10th March 2026: Here are few of the stories that caught our attention over the week with some of our takeaways
Story 1
What Gen Z Actually Wants From a Hotel (It’s Not Your Loyalty Programme)
According to Team Cafeteria’s ‘2026 Travel Reset’ report — based on 197 hours of discourse among more than 900 Gen Z participants — this cohort is emerging as one of the most commercially consequential traveller segments of the next decade, and their priorities bear little resemblance to those of the generations that preceded them.
The headline finding is what the report calls the ‘Coach to Michelin’ model. Gen Z aggressively trades down on flights and transport — 37% of 18–21-year-olds actively seek the cheapest safe flight option — in order to maximise spending on food (50% splurge) and experiences (51% splurge). Hotels occupy a curious middle ground: 29% splurge on accommodation, 28% actively seek the budget option, and the remainder sit somewhere in between. The lodging decision, in other words, is rarely the centrepiece of the trip.
Direct Booking Is Gaining Traction — For the Right Reasons
One in four Gen Z travellers now bypass OTAs entirely, booking directly with airlines and hotel websites. This is not, however, a loyalty-driven behaviour in the traditional sense. The motivation is control and reliability — they want to know exactly what they’re getting and who to contact if something goes wrong. This has direct implications for how hotels position their direct channel: the conversation should be less about points and more about certainty.
That said, the report confirms that traditional loyalty programmes are the least valued hotel perk by a wide margin — 65% of Gen Z rank them last. What they actually want: free breakfast, fast Wi-Fi (viewed by 63% as a non-negotiable baseline), early check-in, and late check-out. These are tangible, immediate, and easy to deliver.
Human Service Still Wins
Perhaps the most commercially counterintuitive finding for an industry investing heavily in self-service technology: 31% of Gen Z cite ‘friendly staff’ as the single most memorable element of a great hotel stay — twice the number who cite operational efficiency. The stories that travel furthest in their social networks are those involving staff who go beyond the expected: free treats, creative problem-solving, genuine warmth. In an era of chatbots and automated check-in, this cohort still responds powerfully to human connection.
The booking discovery process is equally telling. TikTok and Pinterest dominate the inspiration phase — particularly among female travellers (68% vs 44% of males) — but decisions are only finalised after rigorous cross-referencing via Google Reviews, Reddit, and peer validation. For hoteliers, this means the quality of your digital footprint across review platforms is inseparable from your distribution performance. A compelling TikTok presence means little if the reviews don’t hold up under scrutiny.
Japan has emerged as the definitive ‘must-do trip’ for this generation, with Europe — led by Italy, France, and Greece — remaining the dominant collective region. As this cohort moves toward greater financial independence, hotels that understand their value framework — not just their demographics — will be best positioned to earn their business.
Story 2
The US Government Is Now Investigating AI Pricing in Travel
As reported by PhocusWire, the U.S. House Committee on Oversight and Government Reform has launched a formal investigation into the use of artificial intelligence and consumer data to set what it terms ‘surveillance pricing’ — the practice of using harvested personal data to identify a consumer’s maximum willingness to pay and price accordingly.
The committee, chaired by James Comer, has issued formal document requests to the CEOs of Booking Holdings, Expedia Group, Lyft, and Uber Technologies. The central concern is not dynamic pricing per se — the committee explicitly acknowledges the legitimate role of AI in aligning prices with demand and operational costs — but whether personal data is being used to charge individual consumers more than others for identical services, without their knowledge.
The ‘Black Box’ Problem
The committee’s core objection is one of transparency. When a consumer sees a price, they have no visibility into how it was calculated, what data was used, or whether someone sitting next to them on the same flight paid significantly less for an identical seat. The committee cited a Washington Post investigation into Uber’s pricing in which the same trip, taken within the same timeframe by two different users in New York, showed a 221% price difference.
The companies under investigation have pushed back firmly. Expedia Group stated it does not increase prices based on user data or personal characteristics. Uber described its pricing as purely reactive to marketplace factors such as driver availability and proximity. Delta Air Lines, responding to a parallel Department of Transportation inquiry launched in August 2025, stated that its AI pricing partnership with Fetcherr does not produce individualised prices based on personal data.
