What Next For A Brand’s Online Presence in The AI Era?

Like clockwork, marketers have another shiny new thing to contend with. But this one looks less like a passing trend than a structural shift in how people decide what to buy.

Consumers are already asking AI what they should purchase, which brands to consider, and where to find them. That is forcing marketers to confront an uncomfortable question: what happens to brand discovery when the customer no longer needs to visit their website?

Generative Engine Optimisation (GEO), is currently valued at $1.09bn, with forecasts projecting the market could reach $32.92bn within the next decade. If anything close to that growth materialises, it will reshape how brands are discovered, compared, and considered online.

The implications stretch far beyond search. Ecommerce itself is expected to exceed $19.8tn by 2035, driven in part by mobile adoption and agent-assisted buying. For brands, the commercial value of their own digital platforms remains enormous. But the role those platforms play may be changing.

Major AI companies are also exploring advertising as a means of monetising their services. OpenAI has begun introducing advertising into ChatGPT, signalling the emergence of another battleground for brand visibility.

As DEPT associate SEO/GEO director Daniel Moreno puts it: “The brands that get cited are most likely to win the consideration set — everyone else is invisible.”

He adds: "What's fascinating from a consumer behaviour perspective is that AI search is expansionary. The nature of AI platforms allows consumers to think and search in completely new ways that traditional search never really supported. A recent Harvard research piece, done together with OpenAI, showed that around 48 per cent of AI searches are new behaviours; that's roughly half of all searches being queries the model hadn't seen before."

AI is compressing the customer journey, gathering information in seconds and changing how consumers research products and services. Marketers accustomed to keywords, rankings and click-through reports are entering an environment where much of that journey may happen inside a conversation they cannot see.

While still early, it is already proving compelling as a marketing engagement tool. According to recent SimilarWeb research, over half of AI-influenced visits to sites still come through search. AI influenced visitors to brand sites are being directed through intent, staying on page for double that of non AI-influenced users.

So while the value is already evident, increasingly, the challenge is not simply getting somebody onto a landing page. It is getting the brand into the answer in the first place.

The invisibility problem

Even in GEO’s early stages, zero-click behaviour is creating pressure on brands to understand how generative search works.

ChatGPT has surpassed 900 million weekly active users worldwide, while McKinsey projects that 75 per cent of Google searches will surface an AI Overview by 2028.

“That's a fundamental change in how visibility gets earned,” says Mirza Germovic VP, AI Platform Evaluation and Offering Development Lead for Edelman. “If a brand isn't showing up in the answer itself, it's effectively invisible for a growing share of searches. This isn't a future problem. It's reshaping how brands are discovered today, which is exactly why marketers need to treat it as a strategic priority rather than a wait-and-see trend.”

Becky Jenkins, organic performance director at TMW, part of Accenture Song, similarly argues that as consumers increasingly trust AI to simplify their decisions, brands that are absent from an LLM’s research may never make the shortlist.

Accenture’s Consumer Pulse Research suggests nearly two-thirds of UK consumers expect AI to influence at least half of their spending decisions in the next year.

“While that trust isn’t 100 per cent and customers are still spending a lot of time validating, a lot of that validation revolves around the brands AI has already shortlisted, instead of expanding out to other options. Being an option at the beginning must be the priority,” continues Jenkins.

There is another consequence: greater transparency.

Jenkins argues that, as AI draws on information from across the web, established brands may find it harder to rely on scale, authority or polished marketing to outweigh poor reviews and unfavourable comparisons. Those weaknesses can become part of the answer too.

What is a brand website for now?

Sebastian Micozzi, Bacardi’s former digital commerce chief, cautions that brands’ responses to GEO — and the future role of their owned platforms — will differ by sector. But the threat posed by zero-click behaviour is broadly shared.

If an AI engine provides the answer upfront, a consumer may never reach the brand’s website. That matters particularly for businesses that depend on clickstream data to build customer profiles or nurture prospects.

It also creates a difficult question around how much access brands should give AI engines to their sites.

Historically, allowing Google to crawl a site extensively helped its pages surface in search. With generative engines, however, greater access can also make it easier for an AI system to extract information and reproduce it within an answer, potentially removing the reason to visit the original platform.

“You have to raise the defences a little bit higher, if you want to play the ranking game,” he advises brands.

The challenge is finding the balance: making enough information accessible to earn citations and credibility without allowing the AI experience to replace the brand platform altogether.

