you should “ignore most GEO/AEO hacks like chunking and llms.txt.” The gimmicks that flood LinkedIn every week are mostly noise. The fundamentals – clarity, credibility, coverage – are the whole game.
The consistency problem nobody warns you about
There’s a genuinely weird wrinkle to all this. Research from SparkToro found that if you ask ChatGPT to recommend brands in a category 100 times, there’s less than a 1-in-100 chance of getting the same list twice. So checking your AI visibility once and declaring victory (or defeat) is meaningless — you need to test the same prompts repeatedly and track how often you tend to appear.
Our own Peter Fox is blunt about the technical groundwork most brands skip:
“Half the brands worried about AI visibility haven’t checked whether the crawlers can even read them. Before you obsess over prompts, confirm GPTBot and friends aren’t blocked in robots.txt, your key answers render server-side without JavaScript, and your entity data is consistent everywhere. Fix the plumbing first — the citations follow.” Peter Fox, Head of Dev Team:
He’s right that GEO rewards the unglamorous stuff. An AI can’t cite a page it can’t fetch or render.
A realistic starting point this week
You don’t need a six-month transformation programme. Pick your ten most commercially important questions, run each through ChatGPT, Perplexity and Google’s AI Overviews, and note whether you appear and who does. That gap analysis becomes your roadmap. Then chase mentions on the sources already showing up in those answers, tighten your on-page structure, and re-test monthly. GEO is a compounding game, not a switch you flip.
FAQ
Is GEO replacing SEO?
No. They’re complementary. SEO still drives the bulk of trackable organic traffic; GEO wins presence inside AI answers, which increasingly shape what people click in the first place. Do both – the underlying work overlaps heavily.
Do I need special “AI schema” or an llms.txt file?
Standard, accurate schema markup helps machines understand your content. But dedicated AI-only files and “chunking hacks” are widely regarded as overhyped — several leading practitioners advise ignoring them. Solid fundamentals matter far more.
How do I even measure AI visibility?
Test consistent prompts repeatedly (not once) across ChatGPT, Perplexity, Gemini and Google AI Overviews, and track how frequently you’re cited. Dedicated AI-visibility tools exist now, but manual spot-checks are a fine starting point.
Why do brand mentions matter more than links for AI?
Recent data shows branded mentions correlate far more strongly with AI Overview visibility than backlinks do. AI models appear to weigh how consistently the web discusses an entity, not just who formally links to it.
How long until GEO efforts show results?
Expect weeks-to-months, not days. Getting mentioned on trusted sources and building topical depth compounds gradually. Because AI answers vary run-to-run, look for trends over time rather than instant wins.
Which AI engine should I prioritise?
Start where your audience actually is. For most B2C and general queries, Google AI Overviews reach the largest audience; for research-heavy and B2B queries, ChatGPT and Perplexity punch above their weight. Test all three before deciding.
Ian, Becky and Peter run the team at 3way.social, where we spend an unhealthy amount of time thinking about how brands earn visibility across search and AI. Come say hello.
