When guests ask ChatGPT, Gemini or Claude where to stay, one hotel in your market may win the recommendation. Rose-Brook shows you who, why, and what to fix first — with every AI answer saved as evidence.
Captured benchmark answer · Birmingham, UK
3 models · 29 Jun 2026 · verbatim
A guest asked
“I’m planning a trip to Birmingham and want a hotel with strong local character and dependable service. Which hotels should I consider?”
Gemini answered
saved word for word
“Recommended hotels:”
Three names. The answer ended there.
The same question, three models
8 hotels named across all three answers. Birmingham has hundreds.
Rose-Brook read
Hotel du Vin appears first in two of the three captured model answers. The Grand leads the Gemini answer.
Evidence gap
The saved answer names hotels but gives no service or local-character proof for the ranking.
First action
Compare the public proof each leading hotel gives AI for those two guest priorities.
Hotels that weren’t named missed this recommendation — that does not prove a lost booking.
Guests ask this about your city every day. Rose-Brook runs the questions, saves every answer, and shows whether your hotel is in them — and what to fix if it isn’t.
How the benchmark works
01
We run the questions your guests actually ask — across ChatGPT, Gemini and Claude — and rank every hotel in your competitor field under the same conditions.
02
Every gap comes with the evidence behind it and the first practical action: the parking proof, the family-room detail, the arrival information AI could not find.
03
The benchmark re-runs on schedule. You watch your score move against the same field, and you see whether the fixes changed the answer.
Our UK benchmark sample
295 scored hotel observations across our accumulated UK benchmarks. Snapshot: 1 October 2026. Repeated hotel observations may be included; this is not a census of all UK hotels.
Next step
Your hotel, your real competitors, your city. The first read shows where you rank, which competitor is taking the recommendation, and the first fix worth making.
£39 Entry
Monthly ChatGPT benchmark.
£99 Pro
Weekly Pro coverage across ChatGPT, Gemini and Claude.