A new discipline called Generative Engine Optimization — or GEO — is quietly reshaping how companies get noticed in the age of AI search. Unlike traditional SEO, which tweaks website content to climb Google rankings, GEO focuses on earning citations inside the answers that AI engines deliver. McKinsey research shows that roughly half of consumers now use AI-powered search, and 44% call it their primary way to look for information, compared to 31% who still prefer traditional search. For brands that ignore this shift, the consulting firm estimates a 20 to 50 percent drop in traffic from conventional search channels as decision-making moves to AI platforms.
Why most brands are flying blind
Despite the stakes, only 16% of brands systematically track their performance in AI search, according to McKinsey. The rest are operating without a clear picture of whether their products, services, or leadership appear in the answers ChatGPT, Perplexity, or Google's AI Overviews generate. The gap matters because the mechanics of discovery are fundamentally different. In an AI engine, a source is either cited or it's not — there's no middle ground like a second-page ranking. That binary reality means a single mention in a credible outlet can become a permanent citation for every relevant query, while silence means invisibility.
The earned-media advantage
A 2026 academic analysis of generative search found a systematic and overwhelming bias toward earned media over brand-owned and social content. Brand-owned content accounts for only 5 to 10% of what AI engines draw from; the remaining 90%-plus comes from external, earned sources. That flips the old marketing playbook. Instead of spending heavily on branded blog posts or social campaigns, companies now have a stronger incentive to land placements in independent news outlets, trade publications, and analyst reports. Every mention in a credible venue becomes a potential citation for an AI engine answering a question about the company or its category.
Measuring what AI sees
The shift has created a new metric: LLM Performance, as measured by tools like the Outset Media Index. The metric tracks how well outlets perform in AI-driven discovery, giving PR teams a data-backed way to pick which publications will actually move the needle. GEO, in essence, optimizes the broader set of external sources an AI engine reads to influence its synthesis — a fundamentally wider net than SEO, which optimizes only the pages a brand controls. For communications departments, that means the job isn't just pitching stories anymore; it's pitching stories that are structured and placed in outlets likely to be cited by large language models.
What comes next
The data from McKinsey and the 2026 analysis makes one thing clear: the window to adapt is narrow. Brands that start tracking their AI search footprint now have a chance to shape how they're represented before the technology becomes the default gateway for consumers. The question that remains unanswered — and that will drive strategy for the next year — is whether most companies will treat GEO as a temporary experiment or a permanent restructuring of how they think about visibility.
