How is AI impacting search and how can adland help brands navigate this?
For consumers, search has seemingly never been easier. For brands, it’s never been harder.
Change has arrived. I’m no longer using a search engine to search for supplies for the end of the world when AI becomes sentient - I’m using generative AI itself. At every stage of the purchasing journey - whether via ChatGPT, Gemini, Copilot, Perplexity or Google’s AI Overview - AI is now shaping how discovery, consideration and decision-making happen.
At the awareness stage, instead of googling “best end of the world supplies,” I’m asking, “What should I consider when stockpiling for an AI vs human war?” At the consideration stage, rather than trawling through camper bed product specifications, I’m asking a chatbot for personalised recommendations for my war-torn back. And at the decision stage, I’m asking for final reviews and prices, neatly analysed in a table, instead of scratching my beard and juggling multiple browser tabs.
This might sound like a lofty metaphor, but it reflects a very real behavioural shift. According to McKinsey, 73% of people are using AI-powered search at the awareness stage, 60% at consideration and 61% at decision. Zooming out, half of all consumers now use generative AI to search, and around half of Google searches already return AI summaries. Usage spans all age groups, including older generations, and by 2028 McKinsey predicts that $750 billion (£559 billion) in US revenue will flow through AI-powered search environments.
For people, this makes life easier. Search becomes faster, more intuitive and more personalised. For brands, it does the opposite.
This shift is already impacting unprepared brands, many of which are predicted to lose between 20% and 50% of traffic from traditional search channels. Clicks are arriving later in the funnel - if at all - as AI summaries increasingly shape decisions before users ever reach a website. Zero-click search is no longer a future problem - it’s a present reality.
Crucially, AI-powered search is forcing brands to confront new fundamentals. In a world where models synthesise information on a user’s behalf, visibility is no longer just about being present - it’s about being trusted, contextually relevant and consistently represented wherever AI systems look for signals. Brands must also make life easier for the consumer, delivering clear, accurate and useful answers that fit real situations, not just optimised pages designed to rank.
For media and creative agencies, the challenge is turning those brand imperatives into action. Visibility in AI summaries and LLM-generated recommendations demands new GEO (Generative Engine Optimisation) approaches - but not at the expense of SEO fundamentals. GEO is not a replacement for search optimisation so much as another channel layered on top of it, requiring agencies to rethink content, distribution, authority signals and measurement in a world where influence increasingly happens before the click.
These questions for adland are now unavoidable. How are agencies preparing clients for the disruption AI-powered search will bring? What new capabilities, roles or workflows are emerging to support GEO? Where are the biggest risks - and the biggest creative or media opportunities - in a world where discovery is increasingly mediated by AI? And what must agencies do now to ensure brands remain visible, relevant and recommended when answers increasingly come without clicks?
We asked a host of agency AI and search specialists to find out.
The impact of AI on search
Search has become less of a neat little journey from question to answer and more like a pub crawl across platforms, with AI doing the remembering for us.
Francis Roshan Kuttivelli
Head of Technology
M+C Saatchi Performance
We're seeing a big shift in the way users search. Their behaviour is no longer linear, moving from a search query to a webpage. Now, we see users navigating between TikTok, Google, Reddit, YouTube, and AI overviews that can summarise everything and make the decision-making process a lot shorter.
We're no longer operating in an era of search engine rankings, either. Traditional SEO still matters. But for brands that want a presence across LLMs and AI summaries, SEO alone is not enough.
"We’re now living in a “zero-click” environment where the real question for brands is, “Are we cited?” And if a brand’s reputation isn't verifiable across multiple sources, AI doesn’t surface it."
Alex Walker
Managing Director
Havas Market UK
"Twenty years ago, the industry panicked about the death of print. Ten years ago, it was display. Now it's organic search. This is a familiar pattern - it’s not death, it's evolution."
Sam Hadley
SEO Specialist
Arke Agency
We’re preparing clients for AI-powered search by being clear about what is changing and what isn’t. While AI is often framed as disruptive, we see it as a positive reset; an opportunity to refocus on the fundamentals that drive sustainable growth.
"Discovery is diversifying: consumers are using AI tools for deep research and social platforms for trusted opinions. Classic SEO metrics like clicks and impressions no longer tell the full story. Brands now have a new audience to consider: agentic AI systems that synthesise signals from across the web."
What hasn’t changed is just as important. Google remains critical; data shared by Similarweb at BrightonSEO in October 2025 shows that 97% of ChatGPT users still rely on Google when they’re ready to convert. Human audiences still matter too. Understanding what motivates them, what they value, and how they make decisions remains essential, particularly as AI agents are trained on those same preferences. Reputation also continues to be a decisive signal. Consistent, authoritative content, positive sentiment, and off-site validation all compound over time.
