It’s a Tuesday night in Bangkok. A first-time runner sits on the bed, phone in hand, trying to pick a pair of running shoes for her first marathon.
A year ago, this would have looked very different. She would have opened Google. Read five reviews. Watched two YouTubers comparing cushioning and drop. Scrolled Pantip for honest opinions. Opened ten tabs side by side — ASICS, Hoka, Nike, On, Brooks. An hour later, she would have made a decision.
Tonight, she opens ChatGPT instead. She types one sentence: “What’s the best running shoe for a beginner training for a first marathon?”
Three brands come back. Confident. Synthesized. The side-by-side weighing — stride, budget, first-marathon goals — has already been done for her. Thirty seconds later, she clicks “buy.”
The other dozen brands she would have considered a year ago? They never appeared.
This is not just one runner. It is happening across millions of small decisions every day. Throughout 2025, traffic from Google to publisher websites dropped by roughly a third globally. When AI summaries appear at the top of a search result, click-through rates roughly halve. For a growing number of queries, users never visit a website at all. They get what they came for inside the AI itself.
For 25 years, the internet ran on a simple model — ten blue links, and the user chose. That model is quietly being replaced by something different — one synthesized answer, and the user trusts it.
For consumers, this is faster and easier. But for everyone trying to sell something, the rules of the game have just changed.
For more than two decades, brands played one game — SEO, or Search Engine Optimization. The goal was simple. Be on the first page of Google, ideally in the top three results. Whole industries were built on this game. Whole budgets, whole job descriptions, whole strategies.
But that game assumed something that is no longer true — that the customer would see a list and make a choice.
The new game is GEO, or Generative Engine Optimization. It is not about ranking on a list. It is about being named in the answer. When ChatGPT responds to a question about brands, recent data shows it typically names only three or four. If your brand is not one of them, you do not exist for that customer.
Not “ranked lower.” Invisible.
The difference is brutal. SEO was a competition for attention. GEO is a competition for mention.
It’s worth being clear: SEO is not going away. The fundamentals that made SEO work — having a site that is well-built, trustworthy, and easy for search engines to read — are exactly what GEO also draws on. A brand that has spent a decade on SEO is not starting from zero; it is holding the foundation GEO is built on. What is changing is that SEO alone is no longer enough. A brand can rank well on Google and still be absent from AI recommendations, not because its product got worse, but because AI is weighing a different, broader set of signals. Meanwhile, brands that may not rank well on Google at all can be the ones AI cites again and again, because their content is organised clearly, their authority signals are strong, or trusted third parties keep mentioning them.
The brands that win in the GEO era are the ones AI trusts — and AI’s trust is built differently from a customer’s trust. It is built through citations from authoritative sources, consistent presence across the web, and substantive content that compounds over time. It is also built through content that AI can actually read. Think of it like organising a kitchen: if the ingredients are in clearly labelled jars on a shelf, anyone who walks in can cook. If they are dumped into one giant box, even a trained chef takes an hour to find anything. Content works the same way. A page that is clearly organised — with plain headings, short sections, and information laid out where it belongs — is a page AI can summarise and quote. A page where the same information is buried inside long, flowing prose is a page AI will quietly skip past. Marketing copy alone does not earn it.
The way we measure success is changing too. For 25 years, marketers measured everything in clicks. Click-through rates. Bounce rates. Conversion funnels. Every metric assumed the customer would come to your website.
In the GEO era, many of those clicks never happen. The metric that matters is no longer “how many people clicked,” but “how often does our brand appear in AI answers?” Citation share. Mention frequency. Position within the response. The new scoreboard is harder to measure and far less familiar than the old one. But it is the scoreboard that matters now.
For Thai brands, this shift carries an extra layer of urgency. AI models are trained largely on English-language content, and the sources they cite are mostly global. A Thai brand can have decades of trust domestically and still be invisible in an AI-generated answer, simply because it is underrepresented in the data the AI was built on.
At SCBX, this is part of why we invest in research, open publications, and thought leadership — not as a marketing exercise, but as the foundation of being recognized in the new knowledge layer of the internet. The brands that contribute to what AI knows are the brands AI will continue to cite.
The runner laces up her new shoes. The brand she chose is happy. The dozen brands she did not even consider? They never knew they lost.
That is the new silence in the age of AI. The moment a decision is made, and the brands that were not mentioned simply were not there.
Writer:
