Hotels Face New AI Challenge in Travel Discovery Shift

As travelers increasingly shift from traditional search engines like Google to artificial intelligence-powered platforms, the visibility of hotels in the travel discovery process has become more complex. Cloudbeds, a hospitality software provider, is tackling the issue by analyzing how upscale hotels appear in AI-generated search results, offering guidance to hoteliers navigating this fast-changing landscape.

In its new report, “The Signals Behind Hotel AI Recommendations,” Cloudbeds reviewed 145 high-performing hotels across key global markets, including London, Bangkok, and Barcelona. The properties were selected based on their strong performance on traditional search platforms and travel sites like Tripadvisor, Expedia, and Booking.com. Using hundreds of automated queries across leading large language models—ChatGPT, Gemini, and Perplexity—the company aimed to uncover what drives AI engines to recommend certain hotels.

The findings reveal that branded hotels from major groups dominate AI recommendations, making up more than two-thirds of the results. Cloudbeds CEO Adam Harris noted this mirrors the early days of SEO, where brands had to guess what content would boost visibility. Today, it’s about understanding the inputs AI models rely on to make recommendations—something even developers at OpenAI admit isn’t always clear.

Online travel agencies hold significant influence, accounting for 55.3% of cited sources in AI responses. Tripadvisor, Expedia, and Booking.com lead the field due to their existing partnerships with LLM platforms. The study also identified that properties with exceptional guest ratings and high review volume consistently ranked higher. Broad digital exposure matters too: nearly all hotels recommended by AI appeared on YouTube, Reddit, and travel blogs, reinforcing the importance of multi-channel visibility.

Cloudbeds emphasized that while AI models appear consistent in how they rank hotels, even small changes to prompts can affect outcomes. This suggests that refining how a hotel’s digital presence is structured and presented can make a measurable difference. Harris urged hoteliers to think beyond basic property listings, encouraging them to build knowledge bases that include information about nearby attractions, shopping, and transport options. These local details, when formatted correctly, help AI platforms understand and recommend a property more effectively.

In response, Cloudbeds is developing tools, including a prompt-based system that allows hotels to structure their data for LLMs. The aim is to empower hotels to control their narrative rather than relying on third-party platforms. Additionally, the company is working on a product that aggregates user-generated content to further build trust with consumers.

Lighthouse, a hospitality intelligence firm, also announced a new Connect AI platform that helps hotels surface their real-time data to AI models. It enables LLMs to access live availability, rates, and property details, and facilitates direct bookings through AI agents.

Ultimately, Harris believes embracing what makes a property unique—through storytelling, guest experiences, and local flavor—is the path forward. Generative Engine Optimization, or GEO, he says, will reward the same authenticity that SEO once did. As AI becomes more central to travel planning, hotels that build strong, consistent digital footprints will be better positioned to attract and convert future guests.

Tags: phocuswire, SyndicEx250629B

Hotels Face New AI Challenge in Travel Discovery Shift

Related Posts