That raises a more fundamental question. If generative engines increasingly replace websites as information utilities, what should persuade consumers to visit them?

The answer may lie in offering something AI cannot easily reproduce: entertainment, community, exclusive material, events, richer storytelling or experiences deliberately kept within the brand’s own environment.

Soizic Sycamore, managing director for Serviceplan's Plan.Net UK, advises that brands shift from promotional copy to answer‑led content that reflects real customer questions, decision points, and concerns, especially as consumers now complete much of their research through AI before arriving on site. They also need to structure and connect content so AI can interpret it easily, treating websites as coherent knowledge hubs rather than isolated pages. Finally, GEO demands authority beyond owned channels: social conversations, reviews, communities and expert commentary all shape how AI understands and validates a brand.

"The brands which succeed will be creating the most useful, trusted and credible sources of information in their category," Sycamore adds.

Rightmove approaches that question from a position of unusual strength.

One of the UK’s most popular websites, it attracts around 200 million visits a month. Its CMO, Matt Bushby, says around 85 per cent of traffic is direct, while Rightmove itself is the biggest search term in UK property.

Even so, the business is preparing for changing search behaviour. Traffic from LLMs currently accounts for less than 0.5 per cent of visits, according to Bushby, but he expects that figure to increase.

“The big thing we're focused on is having the right data and the right understanding of what's going on. So even though it is unpredictable, we can see what's happening and we can start to work out what happens as a consequence of changes to the platform,” he explains.

That is already influencing how Rightmove thinks about content.

“People are searching for things like ‘How do I get a mortgage?’ So how do we think about our content and how it helps to solve and answer some of those consumer questions is a big thing worth thinking about,” he adds.

The objective, then, is not simply to optimise for machines. It is to understand which questions consumers are now asking those machines — and decide where the brand can credibly provide the answer.

Your brand is what the internet says it is

GEO visibility is not determined by a brand’s owned platforms alone.

Generative systems draw on information circulating across the wider public internet, making sources such as Reddit, Wikipedia, reviews and online forums potentially important signals in how a business is understood and represented.

WPP Media SVP of data strategy Ollie Betts says agencies are already examining potential advertising opportunities within ChatGPT, analysing where brands might fit naturally within the context of users’ prompts.

At the same time, he says strategies are broadening across paid, owned, shared and earned media as marketers consider how everything from websites to creator coverage can influence a brand’s presence in generative results.

“Brands are at the beginning of that journey,” he says, before underlining the need to monitor how brand estates including apps and websites being integrated into platforms such as Anthropic and ChatGPT.

“There are brands who can make an API connection between the two, but it's whether that is relevant for the brand, whether it's a strategic fit. I think the New York Times block a lot of those GEO partners from getting access as they feel that it's a threat or it's not a strategic fit for them as a business. So there's a sort of vetting process involved here,” Betts continues.

Yet Jenkins warns against responding by simply producing more content.

“The industry fell into the ‘volume over quality’ trap before, and teams have put years of work into fixing that. Falling into it again risks taking a very costly step in the wrong direction.”

She adds that such an approach “risks undermining the unique and valuable role each channel plays in communicating with your audience”.

GEO activity, she argues, needs to sit within existing channel strategy rather than becoming another isolated optimisation exercise.

"Paid media will not become less important; it will become more strategic," states Plan.Net's Sycamore. There will be a shift from demand capture to shaping awareness, credibility, and preference much earlier in the journey, as AI increasingly influences research and comparison. Instead it will focus on building mental availability and amplifying authoritative content while supporting discoverability across fragmented AI, search, social, and commerce ecosystems.

Brands must now win the answers

Online measurement will have to change too.

If AI increasingly supplies the answer inside a chat interface, familiar SEO metrics such as rankings and click-through rates capture only part of what has happened.

“Brands need to start tracking whether and how they're being cited or referenced within AI-generated answers, alongside share-of-voice within those answers relative to competitors, and, critically, whether that visibility is translating into brand consideration and conversion downstream, even without a click. It's a shift from a ‘did they land on my page" mindset to a "was I part of the answer’ mindset,” outlines Edelman’s Germovic.

That distinction may ultimately matter more than the GEO label itself.

The mechanics of search are changing, but the marketer’s task remains recognisable: earn consideration, build preference and make the brand easy to choose. What is disappearing is the assumption that consumers must pass through a brand’s own digital front door along the way.

The question for marketers is no longer simply how to rank: it is how to make sure the brand is part of the answer.