Written by Shamus Smith, Founder, LuxDirect
LuxDirect analysed more than 9,000 valid AI responses across four major platforms in April 2026: ChatGPT, Gemini, Grok, and Perplexity, with Claude reported separately on a partial sample. The study tested live web-enabled AI responses against the same 25 London luxury boutique hotels and query set used in LuxDirect's earlier London training-data study. The result was clear: live web search concentrated visibility rather than broadening it. The top five hotels captured 71.0% of all AI mentions, up from 57.2% in the earlier training-data study. The top four alone captured 64.3%. When a response named a booking route, OTA brands appeared more often than hotel direct sites. Half of all responses named no bookable route at all. All figures are from structured API queries, not consumer chat apps, and measure what AI responses named, not clicks or bookings.
Findings are based on structured query testing across four full-coverage AI platforms via API, with a fifth platform reported separately on a partial sample. For this study, "live web" refers to web-enabled API responses generated at query time, not static model memory or consumer app behaviour, and real-user behaviour in the consumer apps may differ. The training-data comparison draws on LuxDirect's earlier London study, which used the same 25 hotels and 45 prompts; that study covered six platforms including Google AI Mode, where this live web study covers the four with full coverage, so the comparison holds the hotels and queries constant rather than the exact platform set. This is not search ranking; it is how AI systems construct recommendations when drawing on the live web. It is the latest in LuxDirect's series of UK luxury hotel market studies, following London, Edinburgh, Cambridge, and Manchester.
Live web search concentrated visibility rather than broadening it
The pre-study hypothesis was simple. Live web retrieval should surface fresher, more varied sources, and so broaden the field of hotels AI recommends. The data showed the opposite.
Across more than 9,000 valid responses, the four full-coverage platforms produced 793 hotel mentions. The distribution is far from even. The top four hotels captured 64.3% of all AI mentions. The top five captured 71.0%. The bottom five collectively held 1.5%.
In the earlier London training-data study, the top five held 57.2% of mentions. Live web search did not dilute that concentration; it intensified it, a 13.8 point increase at the top five. Observed patterns suggest live web retrieval amplifies hotels that already carry dense reference material, rather than levelling the field.
Full visibility rankings: all 25 properties
Hotel identities are anonymised. Full named reports are available to clients.
| Rank | Property | ASOV | Mention rate |
|---|---|---|---|
| 1 | Property A | 23.0% | 41% |
| 2 | Property B | 16.5% | 29% |
| 3 | Property C | 13.6% | 24% |
| 4 | Property D | 11.2% | 25% |
| 5 | Property E | 6.7% | 15% |
| 6 | Property F | 4.7% | 11% |
| 7 | Property G | 3.8% | 8% |
| 8 | Property H | 3.2% | 7% |
| 9 | Property I | 2.6% | 6% |
| 10 | Property J | 2.5% | 6% |
| 11 | Property K | 2.4% | 5% |
| 12 | Property L | 1.5% | 3% |
| 13 | Property M | 1.1% | 3% |
| 14 | Property N | 1.0% | 2% |
| 15 | Property O | 0.9% | 2% |
| 16 | Property P | 0.9% | 2% |
| 17 | Property Q | 0.8% | 2% |
| 18 | Property R | 0.8% | 2% |
| 19 | Property S | 0.8% | 2% |
| 20 | Property T | 0.6% | 1% |
| 21 | Property U | 0.6% | 1% |
| 22 | Property V | 0.5% | 1% |
| 23 | Property W | 0.4% | 1% |
| 24 | Property X | 0.0% | 0% |
| 25 | Property Y | 0.0% | 0% |
Top four share: 64.3% · Top five share: 71.0% · Bottom five share: 1.5%
Two properties registered zero mentions across all responses, on every platform and in every phase, despite both being established London luxury hotels. AI recognition of these properties exists; AI recommendation does not.
Mean mention rates compressed under live web search
Live web mode produced materially fewer target-hotel mentions per query than training mode, while concentrating the mentions it did produce among a small group of anchor properties. Across the four full-coverage platforms the most generous, Gemini, named a target hotel in 10.5% of responses; the strictest, ChatGPT, in 6.2%. The commercial reading is direct: as more travellers move to AI tools that search the live web, the visibility gap between the top four and everyone else widens, not narrows.
Platform divergence: the same hotel can be strong on one platform and absent on another
A hotel's AI visibility is not a single number. It varies substantially by platform. The table below shows the percentage of each platform's responses that mentioned each property, across all query phases combined.
| Property | ChatGPT | Gemini | Perplexity | Grok |
|---|---|---|---|---|
| Property A | 27% | 48% | 37% | 51% |
| Property B | 22% | 28% | 30% | 35% |
| Property C | 28% | 33% | 24% | 24% |
| Property D | 24% | 29% | 26% | 23% |
| Property E | 10% | 18% | 18% | 15% |
| Property G | 6% | 16% | 6% | 5% |
| Property K | 3% | 11% | 3% | 4% |
Selected properties shown. Figures are the percentage of that platform's responses naming each property across all phases.
Gemini was the most generous platform in this study, surfacing the leading property in 48% of its responses against 27% on ChatGPT. Several mid-table hotels showed the same split: visible on Gemini, near-absent on ChatGPT or Grok. The platform a traveller opens determines whether a hotel exists in their consideration set at all. Monitoring a single platform produces a misleading picture of where a hotel stands.
