Here are few of the stories that caught our attention over the week with some of our takeaways
Story 1
$22.6 Billion in a Single Day: What the Middle East Conflict Means for Hotel Commercial Performance
Reporting from uk.investing.com and Reuters confirms what many of us watched unfold in real time — the escalating conflict involving the United States, Israel, and Iran triggered the most severe single-day sell-off in travel stocks since the pandemic, wiping $22.6 billion in combined market value from 29 leading travel companies on 2nd March 2026.
The Geographic Exposure Problem
For hotel groups, the damage was not uniform — it tracked almost precisely to Middle East room count exposure. Accor fell approximately 9%, driven by the fact that roughly 10% of its global rooms sit in the Middle East, with luxury properties in Dubai and Bahrain directly in the line of fire. IHG dropped between 4% and 5%, its high concentration in Saudi Arabia and the UAE compounding matters alongside a concurrent CMA investigation at home. Meanwhile, Marriott and Hilton each declined around 3% — significant, but cushioned by the relatively small share of their portfolios tied to the region.
The Hub Closure Cascades
Dubai International Airport — which processed 92 million passengers in 2024 and ordinarily handles over 1,000 flights daily — remained closed for a third consecutive day. The airspace over Iran, Iraq, and Kuwait sat empty. Chinese airlines cancelled more than 25% of Middle East flights. Cathay Pacific pulled all regional services. Singapore Airlines suspended Dubai operations through 7th March. This is not a localised disruption; it is a systemic shock to the global transit network that underpins international travel demand.
The Oil Price Multiplier
Oil prices jumped as much as 13%, reaching their highest levels since January 2025. Jefferies analysts estimate that a 5% increase in fuel costs could reduce 2026 earnings for Delta and United by 5% to 10%. For American Airlines, profits could fall by approximately 35% under those same conditions. Fewer profitable long-haul flights means reduced premium cabin demand — and for hotels in gateway cities dependent on connecting traffic, that matters.
The Broader Commercial Implication
The Strait of Hormuz carried only five oil tankers on 1st March, against a recent daily average of approximately 60 (As of 3rd March the Strait is completely closed). S&P Global Research suggests that if this pace continues for a week, it will constitute the largest oil supply disruption in history. Hotels in the Middle East are managing force majeure scenarios in real time. For revenue managers globally, the short-term calculus has shifted — corporate travel budgets will be scrutinised, group business may defer, and leisure travellers in affected corridors will look for alternatives. The recovery timeline, analysts currently believe, extends across several weeks at minimum.
Story 2
From RevPAR to Revenue Per Square Metre: The Juyo Analytics 2026 Hospitality Trends Report
The Juyo Hospitality Analytics Trends Report 2026 provides one of the more grounded assessments of where hotel data strategy actually stands — and where the gaps remain painfully wide.
The Talent Problem Is Still the Data Problem
As of early 2025, 65% of hotels reported staffing shortages, with employment levels approximately 10% below pre-pandemic figures. The industry has historically under-utilised Business Intelligence tools not because the tools were inadequate, but because the people needed to interpret them were not there. The report’s response — upskilling existing staff, creating dedicated “Commercial Analytics Leader” roles, and tapping non-traditional labour pools — is directionally correct, though it requires sustained investment that many hotel companies have been reluctant to commit to.
Self-Service BI: The End of the Weekly Report
The shift toward interactive, self-service analytics platforms is arguably the most operationally significant development covered in the report. Hilton’s implementation of a centralised reporting hub to push key data to managers across the entire company is a telling case study. When on-property managers can adjust pricing or staffing based on live dashboard indicators rather than waiting for IT-generated reports, the entire commercial decision cycle compresses. The governance requirement — a single source of truth with consistent metric definitions — sounds straightforward but remains the piece that most hotel groups handle poorly.
Metrics That Actually Reflect Profitability
The report’s push to supplement RevPAR with profit per guest and revenue per square metre deserves serious attention. The first metric accounts for the actual cost of acquiring and servicing a booking — something that makes the difference between a high-occupancy quarter and a profitable one very clear, very quickly. The second treats the hotel as a real estate asset and forces honest conversations about whether meeting rooms, spas, or retail spaces are carrying their weight.
AI Oversight Is the Part Nobody Wants to Talk About
The report acknowledges that AI effectiveness is entirely dependent on data quality, and that human oversight — revenue managers reviewing AI suggestions to catch anomalies — remains non-negotiable. This is a more honest framing than most vendor narratives offer. AI-first analytics inverts the traditional BI model: rather than predicting questions in advance, hotels describe an objective in natural language and the system generates the analysis. That is genuinely useful. But the hotels that will extract value from it are the ones that have already done the unglamorous work of cleaning their data and defining their metrics consistently.
Story 3
The Tool Sitting in Every Revenue Manager’s Dashboard Is Now Under Investigation
The Guardian reports that the UK Competition and Markets Authority has formally launched an investigation into Hilton, IHG, and Marriott International, alongside real estate data firm CoStar — specifically examining whether the three hotel groups have been sharing competitively sensitive information through STR, CoStar’s widely used benchmarking platform.
What the CMA Is Actually Alleging
The regulator’s concern is not that hotels are using benchmarking data — it is that sharing occupancy rates, average daily rates, and RevPAR figures through a common third-party platform may reduce the natural uncertainty of a competitive market. By making it easier for rivals to anticipate each other’s pricing behaviour, the CMA argues, the incentive to compete aggressively on price is diminished. The result, in their framing, is coordinated behaviour that keeps rates artificially elevated for consumers. Companies found in breach face fines of up to 10% of total global annual revenue.
