Hotels just got a dashboard for agentic travel
Lighthouse acquires Hotelrank.ai, giving hotels their first real view into how AI platforms rank, route, and recommend their properties
A hotel can be recommended by ChatGPT and lose the booking to Booking.com in the same sentence. The AI mentions the property by name, describes it accurately, links to an OTA page. The guest books. The hotel pays commission and never knows the recommendation happened.
That single mechanic — whether an AI assistant surfaces a direct booking link or an OTA link — is the commercial question Hotelrank.ai was built around. This week, Lighthouse acquired the company, folding AI visibility analytics into its Connect AI distribution platform.
What the product tracks
Hotelrank.ai runs structured queries across ChatGPT, Gemini, Perplexity, and others at regular intervals — simulating how travelers with different profiles and intentions actually prompt these models. It records whether the hotel appears, where it ranks, what the model says about it. And then the number that determines whether this channel is paying its way: direct booking link or OTA link.
Competitive benchmarking sits on top of that. Not just how a property is performing, but how it stacks up against the hotels AI recommends instead. The perception layer tracks what attributes a model associates with the property — which may or may not match how the hotel actually wants to be understood.
Before the acquisition, Hotelrank.ai had done serious methodological work. Published research, over a million citations analyzed across multiple models and cities. Not a demo dressed as a platform.
What's strong
The OTA-vs-direct link split is the right commercial unit of measurement. Mention frequency flatters hotels that are getting routed into OTA inventory every time they appear. Where the guest lands is the question that connects AI visibility to hotel economics — and Hotelrank.ai oriented its product around that from the start.
Their research also turned up something worth paying attention to: breadth of presence across multiple sources consistently outperforms depth on any single platform. A hotel visible across Google Maps, TripAdvisor, Booking.com, and its own website appears in AI recommendations more reliably than a hotel that dominates just one. The models use a source-aggregation method that rewards coverage over concentration. Knowing that changes where you spend the optimization effort.
Then there's the volatility data. The top AI recommendation for a given query holds roughly half the time. The channel shifts. A strong visibility score today is a snapshot, not a position. That's a more honest framing of what measurement can and can't tell you — and one that distinguishes a product with real research behind it from the ones selling certainty.
What it doesn't fully solve
The optimization levers Hotelrank.ai's own research points to — complete OTA listings, structured schema markup, Google Business Profile accuracy, review recency — are existing hygiene work that AI has made more consequential. The dashboard shows whether you've done it and how it's performing. The levers themselves aren't new.
The harder structural problem sits underneath. AI constructs hotel descriptions by synthesizing sources it has already indexed. The hotel can shape those sources — but can't rewrite the synthesis. Improved visibility scores are real, but how much reflects intentional hotel action versus model updates, competitor changes, or query drift is rarely clean to isolate. The measurement is honest about this. The marketing around AI optimization often isn't.
What becomes possible
For the independent hotelier who has watched agentic travel form from the outside — uncertain whether they even exist in it — the more interesting implication is what becomes possible once you can see the data. Not a new strategy. A new starting point. Which AI platforms mention you. Which don't. Whether the links that surface point home or to an OTA. How the model describes you against the hotel down the street. These are questions commercial teams have been asking without any way to answer them.
A dashboard doesn't win the channel. But it ends the guesswork — and for a hotelier who has spent two years wondering whether agentic travel is happening to their property or passing it by, that's not a small thing.
by Markus Busch, Editor and Publisher of hospitality.today
Read also: What direct AI hotel distribution looks like
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 →