Eighteen months of research from brand consultancy 21st Century Brand reveals that AI systems choose which brands to recommend based on four specific levers - and most companies are falling short on at least two of them. The findings arrive as roughly 50% of web users now search using large language models, organic search traffic is projected to drop by half before 2028, and AI-qualified visitors already convert at 4.4 times the rate of traditional search traffic, according to data from Semrush.
The shift changes who - or what - marketers are actually building their brands for. As the research puts it, "The principles have not changed, but the evaluator has." The brands winning in this environment are not outspending competitors. They are leaving signals across the open web that AI systems can interpret with confidence.
The four levers that drive AI recommendations
The study, which drew on millions of AI prompts across dozens of categories, identified four variables with the greatest impact on brand visibility inside AI-generated responses: coherence, currency, authority, and advocacy.
Coherence means presenting a consistent, differentiated narrative that both people and machines can understand. Notion generates more than 96,000 AI mentions per month, reaching an audience of 575 million, because its value proposition aligns so tightly with product experience that AI systems describe the brand in near-identical terms across 15 countries and multiple languages. Coherence, the researchers note, is ultimately a product and strategy problem - not something marketing can fix alone.
Currency reflects how actively a brand participates in the conversations AI systems draw from. Lego reaches a monthly AI audience of 1.2 billion across 10,000 live topics. The driver is a business model engineered for continuous cultural momentum: a steady cadence of launches, partnerships, and community activity that gives people reasons to talk about the brand every week. Currency cannot be turned on for a campaign. It must be built into how the company operates.
Authority comes from being endorsed by credible voices across a category. Shopify grew its AI mentions to an audience of 1.1 billion, increasing mentions 360% in six months. That reach is supported by more than 119,000 cited pages spanning developers, partners, reviewers, and communities. The lesson is that durable authority increasingly depends on a broad, distributed validation ecosystem rather than a handful of high-status endorsements.
Advocacy is about creating experiences that customers describe in positive, specific terms - the kind of language AI systems recognize as genuine sentiment. Patagonia maintains a global AI visibility score of 69 out of 100 not because of its activism alone, but because product values like durability, repairability, and lifecycle are characteristics people encounter and talk about. What the brand says and what customers say about their experience are aligned, and AI picks up on that gap - or lack of it.
Why marketing can't go it alone
Most marketing teams are not yet saying this out loud, but a marketing department cannot improve performance across all four levers by itself. Coherence demands product and marketing alignment around a value proposition that matches customer experience. Currency requires upstream commitment from leadership to build a company that generates ongoing cultural relevance. Authority needs an ecosystem strategy that extends far beyond PR. Advocacy forces an honest reckoning with the distance between brand promise and product reality.
"The four levers are a brand-building framework, while acting on them is a whole company challenge," the research concludes. The CMOs who recognize this earliest will use it as a mandate to play a more central role in how their organizations communicate and operate. Brands that act now will compound an advantage quickly, because AI systems favor consistency, credibility, and genuine advocacy - exactly what customers want, too.
Why this matters for marketing professionals
Marketers have spent years optimizing for algorithms that rank pages. Now they must optimize for systems that evaluate brands holistically, drawing on thousands of signals in milliseconds. The skills required are still emerging, but the direction is clear: brand strategy must become a cross-functional discipline, not a departmental output. For marketing leaders ready to build that muscle, an AI Learning Path for Brand Managers offers a structured way to understand and act on the dynamics reshaping brand discovery. The evaluator has changed, and the brands that adapt their operating model - not just their marketing mix - will be the ones AI recommends.
Your membership also unlocks:
