Why GEO Isn't Enough; The Case for AI Brand Alignment

The Wall Street Journal ran a smart piece this week on how AI is rewriting the rules of search. It's worth reading. It captures something real about the moment we're in, and it represents where most of the industry's thinking currently lives.

Which is why it's worth a closer look.

The piece anchors on a McKinsey projection: by 2028, consumers will spend $750 billion on goods and services they find through “AI-powered search.” That number frames the conversation clearly: consumers rely on AI models for more than just helping them draft emails and presentations. AI models are influencing what they buy, and which brands they choose.  

As a result, brands are increasingly at the mercy of what AI models tell their customers. This is a narrative that most companies have never audited and can't currently see.

The industry's response was the rise of Generative Engine Optimization. GEO. AEO. The names vary; the work is growing. As the name suggests, it’s modeled after Search Engine Optimization. The playbook is: track which prompts mention you, monitor citation share relative to competitors, structure your content so models can parse and retrieve it. This framework is the SEO playbook, but replacing “keyword” with “prompt.” 

Treating AI models like a new search engine only takes you so far. In reality, they are more like a new influencer. Rather than optimizing for AI, marketers must persuade AI models to be an advocate for their brand.

The WSJ piece cites research showing that 40 to 60 percent of domains referenced in AI responses were completely different just one month later—identical prompts, entirely different sources. This is the canary in the coal mine for the “SEO for AI” approach. Citation share is inherently unstable because it's downstream of something more fundamental: what the model has come to believe about your brand, and why.

Models don't rank. They reason. They've formed opinions based on everything they've absorbed: product documentation, G2 reviews, Reddit threads, third-party comparisons, coverage from two years ago. They have read the entire internet, and from all of it the model has built a picture. Beliefs like "enterprise-ready but probably overkill for small teams." Or "strong integrations, inconsistent support." Those beliefs are live right now, shaping how AI models talk to a brand’s customers  at scale.

GEO tools measure what the model said. AI Brand Alignment asks why. This is the distinction Unusual is building around. 

"A search engine rewards the right signals in the right places. An AI model reads for meaning, triangulates across sources, and notices when your story doesn't hold together. Most brands are still thinking about AI visibility the way they thought about SEO—find the pattern, reverse-engineer it, publish more. That's an incomplete approach. What actually moves the needle is understanding why the model believes what it does, finding the root cause, and correcting it with honest proof. That's alignment. It's a fundamentally different kind of work." — Keller Maloney, Founder, Unusual

The average ChatGPT conversation is eight messages long. Buyers don't decide on the first response. They narrow. They add constraints like budget, integrations, team size, compliance requirements. Each follow-up is a moment where the model reaches into its knowledge and beliefs and makes a judgment call. Prompt share captures the first response. It tells you almost nothing about what wins the recommendation when real constraints are on the table.

That gap between first mention and final recommendation is where alignment work lives. Raunak Chowdhuri, co-founder and CTO of Reducto, experienced it firsthand. Despite serving Fortune 10 clients and raising $108 million, ChatGPT was characterizing his company as a niche startup tool. And that was traceable to specific signals the model had absorbed and weighted. 

"ChatGPT’s understanding of Reducto was stuck in the past. Once Unusual showed us which signals had shaped that perception, we knew exactly what to fix." — Raunak Chowdhuri, Co-Founder & CTO, Reducto

The practical implication is straightforward: you can't correct what you haven't diagnosed. Before publishing another piece of content aimed at improving AI visibility, it's worth understanding why the model currently believes what it does about you—and whether the signals driving that belief reflect the company you actually are today.

GEO tells you the score. AI Brand Alignment tells you what's driving it and how to fix it.

For $750 billion in decisions, that distinction is worth understanding now.