Guests Now Trust AI to Find a Hotel — Just Not Yet to Book It
Adobe Analytics shows a 194% year-over-year rise in AI-source traffic to travel sites, part of a three-stage breakdown of how AI is rewriting hotel discovery, browsing, and booking — with adoption strongest at discovery and most cautious at the final booking step.
Travel sites saw a 194% year-over-year increase in traffic from AI sources in May 2026, according to Adobe Analytics data cited in a new Hospitality Net piece — drawn from an analysis of 8 million visits, and part of a three-stage argument for how AI is rewriting hotel booking. Coalition Technologies’ Jordan Brannon breaks the guest journey into discovery, browsing, and booking, and finds adoption moving at three different speeds. In discovery, AI is now a booking channel in its own right — 43% of U.S. travelers with household income above $100,000 already use generative AI during trip planning. In browsing, Expedia Group research shows 40% of travelers use AI to build itineraries and 53% are comfortable letting it suggest options, meaning hotel websites now need to read well to AI agents, not just humans. Booking itself is the holdout: citing Booking.com’s 2025 Global AI Sentiment Report, only 12% of travelers are comfortable letting AI book independently, even though 65% expect autonomous trip planning to go mainstream soon.
The discovery-stage number lines up with a trend GuestEx has tracked since summer. Wyndham CEO Geoff Ballotti disclosed on a Q2 2026 earnings call that roughly 60% of his company’s guest searches now happen inside an LLM rather than a search engine — a statistic concrete enough that Wyndham built its ChatGPT app specifically to capture that demand before it leaks to a competitor’s surface. A separate Amadeus survey of 6,000 travelers found 69% now trust an AI-generated summary enough to book without further research, with that trust running as high as 87% in India. Read against Brannon’s three-stage breakdown, the pattern holds across three independent sources: guests already trust AI to find and shortlist a hotel; they’re still cautious about letting it click “confirm.” For operators, that’s a narrower, more specific optimization target than “be AI-ready” — get found and get shortlisted, and expect the human to still make the final call for a while longer.
Source: Hospitality Net — Your Next Guest Is A Robot: How AI Is Rewriting The Path To Booking Auto-generated brief — verified before publishing.