Can AI Recommend a Destination Nobody Is Talking About?

Schema may help AI understand your destination. Earned media helps AI trust it.

As we all settle into the reality that AI is becoming a major source of travel discovery, the role of travel PR is coming into sharper focus.

Our value is evolving in two important ways: helping destinations stay visible to AI and creating the third-party citations AI systems rely on.

Earned Media 2.0

Think of it as earned media 2.0.

AI relies on information that already exists across the web. If a destination isn't being written about, reviewed, discussed, linked to, or mentioned by credible sources, AI has very little evidence to work with.

And if AI can't find you, it can't recommend you.

That's why I believe we need to think about the entire digital ecosystem. Earned media remains a critical part of the puzzle, but so are Reddit discussions, YouTube videos, Facebook groups, Wikipedia entries, destination websites, and other places where travelers share information. SEO isn't dead either - Google recently made that clear - and AI won't cite content that isn't relevant or discoverable.

For PR professionals, that means the value of media coverage may be expanding rather than shrinking. Every credible mention becomes another signal that helps a destination become more visible across the web.

Being Visible vs. Being Understood

Being mentioned online is a good start. But being understood by AI is another challenge altogether.

We're entering a phase where how information is presented may become almost as important as the information itself.

I wouldn't be surprised if, in the future, we create two versions of every press release: one for journalists and readers, and another structured specifically to help AI understand and surface the information.

Alongside that version, we might provide schema markup, FAQs, authority sources, relevant links, and other structured elements that publishers could easily add to their websites.

If that sounds like extra work for writers and publishers, consider the upside. The more structured and authoritative their content becomes, the more likely they may be to appear in AI-generated answers as well.

Of course, AI is evolving so quickly that it's possible schema becomes less important over time as models get better at understanding content directly from the page. But for now, structured data remains one of the clearest ways to help AI understand what it's looking at.

My Lalaland Experiment

To test some of these ideas, I created a fictional destination called Lalaland in Europe and published a webpage about it using schema markup and structured content. (You can check it out here)

So far, that's all I've done.

I haven't invested in SEO, generated media coverage, or actively tried to create discussions elsewhere online. I'm simply monitoring what happens.

If Lalaland eventually appears in AI recommendations, that will tell me something interesting about the power of structured data alone.

If it doesn't, it may reinforce what many of us are beginning to suspect:

Schema helps AI understand a destination. Third-party validation helps AI trust it.

Either way, I'll keep monitoring the experiment and share what I learn.

One Thought I'm Leaving You With

For now, one conclusion feels increasingly clear:

AI can't recommend a destination nobody is talking about.

And for PR professionals, that means the value of earned media across a diverse mix of publications: from Tier 1 outlets to niche blogs and specialist websites, may be more important than ever.

Final Note: If you’re exploring how to make your content more AI-readable, I’ve developed a simple schema tool to help brands structure their information more effectively: pcpublicrelations.com/aischematool