Branding, AI and the risk of broken brands

Mental availability is the key for any brand: being front of mind at the moment of purchase. 

How you get there – how you make yourself the inevitable choice as a brand – was always the focus of marketing, but has become a wholly new challenge. AI is now an active gatekeeper between customers and businesses and is starting to limit what gets served up to browsers in the first place: 68% of Google searches end without a click to another website. 

In traditional search, people still worked from their own memory. They saw a list of results, clicked, compared and made a choice. In AI-led discovery, the machine increasingly serves a smaller set of options. It changes the purchase journey, and makes it harder for marketers to see, track or influence what happened along the way. Nearly two-thirds of LLM visibility comes from long-term brand equity, according to recent research from marketing effectiveness expert WARC [What really drives LLM visibility, and what brands can do about it, 2026]. “Brand building is coming back into fashion, as the second-order effects of generative AI start to become clear,” the report states. “This time, it’s much bigger than advertising.”

This new paradigm means that the fundamentals of brand are more important than ever – it has become one of the most crucial tools to strengthen mental availability. To decide which brands are visible, humans need to feel – and brands still need to deliver emotion. But at the same time, the machine is looking for coherence and signal density: the same visual and verbal cues showing up across websites, retail conversations and social channels. Fragmented or contradictory signals make a brand harder to read.  The best brands have always been built through repetition, coherence and ownable signals, over time. LLMs now make the consequences of good – and bad – brand building even clearer. 

If a brand has the right triggers and associations for humans and the coherence and density of signals for machines, it will start to become inevitable. If there’s a noticeable gap between human appeal and machine legibility, however, then you have a problem. Ask AI for a ‘safe car’ and it will recommend Volvo; ask for a luxury car recommendation, and Jaguar doesn’t appear. Its fluctuating brand narrative over the past years reads transition, not authority to the machine.  If you want to close that gap, here is what good brand looks like in the world of AI discoverability…

Ownable assets are a must-have
Mental availability is not what the brand evokes in people. It’s what evokes the brand: the cues, assets and triggers that bring it to mind. Such distinctive assets have always been part of marketing must-have, with some of the most high-profile marketing experts like Byron Sharp and Mark Ritson among their biggest advocates. They are what trigger memory, associations and ultimately emotion in humans. MasterCard’s visual consistency in its red and yellow overlapping circles, for example.

But to truly boost mental availability through AI discoverability, it’s no longer enough for assets to be distinctive, they need to be ownable. Adidas’ three stripes are powerful in a sporting context, but only when presented in its recognisable brand shapes are they truly ownable. Coca Cola red may be distinctive within the drinks category, but the script and bottle contour are more ownable. This distinction matters, as AI sees patterns and recognises whether they are genuinely attributable to a brand. 

In our analysis of The Grocer’s Britain’s Biggest Brands in the UK, we found that in order to secure mental availability with not just humans but also AI, a brand needs at least three distinctive, ownable assets to be able to create a language that codifies the brand everywhere. 

Own a category entry point, not a vague purpose
Distinctive assets create availability, but they also need to be attached to something relevant. That means owning a category entry point: the reason someone enters the category in the first place. In sport, that might include performance, wellness, fashion or comfort. In confectionery, KitKat owns the ‘break’ rather than trying to convince me about the taste of their chocolate, let alone to tell a wider story about provenance and social good, 

We have to be wary about brand purpose. The point is not to chase every cultural conversation or engineer a higher-order role in people’s lives – one that the consumer in reality doesn’t care about. The point is to understand what people are looking for at the moment of choice, then attach the brand’s assets and narrative to that trigger.

You see the importance of this in the sports apparel category. Nike remains one of the strongest brands in the world for distinctive – and ownable – assets: the Swoosh, the ‘Just do it’, the Jordan silhouette. Yet, on category entry point, it is being beaten by competitors: by Lululemon in wellness; by smaller brands like Brooks in technical performance. 

In addition, Nike’s narrative has become less coherent over the years. ‘Just do it’ established a powerful idea, but later campaign messages such as ‘Believe in something’, ‘Runners welcome. Walkers tolerated’ and ‘Why do it?’ literally pull against that core. To a human, each may feel interesting in the moment; humans join the dots to form an overall impression. To a machine, however, they create semantic contradiction. It reads lack of coherence. The result is that AI systems consistently favour other brands over Nike, as a recent Harvard Business Review article highlights. 

Let advertising support the brand, not overwrite it
That coherence is crucial when it comes to wider communications. Advertising, for example, needs to find fresh, interesting ways to reinforce the brand narrative, not reinvent it in every campaign cycle. Marketers may tire of a message quickly because they see it every day; but consumers often take years to properly register it.

This is especially important now, because AI sees everything. A fleeting social post, a sales deck that tells a different story or a shopper message that pulls away from the core all add to the brand’s digital footprint. What feels tactical internally can quickly become incoherence externally.

The same is true for claims. Brands cannot rely on bold statements or trend-led positioning if those claims are not supported elsewhere. The machine can pick up gaps between what a brand says and what is evidenced across its wider footprint. Verifiable claims are therefore part of discoverability, not just compliance.

Brands with ownable assets, relevant category entry points, coherent messaging and credible claims are already building both human equity and machine legibility. The likes of Red Bull, Heinz, Fairy and Cadbury are ahead of the game: what they say, what they claim and how they look has remained consistent over time. Consumers recognise them, machines can read them, and both find what they expect. They are the ones that will continue to be easier for machines to index, easier for people to remember and, ultimately, more inevitable at the moment of choice.