Why This Matters to Hotel Distribution
The investigation has broader implications for the hotel industry, even for those not directly named. Revenue management platforms that use consumer behavioural data — browsing history, device type, repeat visit patterns — to influence rate presentation are operating in territory that is attracting increasing regulatory scrutiny. The practice of showing different rates to different users based on inferred willingness to pay sits in a similar conceptual space to what Congress is currently probing.
The outcome of this investigation is unlikely to be swift. But the direction of travel — greater regulatory interest in algorithmic pricing opacity — is now clear. Hotels and their technology partners would do well to assess their own data practices now, rather than waiting for the scrutiny to arrive at their door.
Story 3
ChatGPT Stepped Back From Booking — What That Means for Hotels
Two developments reported this week — one from Travolution, one from Hospitality.today — together paint a picture of where AI-driven hotel distribution is actually heading, as distinct from where it was hoped to go.
On 6th March, Lighthouse announced the launch of ‘The Hotels Network,’ a direct booking application integrated into the ChatGPT App Directory. Powered by the Connect AI engine and utilising the Model Context Protocol (MCP), the app creates a real-time data bridge between a hotel’s own systems and the ChatGPT interface. Hotels retain full control over their brand narrative — verified descriptions, imagery, live rates — and the model operates on a flat-fee subscription with zero commission. When a user queries ChatGPT about a property indexed by Connect AI, they are routed directly to the hotel’s official booking link.
OpenAI Steps Back From the Transaction
Simultaneously, OpenAI quietly stepped back from its earlier ambitions to facilitate native purchases within ChatGPT. Rather than closing the transaction itself, the platform has opted for a ‘recommend and send’ model — directing users elsewhere to complete their booking. The commercial logic is straightforward: managing global payment rails, merchant agreements across hundreds of jurisdictions, real-time inventory synchronisation, and the disputes and cancellations that follow is an extraordinarily complex undertaking. OTAs have spent decades and billions of dollars building exactly this infrastructure, and it serves as a formidable barrier to entry.
Markets read the signal clearly. Booking Holdings shares rose more than 8.5% on the news. The OTA moat — operational complexity — held firm, at least for now.
The Metasearch Parallel
Industry analysts have drawn a direct comparison to the emergence of metasearch. When Kayak, Trivago, and Google Hotel Search arrived, many predicted they would commoditise OTAs and restore balance to hotel distribution. Instead, Booking.com and Expedia absorbed those channels, pouring marketing spend into dominating visibility there just as they had on Google Search. The risk with AI discovery tools is the same: OTAs have both the financial resources and the operational urgency to do it again.
As Markus Busch, Editor of Hospitality.today, noted this week, the window in which a genuinely neutral AI transaction layer might have given hotels direct access to guests — without Booking.com as the obligatory intermediary — was always narrow. It has now narrowed further.
For hoteliers, the practical implication is time-sensitive. Google is currently developing agentic booking capabilities within Gemini. Amazon is building Alexa+ with transaction capabilities. Apple is integrating Gemini into a rebuilt Siri across two billion devices. OpenAI’s retreat from native transactions is a pause, not a permanent exit. Hotels that use this window to invest in direct booking infrastructure — clean data, indexed content, fast and reliable booking engines — will be better placed when the next cycle of AI-driven commerce matures. Those that don’t will find themselves once again dependent on whoever controls the interface.
Story 4
Kayak’s New Campaign Is Really About the Mental Cost of Booking Travel
Marketing Dive reports that Kayak launched a new brand platform in February 2026 called ‘Got That Right,’ developed with agency Rethink — and while it is ostensibly a campaign about price comparison tools, it is more accurately a response to a specific psychological shift in how people experience the booking process.
The insight driving the campaign is blunt: 66% of consumers find booking travel stressful, and approximately half of Americans report feeling overwhelmed by the volume of online travel advice available to them. The problem, as Kayak’s SVP of Global Brand Marketing Carolina Montenegro put it, is not a shortage of information — it is an excess of it. Social media travel hacks, influencer itineraries, endless review threads, and algorithm-driven ‘best of’ lists have made the pre-booking phase more cognitively demanding than ever before.
The Campaign’s Demographic Target
‘Got That Right’ is aimed squarely at consumers aged 25 to 45, with two 30-second hero executions targeting millennial-specific behaviours. One lampoons the tendency to feel more comfortable completing large purchases on a desktop computer than on a mobile device. The other depicts a mother who has absorbed so much influencer-speak from scrolling travel content that she can no longer communicate normally — until she uses Kayak and returns to rational decision-making. The humour is gentle, but the underlying diagnosis is pointed: information overload has become a genuine friction point in the booking funnel.