Booking-intent queries surface OTAs more than direct
Routing behaviour shifted sharply by query phase. The study classified every response by which booking routes it named: no bookable route, an OTA brand only, the hotel's own site only, or both. Important: this measures what the AI's response text mentions, not where any booking goes. We did not track bookings, clicks, or transactions.
| Routing outcome | Share of responses |
|---|---|
| No bookable route named | 49.1% |
| OTA brand only | 18.1% |
| Hotel direct only | 13.7% |
| Both | 17.0% |
Two findings stand out. Half of all responses named no bookable route at all. When a response did name one, an OTA brand appeared in 35.0% of responses against a direct site in 30.7%, and the overlap (responses naming both) is large, so this is about which sources surface, not a measured booking split.
That balance shifts at booking intent. An OTA brand appeared in 35.0% of discovery-phase responses, 21.6% of comparison-phase, and 47.1% of booking-intent responses. The hotel may earn the mention at discovery; at booking intent, an OTA brand becomes more likely to appear alongside or instead of the direct site.
| Query phase | OTA presence |
|---|---|
| Discovery | 35.0% |
| Comparison | 21.6% |
| Booking intent | 47.1% |
OTA presence is the percentage of responses in that phase in which an OTA brand appeared.
The voice describing your hotel is increasingly a third party
When AI describes or points to a hotel, it leans on the sources it surfaces. On live web search, those sources are often OTA listing pages, because they are structured, current, and dense with the signals retrieval systems favour. A hotel's own website often is not. The result is an information hierarchy where the reseller's page is the one the retrieval layer cites, so the intermediary's description tends to shape how the hotel is presented.
There is one moment when this can be reversed. When a property first opens, resellers have not yet built out its listings, and the hotel's own site can briefly be the source AI cites most. That advantage erodes as resellers catch up, which makes a launch the right time to embed direct signals.
What hotels can influence
Retrieval systems reward clarity. A property that publishes accurate, structured, machine-readable facts gives AI something confident to anchor to.
- Confirm the website identifies the property as a hotel in its structured data.
- Ensure canonical references describe the current property accurately.
- Treat the direct site as a primary source AI should cite, not an afterthought.
None of this guarantees an outcome. AI systems evolve, and observations reflect current behaviour rather than permanent law. But the direction of influence is clear, and the launch window is finite.
Claude reported separately
Claude coverage was truncated mid-scan, leaving 292 valid responses against 2,272 each for the four full-coverage platforms. On its partial sample, Claude named a target hotel in 27.4% of responses, materially higher than the four-platform figures. This is reported separately and excluded from the cross-platform aggregates above, because comparing a partial sample against full-coverage platforms would distort the headline figures. The pattern is noted for completeness, not used as a study conclusion.
Summary
CS10-London Live Web analysed more than 9,000 valid AI responses across four platforms, testing live web search against the same 25 London luxury hotels and queries used in the earlier training-data study. Live web concentrated visibility rather than broadening it: the top five captured 71.0% of all mentions, up from 57.2% in training mode, and mention rates compressed sharply. When a response named a booking route, OTA brands appeared more often than direct sites, and half named no route at all. An OTA brand appeared in 35.0% of discovery responses, rising to 47.1% at booking intent. All figures are API-derived and measure what the AI's response named, not booking outcomes. The concentration pattern is established but not fixed, and independent properties that address their AI signals now will be in a materially stronger position than those that engage reactively.
Frequently asked questions
Did live web search broaden or concentrate AI hotel visibility in London?
It concentrated it. The pre-study hypothesis was that live web retrieval would surface fresher, more varied sources and broaden the field. The data showed the opposite. The top five hotels captured 71.0% of all AI mentions on live web search, up from 57.2% in the earlier training-data study. The top four alone captured 64.3%.
Why does AI surface OTA pages over a hotel's own website?
Retrieval systems favour sources that are structured, consistent, and machine-readable. OTAs invest heavily in this kind of content, while many hotel websites do not. In the study, when a response named a route, an OTA brand appeared in 35.0% of responses against a direct site in 30.7%, rising to 47.1% OTA presence at booking intent.
When in the query journey do OTAs appear most?
At booking intent. For discovery and comparison queries, OTA presence sat at 35.0% and 21.6% of responses. When a guest signalled transactional intent, OTA presence rose to 47.1%. This measures OTA presence in the response text, not a measured booking destination.
Why is a hotel's launch window important for AI visibility?
When a property first opens, resellers have not yet built out their listings, so the hotel's own site can briefly be the source AI cites most. This advantage tends to erode as resellers catch up. Embedding accurate, structured direct signals during the launch window helps a property hold that lead for longer.
How many London hotels had zero AI visibility in the study?
Two of the 25 targeted properties registered zero mentions across all responses, on every platform and in every phase. Both are established London luxury hotels. AI recognition of these properties exists, but AI recommendation does not.
LuxDirect sits between your hotel and the AI discovery layer. We align what AI remembers about you with the hotel you actually run.
Every week, we systematically monitor how six leading AI platforms recommend your hotel across high intent guest searches. We show you where AI is diverting guests to OTAs, where competitors are outperforming you, and where your positioning is weak or underrepresented. Then we resolve the structural issues driving it and strengthen your direct booking position within the AI layer, systematically reducing dependency on OTA routing.
You do not need an internal technical team. LuxDirect operates as a visibility concierge. You approve. We execute.
Starting at £99 per month. The average luxury OTA commission runs between 18% and 22%. If LuxDirect recovers just one booking per month from AI mediated OTA routing back to your direct site, the service has paid for itself several times over.
Run AI Audit →