SEO vs GEO
Despite the panic, SEO isn’t dead - it’s being built on, forcing brands to rethink optimisation as something broader than rankings alone.
Carl Ebanks
Digital Director
Cogent
The impact of AI on SEO is that it is more important than ever to have great SEO as a core part of your owned content strategy.
"Fundamentally, AI models will utilise underlying data from search results, combined with some of their own modelling to get results."
Gemini obviously uses Google Search, ChatGPT uses primarily Bing data, but combined with its own crawler and a tool like Perplexity, takes a more holistic approach with a focus on trusted sources. All of this leads to the answer that to be found in AI results, you also need to be found in traditional search.
Sam Hadley
SEO Specialist
Arke Agency
As John Mueller of Google said at Search Central Live in Zurich, AI systems rely on search, and there is no GEO or AEO without doing SEO fundamentals well. This is not to dismiss GEO, but GEO requires adaptation and progress, but not reinvention for its own sake.
"The brands that succeed will be those that evolve their strategies while staying anchored in the principles that have always driven effective marketing."
Trust, citation and reputation
When AI does the summarising, brands don’t win by shouting loudest - they win by being trusted, cited and recognisable wherever the model goes looking.
Alex Walker
Managing Director
Havas Market UK
We're preparing clients by reframing measurement entirely. The customer journey now spans TikTok discovery, YouTube research, Reddit reviews, and ChatGPT comparisons before any conversion. We're helping brands measure presence and sentiment across AI platforms, not just clicks from them.
On capabilities, we're investing in AI visibility auditing. McKinsey's research shows brand-owned pages represent only 5-10% of sources AI search draws from. Understanding where you're being cited, and how you're being characterised, is now foundational work. We've built proprietary tools on our Converged operating system to track exactly this.
"The biggest risk isn't declining traffic, it's measurement fragmentation. But there’s opportunity here for fewer visits and dramatically higher intent. Early data shows AI-referred traffic converts significantly better."
To harness these opportunities, agencies must stop treating GEO as an SEO bolt-on. It requires understanding which sources AI platforms trust, building content that earns citation rather than just ranking, and focusing on the human behind the click rather than the click itself.
Carl Ebanks
Digital Director
Cogent
That being said, there are, potentially, some considerations that need to be taken into account, including:
AI ranking analysis is an area traditional tools like Ahrefs and Semrush have been slow to respond to though they are now starting to release solutions to allow brands to monitor their AI results. I'm certain those tools will evolve but we'll be monitoring their ability to accurately measure AI results.
"Copywriting for a page is changing to be more direct with up-front answers to user queries summarised in a single paragraph, with the addition of unique content and some form of data specific to your business, becoming a key differentiator in whether you turn up in results. Simply writing the same thing a different way with better keyword placement will not cut it for AI."
Digital PR and brand mentions will become even more important with the need to have extensive digital footprints across a range of trusted sites. While traditional contextual backlink optimisation and building are still an important tactic in SEO, with AI results we need to extend that traditional approach as much as possible to ensure brands appear alongside specific content and across authoritative third-party publishers.
How adland is responding
In their creative sheds, agencies are retooling - blending data, content, PR, technology and measurement to help brands show up in AI-driven discovery.
Francis Roshan Kuttivelli
Head of Technology
M+C Saatchi Performance
What matters for brand visibility is having a clear, consistent presence that AI systems can trust. And that’s where we have currently pivoted: we have evolved to bring SEO, GEO, PR, content, and technology together. We have started consolidating all these efforts into a single workflow and are looking at it through a brand-demand lens for AI search.
As an agency, we’re delivering new competencies such as name-brand auditing, citation analysis, expert content creation and AI reputation tracking, so we clearly understand how and where our clients show up.
"If brands rely solely on SEO, they might quietly disappear from AI-generated summaries. But that risk is easily avoidable with a concerted shift towards AI visibility."
So, our role is clear: to ensure brands become part of the story whenever AI platforms shape their answers. And we are now well-equipped to help our clients build a reputation that extends beyond their website and establish their distinct, credible presence across multiple platforms.
Leila Seith Hassan
Chief Data Officer
Digitas UK
We moved early. When AI search began to shift behaviour in 2023, we started building what became Model Sight, a product that helps brands understand and optimise how they appear inside large language models. Because we already had strong data, content and creative foundations, this is now part of our standard offering.
Looking forward, we’re helping clients prepare for what comes next. We’re entering a world where people may visit websites less often and new standards like Model Context Protocols (MCPs) mean websites could evolve into repositories of content, data and tools that plug directly into LLMs. Brand interactions may increasingly happen through agents rather than owned platforms. Model Sight helps brands get ahead of this shift by showing what LLMs currently “know,” where the gaps are and what needs to change in their content, structure and messaging so they stay visible and accurate in this new environment.