The Market Has Already Responded
IHG shares dropped as much as 5% on the FTSE 100 following the announcement. Hilton and Marriott each declined approximately 3% on the New York Stock Exchange. CoStar fell around 2%. IHG and Hilton both confirmed they are cooperating fully. CoStar expressed surprise that a platform it describes as longstanding — used by governments and companies for decades — is now under scrutiny. Marriott did not respond to requests for comment.
The Broader Precedent
This investigation sits within a wider CMA push to examine how technology and data-sharing affect competitive behaviour. The regulator recently secured £100 million for affordable housing programmes from seven major housebuilders after finding evidence of shared pricing information. The parallel is deliberate. The hospitality industry has long treated STR benchmarking data as an unremarkable operational input. That assumption is now being tested formally — and the outcome will likely set the framework for how third-party benchmarking tools are used across the sector.
What This Means for Revenue Management Practice
STR data is embedded in commercial workflows across the industry. Revenue managers use it to contextualise their own performance, track competitive positioning, and inform rate strategy. The CMA investigation does not suggest this is inherently illegal — but it does raise the question of whether the specific data inputs and the manner in which they are shared constitute a competitive harm. The next six months of evidence gathering will be worth watching very closely.
Story 4
Agentic Hotel Bookings: Why the Industry Needs to Stop Using the Term as Though It Means One Thing
Hospitalitynet publishes a piece that cuts through one of the more carelessly used phrases in hotel technology discussions right now — and the distinctions it draws have real commercial consequences.
Three Models, Three Entirely Different Problems
Ira Vouk identifies three distinct forms of agentic booking, and conflating them is where most hotel technology strategies currently go wrong.
AI-Assisted booking keeps the human in control throughout. A traveller uses ChatGPT or Gemini to narrow their options, but the final decision and the transaction itself remain with the person — either redirected to a brand site or completed through a universal commerce platform. The power shift here is in discovery, not in commerce.
AI-Mediated booking brings the interaction onto the hotel’s own digital infrastructure. An embedded agent on brand.com handles preferences, loyalty data, and promotional logic. The transaction completes within the hotel’s own environment. This model protects customer data and the payment flow — the hotel retains commercial control throughout.
AI-Executed booking — the genuinely agentic scenario — involves a user’s personal AI agent detecting a travel need autonomously, negotiating directly with a hotel’s AI agent, and delivering a confirmed booking to the human without their active participation in the transaction. This requires machine-verifiable portable identity and delegated payment credentials. It is a fundamentally different technical and commercial architecture.
Why a Chatbot Is Not Agentic Readiness
This is the point the industry most needs to hear. Deploying a chatbot on your website does not prepare your systems for AI-executed commerce. An API partnership with a consumer AI platform does not constitute readiness for autonomous agent-to-agent transactions. The infrastructure requirements for genuine A2A bookings — standardised machine-to-machine APIs, encoded policy frameworks, trust authentication layers, automated contracting logic — do not exist in most hotel technology stacks today.
The Distribution Implication
Each model changes the competitive dynamic differently. AI-Assisted models increase the importance of visibility on consumer AI platforms — the discovery layer now sits outside your control. AI-Mediated models reward investment in direct channel infrastructure. AI-Executed models will ultimately be won or lost on technical standards and the ability to participate in machine-verifiable trust frameworks. Hotels that treat all three as the same challenge will build for the wrong future.
Story 5
AI Is Winning the Inspiration Phase. OTAs Are Still Winning the Booking
Directo’s analysis of AI versus OTA performance between January 2025 and January 2026 provides some of the clearest early data on how conversational AI is actually reshaping travel demand — and where its current limits remain structural rather than temporary.
The Referral Numbers Tell a Specific Story
ChatGPT referral rates grew from near-zero to measurable levels across all tracked travel brands over the twelve-month period. Skyscanner reached 3.6% ChatGPT referral rates. Eurostar hit 3.0%. Vrbo reached 2.5%. OTAs like Booking.com show lower percentages, but their absolute referral volumes remain substantial given the scale of their user bases. Metasearch platforms lead because AI excels at precisely the task they were built for — comparison and structured discovery.
The 15–18% Overlap Finding
Between July 2025 and February 2026, approximately 15% to 18% of OTA sessions also included a ChatGPT interaction. Users are not choosing between the two — they are using AI to generate and refine ideas, then moving to OTAs for real-time pricing, availability verification, and final booking execution. The research loop is sequential, not competitive.
Where the Funnel Breaks
The data is unambiguous on this point. Of all travel-related AI prompts, 75% are informational. Only 5% are transactional. The AI funnel loses users after the shortlisting phase — and the reasons are structural. No reliable real-time inventory access. No trusted payment infrastructure within the chat interface. Fragmented inventory ownership. Significant unresolved questions around consumer trust and liability. These are not gaps that will close with a software update.
The Real Competitive Battle
The most important strategic insight from this analysis is the reframe it offers. The threat to OTAs is not disintermediation — AI is not going to replace the booking flow in the near term. The threat is the evolution of the interface itself. “Generative UI” — dynamically produced comparison tables, visual cards, and interactive maps — has the potential to bridge the gap between discovery and transaction in ways that current text-heavy chat interfaces cannot. The hotel groups and OTAs that will shape the next phase of distribution are the ones investing now in the handoff: the moment when an AI’s advisory recommendation converts into a confirmed booking, and who owns that moment.