For years, digital discovery meant one thing: get found on Google. But that model is shifting. People are still Googling, but more and more are simply asking ChatGPT instead. Large language models don’t send everyone to your website; they summarise, condense, and they might even cite you.
‍

But that doesn’t always translate into website visits – the rising use of LLMs may mean the number of people clicking onto your website actually falls. That can be unnerving if you’re glued to analytics, but it isn’t necessarily a bad sign. It can be a signal that your brand is making the AI shortlist.
‍

A quick visibility sense-check

A simple way to see this in action is to ask an AI assistant like ChatGPT to list “the best B2B marketing agencies in the UK,” “the top fintech PR firms,” or “the best tax lawyers in London.” The answers aren’t scientific and they vary, but they’re often pulled from places that signal authority: media coverage, professional directories and well-ranked, credible sites.
‍

Whether you like it or not, people are using these tools to search – which means visibility is no longer just a game of blue links. If LLMs are shaping the shortlist, your content needs to be in their line of sight too.
‍

Quality over quantity

The genAI era has produced a lot of slop – endless content churn and keyword stuffing that’s harder than ever to cut through. But it has also been a wake-up call. LLMs aren’t impressed by volume. They’re trained on the best of the web: trusted publications, authoritative voices and content that gets read, shared and remembered. Quality matters more than ever. It’s not about flooding the internet; it’s about showing up in the places that count – these are the places models look when they assemble answers.
‍

From pageviews to proof

This is where SEO starts to look more like PR too. For brands today, thought leadership that lands in respected outlets and timely commentary that earns citations is what puts you in an AI model’s line of sight. PR becomes the engine here, shaping distinctive points of view, placing them where they carry weight, and creating a steady drumbeat of credible mentions that models (and humans) can lift with confidence. Content on your own website still matters too – if it’s specific attributable and genuinely useful, it can serve AI models better than a stack of generic posts.
‍

Placed with intent

So, the question becomes practical: are you visible where it matters? Do assistants reliably associate your brand with the capabilities you want to be known for? Are you being cited in the right titles? If not, the answer isn’t more content – it’s creating better content, placed with intent.
‍

If you’d like an audit of how AI tools describe your brand, and a focused plan to strengthen your content and place your brand where it matters, get in touch with the Rostrum team at hello@rostrum.agency

‍