How to optimise your website for the AI layer

For years, we designed websites assuming users would Google something, click a link, and land on our carefully crafted page.

Now? Not so much.

People are asking ChatGPT, Claude, Perplexity—even Google itself—and getting answers instantly. No clicks. No pageviews. No time on site.

That’s not theory. It’s behaviour.

I’m doing it. You’re probably doing it. So are your customers.

We’ve moved from “search and sift” to “ask and receive.” Recent studies indicate that approximately 65% of global Google searches result in zero clicks, with mobile searches exceeding 75%.

This means we must rethink how we show up, not just for people but also for the machines now mediating access to everything.

At Catch, we work with large organisations in government, membership, and enterprise, and we’re seeing this trend reshape how digital platforms need to perform.

This change has structural implications for large organisations, especially those with sprawling sites, service ecosystems, or regulated content.

It's not just about tweaking SEO.

It means rethinking:

• Content governance

• Information architecture

• CMS capability

• API readiness

• Measurement frameworks

It’s system-level design, reimagined for the AI age.

This isn’t about chasing Google algorithms or pleasing robots. It’s about meeting people where they are now — and increasingly, that means meeting them in the AI layer.

Think of it like this:

Your brand is no longer just a site or campaign.

It’s a system of content, data, and structure that needs to make sense to humans and AI.

Some call this shift Answer Engine Optimisation (AEO), also Generative Engine Optimisation (GEO) and Generative Search Optimisation (GSO).

Whatever the acronym, it’s not just about ranking well, but being selected and surfaced as the answer. It’s not a gimmick. It’s a mindset: Helpful, Structured, Trustworthy.

Here are six human-first, AI-aware actions to take right now:

1. Create helpful, honest content

Skip the fluff. Avoid keyword stuffing. Just answer people’s questions in plain language using your real voice and expertise.

Focus on content that’s:

• Helpful

• Specific

• Factual

• Up to date
  (Yes, even that old blog post from 2021 needs attention.)

2. Use clear structure (for people and machines)

Use headings, bullets, tables, FAQs — anything that breaks content into digestible chunks.

Think:
“Could this be lifted into an AI answer box?”
If yes, you’re on the right track.

Remember: AI is trained to emulate human thinking, so if your content is clear, natural, and easy for people to understand, you’re also making it easier for AI to surface.

3. Add context and authority

AI systems are sniffing out signals of trust. Use named authors—link to reputable sources. Add stats, quotes, and references. Don’t just say something—prove it.

Mentions from trusted sources like Wikipedia, academic sites, or established publishers tend to carry more weight in AI training models, so building credibility across the web helps your own content get surfaced.

Google's Search Quality Evaluator Guidelines emphasise the importance of Experience, Expertise, Authoritativeness, and Trustworthiness (E-E-A-T) in content. They highlight that high-quality pages often have a positive reputation and are cited by other authoritative sources.

(Also, don’t be afraid to show personality. People still want to feel like a human is behind the brand.)

4. Optimise with the right tools

AI doesn’t just read—it parses, maps, and compares.

To show up in AI-driven results, your content needs to be:

• Well-structured (clear headings, chunked content, logical hierarchy)

• Semantically tagged (using schema.org, metadata, microdata)

• Publicly accessible (no content behind logins or broken APIs)

That’s less about which CMS you use, and more about how you model and expose content to machines.

That said, platforms like Storyblok, which we work with, make it easier to manage structured, reusable, AI-friendly content at scale. However, the value comes from how you use the tools, not just which ones you choose.

Good old-fashioned SEO principles—like accessibility, content flow, metadata and clarity—are more important than ever. AI rewards clean, clear, machine-readable sites.

5. Make your site callable, not just clickable

In the future, AI assistants might transact without users visiting your site. So start thinking about:

• APIs

• Structured product feeds

• Trust signals and partnerships

You want your business to be part of the answer, not just waiting for a click.

6. Redefine "Performance"

Clicks and traffic still matter—but they’re no longer the whole story. As Reforge points out, traditional metrics like keyword rankings are becoming less indicative of success in AI-driven search.

Smart brands should start tracking:

Brand search volume
Use tools like Google Search Console or SEMrush to see if more people are searching for you by name, which is often a sign that your content is landing somewhere, even if they didn’t click from AI.

AI mentions (where possible)
Try using tools like Perplexity or ChatGPT to ask browsing questions your audience might ask and see if your brand or content shows up in the answer. Mentions—not just backlinks—help train AI to associate your brand with credibility in context.

Voice search data
Look for longer, natural-language queries in your Search Console—these often hint at voice assistant behaviour. Make sure your content is well structured for featured snippets and smart answers.

Referral lifts after publishing
Keep an eye on what happens after key content goes live. Tools like GA4 or Looker Studio can help you track traffic shifts, branded query spikes or repeat visits — even when attribution isn’t perfect.

You won’t get perfect visibility—yet—but you can start spotting the signals. It’s less about clicks, more about influence.

Important note: Optimising for AI doesn’t mean abandoning traditional SEO. In fact, many of the same fundamentals—like metadata, accessibility, content targeting and semantic structure—are even more important now. The difference is, you’re now designing for both people and intelligent systems.

If your digital strategy is still built around “get found on Google”, you’re already behind.

For large businesses and government agencies, ignoring this shift could mean becoming invisible, not to Google, but to the way people actually discover and evaluate solutions today.

The opportunity?

Be the brand that shows up confidently in the new AI-powered discovery layer.

Lead with clarity, structure, and trust. And most of all, with humans in mind.

Want help figuring out how your content, tech and structure are working in the AI layer?

Let’s talk. No hype. Happy to help—no bots, just humans.