Small publishers lost 60% of search referrals. Google beat a publishers antitrust suit. For brands, the game is becoming a source inside the answer layer.
The Wall Street Journal ran a piece in December framing AI search optimization as a nascent, possibly overhyped industry. That framing is already outdated. What is happening is not a new marketing subculture. It is a channel shift — the kind that restructures where and how brands get discovered.
The data no longer supports treating AI search as a curiosity. It demands treating it as infrastructure.
The Numbers Are Past the "Interesting" Phase
Google's AI Overviews now reach two billion monthly users across more than 200 countries. Sundar Pichai called it "one of the most successful launches in Search in the past decade." AI Overviews now appear on roughly 27% of search queries, up from under 4% at the start of 2025. That is a sevenfold expansion in ten months.
Outside Google, the numbers are equally pointed. ChatGPT processes over two billion queries per day. AI-powered chatbots now account for more than 5% of U.S. desktop search traffic, up from 1.3% in early 2024. Perplexity tripled its monthly query volume in under a year. These are not novelty numbers. These are adoption curves.
Gartner predicted traditional search engine volume would drop 25% by 2026. We are on pace.
Visibility Has Moved From Ranking to Retrieval
The mechanics are different in a way that matters. Traditional SEO optimizes for a ranked list. AI search optimization — whether you call it GEO, AEO, or something else — optimizes for retrieval into a synthesized answer. The page does not win a position. It gets selected as a source object, chunked, and woven into a response the user may never click through.
The clearest evidence that these are separate channels: fewer than 10% of sources cited by ChatGPT, Gemini, and Copilot rank in Google's top 10 organic results for the same query. Ranking well on Google does not mean you will appear in an AI-generated answer. And appearing in an AI answer does not require ranking at all.
That decoupling is the whole story. It means companies need a separate strategy for AI visibility, not a bolt-on to their existing SEO program.
Companies Are Already Adapting
The WSJ documented specific companies adjusting their content strategies for AI retrieval. Cars.com made editorial changes to surface in AI-generated answers. Profound, the AI search analytics startup, raised $35 million in a Sequoia-led Series B and now serves over 500 enterprise clients including Ramp, U.S. Bank, Indeed, and DocuSign. These are not fringe operators experimenting. These are large organizations treating AI search as a measurable channel.
The WSJ also told the cautionary side. Lorelight, an AI optimization startup, shut down in October 2025 after its founder concluded the data "simply reinforced the value of time-tested practices for traditional search engine optimization." That tension is real, but the takeaway is not that GEO is fake — it is that the discipline is still being defined, and the organizations that win will be the ones that integrate extractable content practices into their broader content strategy rather than treating it as a separate trick.
What AI Systems Actually Prefer
The content formats AI systems cite most are not mysterious. They are structured, specific, and experience-rich.
FAQs and direct-answer formats. When a page answers a question in a self-contained block, retrieval-augmented generation systems can extract it cleanly. The format does the work.
Descriptive product detail. Specification-level content — dimensions, ingredients, compatibility lists, pricing structures — gives AI systems concrete facts to synthesize. Vague marketing copy does not get retrieved.
Lists and structured comparisons. Ordered, labeled content chunks map naturally to how LLMs compose answers. If your content is already chunked, the system does not have to guess where one idea ends and another begins.
Then there is the third-party layer. Reddit is the most-cited domain across Google AI Overviews, Perplexity, and ChatGPT. Reddit's licensing deals with Google ($60 million per year) and OpenAI (roughly $70 million per year) have formalized what was already observable in citation patterns: AI systems trust discussion-based, experience-rich content. Reddit appears in 97.5% of product review queries on Google.
For brands, this means monitoring and influencing what gets said on forums, review sites, and community platforms is no longer optional reputation management. It is a direct input to AI retrieval.
The Zero-Click Reality
The broader context makes the channel shift even more urgent. SparkToro's 2024 zero-click study found that 58.5% of U.S. Google searches end with zero clicks. For every 1,000 searches, only 360 clicks reach the open web. AI Overviews accelerate that trend: Seer Interactive's tracking shows organic click-through rates dropped 61% on queries where AI Overviews appear.
But here is the counterweight. When a brand is cited in an AI Overview, it receives 35% more organic clicks and 91% more paid clicks compared to non-cited brands. The overall click pool is shrinking, but the brands inside the answer layer are capturing a larger share of what remains.
That is the dynamic that makes this a real channel shift and not a fringe concern. The game is no longer about ranking for keywords. It is about becoming a source object that AI systems retrieve, cite, and present to the user as part of the answer.
What This Means for Content Strategy
Bain & Company's analysis found that roughly 80% of consumers now rely on zero-click results at least 40% of the time. Their conclusion: marketers need to shift metrics from clicks to AI reach. That is not a theoretical recommendation. It is a description of what the successful operators are already doing.
The practical implications are concrete. Structure your content for extraction, not just consumption. Make every key claim self-contained enough to survive being pulled out of context. Monitor what AI systems say about your brand across ChatGPT, Gemini, Perplexity, and Copilot — not just what Google ranks. And pay attention to what your customers are saying on Reddit, forums, and review sites, because that content is feeding the systems that generate the answers.
This is not a subculture. It is the new surface area for discovery. The brands that treat it as such will be visible in the answer layer. The ones that do not will be optimizing for a shrinking click pool and wondering where the traffic went.
James Calder is the editor of The Search Signal, covering AI-powered search, generative engine optimization, and the future of brand discovery.
Small publishers lost 60% of search referrals. Google beat a publishers antitrust suit. For brands, the game is becoming a source inside the answer layer.
Perplexity's Personal Computer turns a Mac mini into a persistent, always-on AI agent that works across your local files and apps around the clock — a new category of ambient computing.
Brave Search now offers two AI-native API endpoints that deliver pre-extracted, relevance-scored web content directly to LLMs — signaling a fundamental shift in how search APIs serve machine consumers.