The campaign arrives during a period of contraction for international travel. YouGov data cited in the Kayak research shows that 43% of Americans who travel internationally reported travelling abroad less in 2025 than the year before, with economic concerns and rising costs cited as primary drivers. The emphasis on price comparison tools within ‘Got That Right’ is a direct acknowledgement of this anxiety.
The Broader Commercial Context
For hotels, there is a useful read-across here. If the booking process itself has become a source of stress — and the data suggests it has — then anything that reduces friction and increases consumer confidence in their decision is commercially valuable. The case for direct booking channels that are genuinely fast, clear, and easy to use has never been stronger. Complexity and choice overload benefit aggregators. Simplicity and clarity benefit direct.
Kayak’s parent company, Booking Holdings, reported Q4 2025 revenue of $6.35 billion, a 16% year-on-year increase, with growth attributed primarily to AI-driven tools and agentic platforms across its brand portfolio. ‘Got That Right’ is as much a competitive positioning statement as it is a consumer-facing campaign: in a crowded and anxious market, the brand that makes the decision feel easiest wins.
Story 5
The AI-First Hotel: What It Actually Takes to Get There
BCG’s latest hospitality analysis, published this week under the title ‘AI-First Hotels: Leaner, Faster, Smarter,’ offers one of the more rigorous assessments of where the industry currently sits on the AI adoption curve — and the gap between ambition and reality is significant.
As of 2025, fewer than 10% of hospitality companies are classified as ‘future built’ — meaning they have cutting-edge AI capabilities generating measurable commercial value. A further 25% are actively scaling AI across multiple functions. The remaining 65% are either experimenting without a coherent framework or have yet to begin in any meaningful sense.
What the Leaders Are Actually Achieving
The report is careful to move beyond aspiration and focus on what is measurable. AI-driven dynamic pricing models are delivering RevPAR growth of up to 15% for early adopters. Predictive housekeeping models — trialled at properties including the Ritz-Carlton San Francisco — have produced a 20% improvement in room preparation efficiency. Winnow’s AI waste-tracking system, deployed at Four Seasons properties, achieved a 50% reduction in food waste within eight months. Marriott’s AI room-assignment engine processed 1.2 million assignments in seconds during its pilot phase. These are not projections. They are documented outcomes.
Three Barriers That Are Slowing Everyone Else Down
BCG identifies three structural obstacles that are holding the majority of the industry back. The first is what they call the investment dilemma: hotel companies consistently prioritise visible capital expenditure — lobby renovations, new F&B concepts, spa upgrades — over data infrastructure investment that produces no immediate visible return. Cleaning guest records, integrating disconnected systems, and building a central customer data platform is unglamorous work that can take six months or more before it generates any measurable output. It rarely wins a board presentation.
The second barrier is system fragmentation. The average hotel operation runs on a patchwork of disconnected platforms — PMS, POS, CRM, F&B management, loyalty — that were never designed to communicate with each other. 80% of hoteliers report spending up to two working days per week manually stitching reports together from these disparate sources. There is no meaningful AI deployment without first solving the data architecture problem.
The third is the skills gap. The travel and tourism sector has AI-competent workers representing just 2.9% of its total workforce — compared with 21.3% in telecoms, media and technology, and 7.5% in financial services. The gap is wide, though growing at 4.7% annually as of January 2025.
The Distribution Dimension
BCG makes one observation that deserves particular attention from commercial and distribution teams. As guests increasingly interact with AI-driven travel assistants rather than browsing OTA listings, brand equity is shifting from name recognition to algorithmic readability. The question is no longer just ‘how do we rank on Booking.com?’ but ‘can ChatGPT or Google AI Overviews accurately answer specific questions about our property?’ Hotels with clean, structured, verified data will appear in AI-generated recommendations. Those with fragmented or incomplete digital profiles will not.
The window to build the necessary infrastructure before AI-native commerce becomes the dominant booking channel is open, but it is not indefinitely open. The BCG framework offers a practical starting point: invest in data architecture first, automate the back office second, and build the guest-facing AI layer on top of a foundation that can actually support it.