The Machines Understand All Fifteen Travel Brands. Five Earn a First Place.

We asked four AI models to research fifteen travel and hospitality brands. Each brand therefore holds 60 recommendation moments. The median score is 69. Aman leads at 83. VRBO sits at 58.

Clarity is not the problem here. Every brand scores 15 or higher on Clarity. Seven brands score 18. The models can state what each of these brands is. Recommendation does not follow.

Eight of the fifteen brands hold a swell position. Their equity is real, and the machines under-reward it. The other seven are aligned. No brand in this category is recommended above its equity.

Five brands earn a first place. Ten brands earn none.

ABOUT THIS READING

We score five attributes out of 20 each. Findability, Clarity, Sentiment, Identity Fidelity and AI Recommendation sum to a score out of 100. Five tiers sort that score. AWESOME runs 83 to 100, STRONG 70 to 82, AVERAGE 56 to 69, WEAK 40 to 55, and INVISIBLE 0 to 39.

This is a repeat reading of the category. Four models answered fifteen buyer questions for each brand, which gives 60 recommendation moments per brand. A position label accompanies each brand. SWELL marks real equity that the machines under-reward. EBB marks a brand recommended above its equity, and ALIGNED marks agreement between the two.

Winners

Seven brands have a Q2 reading, and their median change was 0. Aman rose 8, VRBO 6 and Marriott 4. Hilton held, Four Seasons slipped 1, and Airbnb and CitizenM each fell 4. The other eight brands are new this quarter.

Aman leads at 83 and made the largest gain, 8 points. Identity Fidelity and AI Recommendation each rose 4. All four models name Aman first for the most beautiful places to stay, and it takes 11 first places in 60 answers. None names it for trust, service, family or business travel.

Aman's three Q2 bets did not produce the gain. The character bet split the models instead of uniting them, so the voice gain came from vocabulary alone. Skift published the scale story before Aman did, and the property catalog is still unbuilt.

Losers

VRBO closes the board at 58, though it rose 6 points, the second-largest gain. Identity Fidelity rose 4 and Clarity 3, while AI Recommendation moved from 1 to 2. The models name VRBO on 3 of 15 questions and never first. None names it for family vacations.

The Q2 bets explain both halves. The voice bet half shipped, and the models now return VRBO's mission and owned phrases. The recommendation bet did not ship, because the content it called for does not exist.

Marriott rose 4 points to 80. AI Recommendation rose 4 and Identity Fidelity 2, while Clarity and Sentiment each slipped 1. It appears on 9 of 15 questions and comes first in 5 answers, strongest on business travel. Our Q2 language bet half delivered. Every model now names the right company and its brand ladder, but explains it in fee and scale terms.

Airbnb and CitizenM fell furthest, 4 points each. Airbnb's AI Recommendation fell 5 points, to 11. All four models name it when asked which travel brands come to mind, and one answer in 60 puts it first. Two Q2 bets line up with its other moves. The voice bet half landed and Identity Fidelity rose 3. The narrative bet started but did not land, and Sentiment fell 2.

CitizenM fell 4 points to 60, mostly on Findability, down 3, after citizenm.com began redirecting to Marriott. No model names citizenM in any of the 60 answers. Its Q2 niche bet never ran, because the site it relied on was redirected first.

The tie between bets and moves has a limit. A move that lands on a bet's target does not prove the bet caused it, or that ignoring it did.

The Leaderboard.

The Shelf Holds 153 Brands. The Cohort Holds 26.5% of the Choices.

The models made 404 choices across the fifteen buyer questions. Those choices named 153 brands. The fifteen audited brands hold 26.5% of them. The rest belong to names outside the cohort.

The largest outside names are not hotel groups. Singapore Airlines takes 13 choices. Southwest Airlines takes 11. Belmond takes 10, Delta Air Lines takes 10, and Intrepid Travel takes 10. G Adventures, Rosewood and Six Senses each take 9.

On these fifteen questions, the models name airlines, tour operators and small luxury collections beside the hotel groups. A hotel group that compares itself only with other hotel groups misses that competition.

15 buyer questions x 4 models / audited brands in colour, every other brand in a stepped grey. Toggle between Cohort Only and Full Shelf.