Our AI Labs and Practice, set up in 2022, was designed to anticipate shifts like the move from traditional search to LLM discovery. Model Sight is one of its most successful products and it helps reveal how LLMs perceive a brand today.
"The technical approach has fundamentally changed. We’ve moved from keyword optimisation to contextual optimisation: understanding how models connect concepts, conversations, and sources."
This requires richer, more structured data that signals experience, expertise, authoritativeness and trustworthiness. And where brands once optimised for Google alone, they now need to optimise across many models, each with different behaviours and audiences.
Sarah Treliving
Chief Digital, Data and Technology Officer Goodstuff Communications
Understanding how consumers use AI prompts for search, can be understood directly - with a new range of tools, however in order for a brand to respond to these more specific and conversely more holistic prompts for a task, it requires a brand to understand their consumers, business and category in full technicolour, inspired or extracted from Search, Social, Web, competitors, staff, Reddit and traditional research. Arguably, for the first time all media is being planned from the same source audience insight, which is an exciting prospect for coherent communications and better performance.
AI-driven Search places particular focus on ‘natural language’ derived from the many, regional and category terms used to describe a need or audience - is it tots or toddlers or both? Is it fragrance, scent or perfume? Keywords are binary and are used as generic terms.
"Needs, requests and enquiries facilitated by more expansive content available via AI Search, means ‘searches’ or ‘prompts’ are more personal to our situations and context, and are therefore closer in language to how we speak rather than how platforms have directed us to search in the past."
In terms of execution, we are helping clients take advantage of shifting audiences, via new media planning, execution and measurement propositions and new platform partnerships. Being first to take advantage creates a short-term snatch of share, but the work involved continues to reap benefits in the longer term- from audience understanding, format optimisation and tracking integration.
Helping clients bridge the gap between the ecommerce team and marketing with business cases associated with media spend can help prioritise the new martech integrations needed to accommodate ‘multimodal’ and additional cross-platform tracking and measurement needs of AI.
Risks, opportunities and what comes next
AI makes discovery harder to control - but it also flattens the playing field, creating a rare reset where authority, clarity and consistency are rewarded over scale and spend.
Leila Seith Hassan
Chief Data Officer
Digitas UK
Brands face reduced control over how they appear in LLMs, and competitors can influence that. Nuance, humour and tone can also get lost, creating challenges for brands.
Because LLMs prioritise clear, consistent and structured information over historical SEO or brand dominance, every brand starts with a clean slate. Those who optimise early can establish authority, gain visibility in geo-relevant discovery and even leapfrog long-standing competitors, giving smaller brands a level playing field.
Brands can no longer pay their way out of low visibility. They need accurate, consistent messaging across every function, like marketing, PR, sales, and R&D, because inconsistencies confuse LLMs and can undermine trust.
"LLMs increasingly sit alongside friends, colleagues and professionals as trusted sources, and people expect them to help explore options conversationally and get to answers faster."
Brands that communicate clearly, align internally and reinforce the same story everywhere will earn authority and show up more reliably in AI-mediated discovery.
Fergal O'Connor
CEO and Founder
Buymedia
AI is changing the landscape, not only for search but for the paid ads model that drives revenue for Google.
"The likelihood of LLM models like Chat GPT bringing in paid advertising will happen sooner than we think and agencies must be armed with the best possible consumer and media data to help guide their media choices."
Agencies must have a continual learning loop built into their systems to help learn what is driving impacts in the business. By collecting campaign data, month after month, year after year, they build something incredibly valuable, where patterns become clearer, seasonality becomes predictable, incremental gains accelerate, waste reduces and forecasting improves. This is how organisations scale efficiently, by using data effectively to plan, execute, measure, learn and plan again.
Sam Hadley
SEO Specialist
Arke Agency
To support GEO, we’re investing in tools and workflows that track brand sentiment and visibility across AI-driven environments, focusing on engagement and trust rather than vanity metrics. The biggest risk is chasing short-term tactics or using AI to mass-produce inauthentic content. But the opportunities are great: this is permission to build brands which are true to themselves and unique to others. Focus on what doesn’t change: the needs of your audience plus your organisational strengths equals a future-proofed strategic direction for you or your clients.
"Search is no longer just about being found - it’s about being believed. As AI takes on more of the work of discovery, the brands that endure will not be those stockpiling keywords for a hypothetical end-of-the-world scenario like mine I mentioned earlier, but those that show up clearly, consistently and credibly - long before a click ever happens."
Built with Shorthand