How AI search is changing brand visibility and what you can do about it

People aren’t just looking for a link, they’re asking broader questions and looking for help to narrow their options. They’re having a conversation.

Search is dead. Search isn’t dead. AI search has changed everything or nothing at all – it all depends what you read. The reality is neither is true… or maybe it’s both. Search isn’t going away, but it is being redefined and the rise of AI search means brands need to change the way they think about content and communication.  

Google dominates but search behaviour is changing

First, let’s get some perspective. Despite the rise in AI-powered search tools, Google is still pretty big – according to its own stats, processing more than 5 trillion searches in 2024. So no, not dead yet.

But consumer behaviour is changing, and more people are starting their research somewhere else entirely – on AI tools like ChatGPT, Copilot or Perplexity.

However, the gap is also pretty big. According to data reported by Search Engine Land, even the most popular AI tools only process a fraction of that Google volume. ChatGPT handles around 37.5m search-like prompts a day, a minimal 0.25% market share.

More like a chasm than a gap, but as with so many things about AI, it’s the rate of change that means brands need to pay attention.

Adobe reported seeing the first real spike in generative AI traffic to US retail sites at the end of 2024 – up 1,300% on the year before. Maybe not a surprise, as AI is starting from a really low base, so it’s bound to look impressive. That’s not wrong, but the trend is still going strong. Traffic in February 2025 was still on the up – by 1,200% compared to the previous July, doubling every two months since the end of 2024.

AI search isn’t just new tech, it’s a new kind of conversation

This is where it gets really interesting because AI tools aren’t just a new where, they’re changing how people search and when brands show up in the decision-making process.

Why? Because the intent is different. People aren’t just looking for a link, they’re asking broader questions and looking for help to narrow their options. They’re having a conversation.

As highlighted in The Verge, surveys found more than half (55%) of AI search users say they use it for research, with nearly the same amount (47%) looking for recommendations on what to buy.

And that’s exactly how I used ChatGPT to help me buy a new bike.

What AI search looks like in practice

I’m no bike expert, but it was clear I needed an upgrade. The old one, bought when my son was little, wasn’t up to the job. These days, we’re cycling more as a family, and he’s easily keeping pace. With a 50km weekend ride coming up on our holiday, it was time to find something new. But where to start?

I began by telling ChatGPT how I use my bike, where I go cycling, and what I want to be able to do in the future. I also told it a bit about me and what I value, for example, I enjoy family bike rides but also like time on my own. And, crucially, I don’t wear Lycra and have no intention of doing so!

ChatGPT helped me narrow things down. It explained cycling terms I didn’t know. It compared bike models. Only after I’d decided what I was looking for did I go to Google – not to research, just to check stockists and compare prices.

My discovery happened long before I got to the traditional funnel. And the only content I saw was what the AI showed me.

Why brands need to prioritise showing up over showing off

This isn’t just a change in behaviour, it’s a change in visibility for brands.

According to research presented by Profound at BrightonSEO, only 12% of URLs cited by ChatGPT overlapped with Google’s top-ranking results. That means your organic search ranking really might not matter in the places people are now searching for information.

Helpful, clear information is the key to your brand showing up when it matters.

What content is most trusted by AI search tools

AI search tools – including Google’s AI Overview – tend to cite what’s clear, confident, structured and useful.

When AI tools go off to find that helpful information, they’re looking for something with experience, showing expertise and authority – something they can trust. Yep, they’re suckers for EEAT content – and not just direct from brand websites.

As highlighted by Search Engine Journal, a study found that during the “problem exploration” and “education” stages of a consumer journey there’s a higher percentage of citations from third-party editorial content, as well as plenty of other less obvious places, like UGC sources.

Depending on the AI tool, those UGC sources vary too – YouTube, LinkedIn, Reddit, you name it – wherever people (or brands) are providing helpful information, AI finds results.

These places often dominate AI citations because the content is simple, trustable and written like an answer; the answer to the question you asked your AI search tool!

Your URLs have to be helpful too

And to get technical for a moment, this ‘simplicity’ idea even goes as far as your URL. You can have the best, most informative brand content out there – but if your link structure is really long or full of meaningless numbers, watch out.

And Microsoft has confirmed that descriptive, keyword-rich URL slugs help AI systems understand and surface content more reliably.

What does effective AI-optimised content look like?

All this has shown us that AI search is happening right at the top of the marketing funnel. This is a challenge but also a great opportunity for brands, content marketing and digital PR.

Discovery, comparison, information, guidance – it’s where brands can build relevance and appear in the chats that customers are having with their AI search engine.

But to join in the conversation, you’ve got to have the right content out there, which means hitting some of these key factors:

• Information rich
• Simple headers
• Q&A structure
• Explainers and comparisons
• Semantic, descriptive URLs (eg /compare-gravel-bikes)

And most importantly:

• Helpful, human writing

The brands cited in generative answers don’t always have the biggest budgets, but they write like they know what they’re talking about.

Content teams should lead your AI search strategy

This isn’t a technical SEO problem. It’s a content and authority challenge.

AI tools are choosing which answers to trust, and they’re rewarding the kind of work that comms and editorial content teams are uniquely placed to produce:

• Content that’s clear, referenceable and structured
• Thought leadership and first-party data
• Brand mentions in relevant media
• Digital PR that builds reputation, not just links

Get your content into the new conversation

Search isn’t dead, but in an AI world it has moved into another realm.

There’s a new conversation happening and if you want your brand to be part of it, your content needs to sound like an answer worth repeating.

At Sticky, we help brands get into the right places by creating content worth talking about.

If you want to talk about how AI is reshaping the way your audience discovers and trusts information, we’re here for that conversation.