We Tested 2,800 Travel Queries. Here's What AI Actually Recommends.
At SearchIntel, we ran 2,800 travel queries across ChatGPT, Claude and Gemini covering UK and US markets. We tested discovery queries ("things to do in Barcelona"), comparison queries ("best tours in London") and booking queries ("book Colosseum tickets"). Then we tracked which brands appeared, on which platform, and how often.
The results challenge assumptions about how AI recommends travel brands.
Finding 1: Gemini Dominates Travel Brand Mentions
Gemini mentions travel brands more than any other platform — accounting for roughly 45% of all OTA mentions despite being just one of three LLMs tested.
| Platform | GetYourGuide | Viator | TripAdvisor |
|---|---|---|---|
| Gemini | 168 | 171 | 212 |
| ChatGPT | 126 | 136 | 132 |
| Claude | 85 | 90 | 56 |
Gemini's responses are also 3.5x longer than Claude's (5,000 characters vs 1,375), creating more opportunities for brand mentions — but also more noise. If you're a travel brand tracking AI search visibility, Gemini should be a priority.
Finding 2: Booking Intent Is Where OTAs Win
GetYourGuide and Viator are mentioned 2x more in booking queries than comparison queries. TripAdvisor shows the opposite pattern — it owns the comparison moment.
| Brand | Comparison Mentions | Booking Mentions | Booking Share |
|---|---|---|---|
| GetYourGuide | 116 (31%) | 261 (69%) | 69% |
| Viator | 126 (32%) | 269 (68%) | 68% |
| TripAdvisor | 229 (59%) | 156 (41%) | 41% |
When someone asks "best tours in Rome", TripAdvisor appears 59% of the time. When they ask "book Colosseum tickets", GetYourGuide and Viator dominate. This means different brands win at different stages of the customer journey — and most travel brands have no idea where they stand.
Finding 3: Discovery Queries Rarely Mention Booking Platforms
For "things to do in [city]" queries, GetYourGuide and Viator were almost never mentioned — just 2 mentions each across hundreds of discovery queries. Kayak, by contrast, appeared 128 times.
OTAs don't compete in the discovery phase. That space belongs to destination content, tourism boards and editorial sites. The AI search visibility battle for tour operators starts at the comparison and booking stages.
Finding 4: AI Platforms Geo-Target Brand Recommendations
The UK and US markets see dramatically different brand recommendations:
| Brand | UK Mentions | US Mentions | Difference |
|---|---|---|---|
| Booking.com | 121 | 51 | 2.4x higher in UK |
| Kayak | 94 | 197 | 2.1x higher in US |
| Skyscanner | 54 | 6 | 9x higher in UK |
| Priceline | 1 | 32 | Nearly zero in UK |
European-headquartered OTAs have significantly higher AI visibility in UK queries. US brands dominate their home market. If you're a travel brand operating across markets, your AI search visibility strategy needs to be market-specific.
Finding 5: Each Platform Has Brand Affinities
Not all platforms treat brands equally:
- ChatGPT favours Airbnb — 88 mentions vs 77 on Gemini and 41 on Claude
- Gemini favours traditional OTAs — highest mentions for GetYourGuide, Viator and TripAdvisor
- Claude is the most conservative — lowest overall OTA mention rate, but introduces brands earliest in its responses (at just 8.5% into the answer)
A brand can appear visible on one platform and invisible on another. Measuring only one platform gives you an incomplete picture. Learn more about how to measure AI search visibility across all platforms.
What This Means for Travel Brands
You need platform-specific visibility data. Generic "AI visibility" claims mean nothing when Gemini mentions you 168 times and Claude mentions you 56 times. Each platform requires its own analysis.
Intent matters more than volume. A travel brand that appears in 100 discovery queries but zero booking queries is losing the moments that drive revenue. Map your AI visibility by intent stage, not just total mentions.
Market-specific strategies are essential. A UK strategy that works for Booking.com will fail for Kayak. AI platforms appear to geo-weight recommendations based on brand origin and market relevance.
Monitor continuously, not once. AI models retrain regularly. A brand's Share of Memory — its presence in AI's knowledge — can shift as competitor narratives strengthen. What's true today may not be true next quarter.
