A visibility score is not a booking

The AI visibility tools sell hotels a rising score built on questions the software invented — while the booking goes where it always went

A new category of software wants to tell you how visible your hotel is inside ChatGPT, Gemini and Claude. The pitch lands because the worry is real — guests do ask assistants where to stay, and no hotel wants to be the one the model never names. So you sign up, and a dashboard fills with scores: share of voice, citation rate, a visibility number that goes up or down each week, your property ranked against the competitors down the road.

It looks like the thing you needed. It mostly isn't, and it helps to see why.

Start with where the data comes from. These tools don't watch real travelers talking to real assistants — they can't, those conversations are private. So they generate the questions themselves. The platform invents a set of prompts it thinks a guest might type, runs them across the AI engines on a schedule, and records whether your name comes back. The score on the screen measures how you perform against questions the software wrote. Not questions a guest asked. A made-up exam, graded weekly, with your hotel's grade in bold at the top.

Then look at who the exam was built for. This whole category grew up around business software — the case studies, the language, the default prompts all point at SaaS companies trying to get named when someone asks for "the best project management tool." That world works the way the tools assume: a buyer starts with a category and narrows to a brand. Hotels don't. A guest starts with a place. Nobody shops "boutique hotel" in the abstract and then picks a city afterward — they choose Rome, or a shortlist of towns, and only then compare what's there. Geography is the whole conversation. A prompt generator built for software doesn't model that, so it measures you against questions shaped like someone else's industry.

Now follow what the dashboard tells you to do about your score. The answer is almost always the same: you're missing from these prompts, so make more content to fill the gap. More pages, more posts, more prompts to track next month. For a lean team at an independent hotel, that's the trap. The work is real, the hours are real, and they come straight out of the time that could go to the handful of conversations that actually send someone to your booking page.

That's the part the score quietly hides. A higher visibility number is not more bookings. It is more mentions, in answers to questions nobody asked, measured by a tool that profits when you track more of them. The product being sold is the measuring. The thing your hotel needs is the booking, and the two are not the same — a property can climb the dashboard all year while its rooms fill the way they always did, through the same channels, at the same cost.

None of this means visibility doesn't matter. Being absent when a guest genuinely asks an assistant about hotels in your town is a real problem worth fixing. The point is narrower: a measurement is not an outcome. Watching yourself appear in a synthetic answer feed tells you the software found you. It tells you nothing about whether the guest did, or whether the booking landed with you or somewhere down the chain.

So before the subscription renews, the question worth asking isn't how high the score is. It's which conversations actually move a booking — the real ones, in the destinations you compete in — and whether the dashboard is measuring those or just measuring itself. A score that rises while the rooms fill the same old way isn't visibility. It's a bill for watching.

by Markus Busch, Editor and Publisher of Hospitality.today

Read also: In AI search, the hotel's own website barely counts

Enjoying this analysis? Hospitality.today delivers daily insights on hotel distribution, AI trends, and travel commerce — straight to your inbox. Subscribe for free at Hospitality.today →