KEY FINDING 01

Aman Holds the Top Tier Alone, and It Does Not Lead on Findability

Aman scores 83. It is the only AWESOME brand in the category. Six brands sit in STRONG, and eight sit in AVERAGE. Four Seasons and Ritz-Carlton follow at 81, and Marriott follows at 80.

Aman scores the maximum of 20 on AI Recommendation. It is also named first in 11 of its 60 moments. No other brand comes close on that count. Those are two separate facts, and Aman is the only brand that holds both.

Findability is not the source of the lead. Aman scores 13 there, and seven brands in the cohort score higher. Marriott, Airbnb, Shangri-La and Accor each score 15. The lead rests on recommendation. Aman's Sentiment score of 16 sits below Four Seasons at 18 and Mandarin Oriental at 17.

KEY FINDING 02

Shangri-La and CitizenM Are Never Named

Shangri-La scores 15 on Findability, the joint highest in the cohort. It scores 17 on Clarity and 16 on Sentiment. It scores 0 on AI Recommendation. Across 60 moments, no model named it once. CitizenM shows the same shape with 18 on Clarity and 17 on Identity Fidelity, and 0 named.

Accor repeats the pattern at scale. It also scores 15 on Findability. It scores 1 on AI Recommendation and takes 3 named moments. Being easy to find does not make a brand easy to choose.

Eight brands carry this gap. Kimpton scores 16 on Identity Fidelity and 2 on AI Recommendation. Mandarin Oriental scores 17 on Sentiment, second in the category, and 7 on AI Recommendation. Airbnb shows the same gap higher on the board, with 74 overall, 18 on Clarity, and a single first place.

Brand Fundamentals vs Recommendation.

KEY FINDING 03

Airlines and Tour Operators Take Places on the Travel Shelf

The two largest names outside the cohort are airlines. Singapore Airlines takes 13 choices and Southwest Airlines takes 11. Delta Air Lines takes 10. Both of the two largest airlines hold more choices than Hyatt, which takes 10.

Tour operators hold the next block. Intrepid Travel takes 10 choices and G Adventures takes 9. These companies sell a trip, and the models name them on the same shelf as the hotel groups.

Luxury names outside the cohort take places too. Belmond takes 10 choices. Rosewood takes 9 and Six Senses takes 9. Inside the cohort, Mandarin Oriental scores 67 and Shangri-La scores 62, and both sit in the AVERAGE tier.

KEY FINDING 04

Every Brand Is Understood. Ritz-Carlton Is the Most Distinct.

Clarity has a floor of 15 in this category and a ceiling of 18. Identity Fidelity runs wider, from 14 to 19. Ritz-Carlton scores 19, the highest Identity Fidelity score in the category. Airbnb and CitizenM follow at 17.

Five brands share the floor at 14. Hilton, IHG, Shangri-La, Accor and VRBO all score 14 on Identity Fidelity. The models can describe these brands. The models do not hold a sharp picture of what separates them.

Distinctness alone does not buy recommendation. CitizenM scores 17 on Identity Fidelity and 0 on AI Recommendation. Ritz-Carlton scores 19 and takes 15 on recommendation with 2 first places. A brand must be both distinct and chosen, and the second does not arrive with the first.

Understood vs Distinct.

KEY FINDING 05

Marriott Is Named Most Often. Aman Is Named First Most Often.

Marriott appears in 26 of its 60 moments, more than any other brand in the cohort. It takes 5 first places. Aman appears in 19 moments and takes 11 first places. Presence and primacy are different assets, and the two leaders hold one each.

IHG makes the split plain. It appears in 22 of 60 moments and scores 16 on AI Recommendation. It takes no first place at all. Hilton appears in 19 moments and also takes none.

A brand that the models name often but never place first is present without being chosen first. Four Seasons appears in 20 moments and takes 2 first places. Ritz-Carlton appears in 17 and takes 2. Aman and Marriott take 16 of the 21 first places in the category.

WHAT THIS MEANS

This category does not have a comprehension problem. The machines can state what every one of these fifteen brands is. Clarity never drops below 15. The problem sits one step later, at the moment the model must choose a name.

Eight brands hold equity the machines under-reward, and the work for them is to be chosen rather than explained. The wider shelf names 153 brands, so the competitive set includes airlines, tour operators and small luxury collections. Marriott shows that presence can be won at scale. Aman shows that primacy is won on a narrower and more specific claim.