Key Takeaways
- Gemini accounts for ~45% of all travel OTA mentions — the most brand-generous platform
- Booking intent is where OTAs win — GetYourGuide and Viator dominate booking queries 2:1 over comparison
- TripAdvisor owns comparison — mentioned in 59% of "best X" queries
- AI geo-targets recommendations — Booking.com is 2.4x more visible in UK, Kayak 2.1x more in US
- Each platform has brand affinities — ChatGPT favours Airbnb, Gemini favours traditional OTAs
- Discovery queries ignore OTAs — the AI visibility battle starts at comparison and booking
The Bigger Picture: 21 UK Travel Brands Benchmarked
Beyond our OTA-focused query analysis, we also tested 21 major UK travel brands across ChatGPT, Claude, Gemini and Google AI Overviews to understand the broader market landscape. The results were stark:
- Average AI visibility score: 37 out of 100. Most brands are more invisible than visible across AI platforms.
- 90% of brands received zero citations in Google AI Overviews. The platform that appears directly in Google search results is almost entirely ignoring the travel industry.
- Brand size did not predict AI visibility. Several well-known operators with large marketing budgets scored below 30. Smaller, niche brands with strong review footprints sometimes outperformed them.
The pattern was clear. Google ranking, brand awareness and marketing spend are poor predictors of AI visibility. The brands AI recommends are the ones with the strongest third-party evidence across the web.
Why Travel Is Uniquely Exposed
Travel is one of the most affected verticals for a specific reason: it is inherently recommendation-based. Nobody searches for "buy holiday." They search for "best holiday for..." or "where should I go for..." These are exactly the kinds of questions AI platforms are designed to answer. And when AI answers a recommendation query, it typically names one to three brands. Not ten. Not twenty.
If your brand is not in that shortlist, you do not exist in the traveller's decision process. There is no second page of AI results. The traveller gets an answer and either follows it or refines the question.
68% of Google searches now end without a click. Google AI Overviews appear on the majority of travel-related queries. ChatGPT has over 800 million weekly users. Research from the Deloitte Travel Industry Outlook confirms that Gen Z and millennial travellers are adopting AI-powered trip planning at accelerating rates. This is not a trend on the horizon. It is happening now.
Actionable Steps for Travel Brands
1. Build structured data that AI can read
Implement Organization, FAQ and Product schema across your website. TourProduct and TravelAction schema types are particularly relevant for tourism operators. Without structured data, AI has to guess — and it often guesses wrong or skips you entirely.
2. Create FAQ content that matches traveller questions
AI models heavily favour FAQ-style content when generating travel recommendations. Build comprehensive FAQ sections that answer questions like "What's the best time to visit Patagonia?" or "How much does a guided safari cost?" Use natural language, be specific, and include real numbers — costs, durations, group sizes.
3. Strengthen your third-party citation footprint
This is the single most important factor. AI models build confidence in brands through independent corroboration:
- Review platforms: Active, recent reviews on TripAdvisor, Trustpilot, Google Reviews and sector-specific platforms like Feefo.
- Reddit and forums: Genuine mentions in travel subreddits and forums like Lonely Planet's Thorn Tree. AI models pull heavily from Reddit.
- Travel publications: Coverage in Conde Nast Traveller, Wanderlust, The Telegraph Travel, or specialist outlets in your niche.
- Directory listings: Presence in ABTA, AITO and other authoritative travel directories.
4. Invest in review signals
Review volume, recency and sentiment directly influence AI recommendations. A brand with 2,000 TripAdvisor reviews from the last 12 months sends a far stronger signal than one with 200 reviews from three years ago. Make review generation part of your customer journey, not an afterthought.
Frequently Asked Questions
What is AI-powered travel search?
AI-powered travel search refers to how platforms like ChatGPT, Google Gemini, Claude and Google AI Overviews generate direct travel recommendations instead of showing a list of links. Travellers ask natural-language questions and receive synthesised answers drawn from reviews, forums, news and third-party sources across the web.
How do travel brands appear in AI search results?
AI models recommend travel brands based on third-party evidence — reviews on TripAdvisor and Trustpilot, Reddit discussions, travel publication coverage, and structured data on your website. Brands with strong review signals and broad third-party citation footprints are far more likely to be recommended.
Can I use AI tools to get personalised travel recommendations?
Yes. Tools like ChatGPT, Google Gemini and Perplexity now function as AI-powered travel search platforms. You can ask detailed questions — "best family beach holiday in Europe under 3,000 pounds in August" — and get personalised recommendations. However, the quality depends on which brands have built sufficient third-party authority for AI to trust and cite.
Why do AI travel recommendations keep changing?
AI models update their knowledge regularly, and the sources they draw from shift over time. A brand recommended today may not appear tomorrow if a competitor gains stronger review signals, fresher content, or more third-party mentions. We have seen brands lose AI visibility within weeks of a competitor launching a PR campaign that generated third-party coverage. This is why continuous monitoring matters.