Overview
Stellagent released the Okinawa edition of its AI Hotel Recommendation Survey on July 9, 2026. We asked ChatGPT, Claude, Gemini, Perplexity, and Microsoft Copilot for Okinawa Main Island accommodation recommendations across five travel scenarios, collecting 125 responses and analyzing recommendation concentration as well as differences by scenario and AI service.
For more information, see the press release on PR TIMES.
The top five properties accounted for 44.2% of AI recommendations for Okinawa Main Island
Background
Generative AI is becoming a new entry point for travel planning by accepting a traveler's companions, budget, purpose, and transportation constraints in a single conversation. For accommodation providers, it is increasingly important to understand not only search rankings, but also the travel contexts in which AI recommends a property and how it explains that recommendation.
The survey used five scenarios — couples, families with children, first-time inbound visitors, groups of friends in their twenties, and multi-generation adult families — and recorded recommended properties, ranks, reasons, and booking paths.
Key Findings
- All 625 recommendation slots across 125 responses could be extracted as property names, covering 84 properties.
- The top five properties received 276 mentions, or 44.2% of all recommendation slots. This concentration was higher than the 31.1% observed in the Hakone edition.
- Halekulani Okinawa led with 122 recommendations and 60 first-place picks. Renaissance Okinawa Resort received 20 first-place picks and was particularly strong for families with children and multi-generation adult families.
- By AI service, Microsoft Copilot concentrated its recommendations on 20 properties, while Perplexity distributed its recommendations across 41.
Survey Overview
| Item | Details |
|---|---|
| AI services | ChatGPT, Claude, Gemini, Perplexity, Microsoft Copilot |
| Survey date | July 2, 2026 |
| Runs | 5 AIs × 5 scenarios × 5 runs, 125 responses (625 recommendation slots) |
| Data captured | Rank, property, rationale, best-fit traveler, booking path, and URL |
| Conducted by | Stellagent Inc. |
Implications for Hotel and Resort Operators
Operators should consistently publish concrete information about rooms, dining, activities, children's facilities, and access so AI can explain why a property fits a specific travel scenario. When recommendations concentrate on a small number of properties, it is also important to monitor the conditions under which a property enters the candidate set and identify missing information.
The full ranking, scenario-level and AI-level analyses, and methodology are available in the AI Hotel Recommendation Survey: Okinawa Edition.
Looking Ahead
Stellagent will continue conducting AI recommendation studies across regions and travel scenarios to make accommodation providers' visibility in AI and their connection to official booking channels measurable.