The AI travel landscape shifted dramatically this week, with ChatGPT pulling back from direct hotel bookings, a landmark Cloudbeds report confirming deepening OTA dependence across 90 million bookings, and BCG publishing a blueprint for “AI-first hotels” that could bypass intermediaries entirely. If you run a hotel, tour operation, or any travel business, these developments directly affect your revenue strategy for the rest of 2026.
Let us break down each story and what it means for your property.
ChatGPT Abandons Direct Bookings: OTAs Win This Round
OpenAI made a significant strategic reversal in March. After months of testing integrated checkout features inside ChatGPT that would let travelers book hotels without leaving the chatbot, OpenAI shifted its focus from completing transactions to product discovery, routing actual checkout to third-party apps like Expedia and Booking.com.
The market reaction was immediate. Shares of both Expedia and Booking Holdings surged on the news.
What This Means for Hotels
This is both good and bad news:
The upside: OTAs are not being disintermediated tomorrow. Your Booking.com listing still matters. The billboard effect (where 65% of OTA visitors later Google the hotel directly) remains intact.
The downside: ChatGPT now functions as a discovery layer that funnels travelers into OTA ecosystems. If a traveler asks ChatGPT for “boutique hotels in Lisbon” and your property does not appear in that initial recommendation, you get pushed downstream into price-comparison territory where commissions eat your margins.
The key insight: discovery is now the battlefield. ChatGPT has 900+ million weekly active users globally. Even if a fraction of those users plan trips through the platform, the volume is staggering. Hotels that control their presence in AI discovery will capture demand before it reaches an OTA.
This is exactly why AI visibility for hotels has become a critical distribution strategy, not a marketing experiment.
Cloudbeds Report: 90 Million Bookings Reveal a Troubling Trend
The Cloudbeds 2026 State of Independent Hotels Report dropped last week with data from 90 million bookings across 180 countries. The findings paint a stark picture for independent operators.
The Numbers That Matter
| Metric | 2025 Change (YoY) | What It Means |
|---|---|---|
| Global Occupancy | -0.6% | Fewer rooms filled |
| ADR (Average Daily Rate) | -5.8% | Lower prices per night |
| RevPAR | -5.4% | Revenue per room declining |
| OTA Dependency | Increasing | More bookings through intermediaries |
| EMEA ADR | +6.0% | Europe bucking the trend |
| Asia Pacific RevPAR | -17.5% | Steepest regional decline |
The central finding: independent hotels are losing ground to OTAs. While branded chains maintain their positioning through massive marketing budgets and loyalty programs, independents are getting squeezed from both sides: declining rates and rising acquisition costs.
Adam Harris, CEO of Cloudbeds, put it directly: “With AI reshaping discovery, OTA dependence deepening, and margin pressure mounting, independent lodging has never needed clarity more.”
The one bright spot? EMEA, where ADR rose 6.0% and RevPAR advanced 3.9%. If you operate in Europe, you are in the strongest regional market for independents right now. But even that advantage erodes if your distribution strategy depends primarily on OTAs taking 15-25% of every booking.
BCG: AI-First Hotels Are Coming
Boston Consulting Group published a major report on AI-first hotels that reads like a roadmap for the next five years of hospitality. The key prediction: OTAs may face disintermediation not from ChatGPT directly, but from AI providers embedded inside banks and credit card companies.
Here is the scenario BCG outlines:
- AI providers partner with global distribution systems (GDS) via commercial API agreements
- These AI assistants get embedded into banking and credit card apps
- A traveler asks their bank’s AI assistant to plan a trip
- The AI accesses hotel inventory directly, bypassing both Google and OTAs
- The booking happens inside the financial app, with the credit card already on file
This is not science fiction. It is a logical extension of what Apple, Google, and financial institutions are already building. When your credit card app can recommend, book, and pay for a hotel in a single conversation, the traditional OTA search-compare-book funnel becomes obsolete.
The Implications for Independent Hotels
The BCG report reinforces something we have been saying at palmtree.ai: your hotel’s content needs to be machine-readable, not just human-readable.
Current hotel descriptions on OTAs consist of a few dozen structured fields: star rating, location, amenities, price. But AI queries are far more nuanced. A traveler asking for “a hot spring hotel with a view of Mount Fuji that is quiet enough for remote work” cannot be matched with structured fields alone.
Hotels that invest in rich, narrative content, structured data (schema markup), and AI-readable formats like llms.txt will be the ones AI systems can actually recommend. Those relying solely on OTA listings with generic descriptions will remain invisible to the next generation of booking interfaces.
Gemini Overtakes Perplexity as #2 AI Referral Source
In a development that flew under the radar, March 2026 data shows Google Gemini now accounts for 8.65% of AI chatbot referrals to websites, surpassing Perplexity at 7.07%. ChatGPT still leads, but the gap has narrowed from 22x to roughly 8x since October 2025.
Why This Matters for Travel Businesses
| AI Platform | Referral Share | Trend | Travel Relevance |
|---|---|---|---|
| ChatGPT | ~65% | Leading but shrinking share | Booking.com/Expedia integration |
| Google Gemini | 8.65% | Rapidly growing | Tied to Google Hotels, Maps |
| Perplexity | 7.07% | Growing but overtaken | Strong for research queries |
| Claude | ~5% | Steady | Detail-oriented travelers |
| Other | ~14% | Fragmented | Grok, Meta AI, others |
For hotels, this means you cannot optimize for just one AI platform. A traveler might discover your property through Gemini (which ties into Google’s hotel ecosystem), research it on Perplexity, and then ask ChatGPT for a final recommendation before booking.
This multi-platform reality is why tracking your AI Travel Score across all major engines is not optional anymore. A property that scores well on ChatGPT but is invisible to Gemini is leaving revenue on the table.
Revenue Hub: “AI Is Reshaping Hotel Demand”
A Revenue Hub analysis published this week crystallized the distribution challenge perfectly:
“They ask ChatGPT for boutique hotel recommendations in Mexico City. They ask Gemini to plan a weekend in Barcelona. They ask Perplexity to compare options for a design-focused stay in Copenhagen. And they get one answer. A shortlist. A recommendation. A curated set of options. If your hotel is not part of that answer, you are pushed downstream into price comparison environments where OTAs dominate.”
This is the clearest articulation of the new distribution reality. AI has moved the decision point earlier in the traveler journey. By the time someone reaches Booking.com, the shortlist is already made. The OTA becomes a transaction tool, not a discovery tool.
The hotels that appear in that initial AI-generated shortlist capture demand at zero commission. The ones that don’t are left competing on price inside OTA ecosystems, paying 15-25% for the privilege.
What Zero-Click Means for Hotel Marketing
Adding to the urgency, recent data from DigitalApplied shows zero-click searches now account for 60-83% of all queries, depending on whether AI Overviews are present.
For travel businesses, this means:
- Your website traffic from Google will continue declining. Travelers get answers directly in AI responses without clicking through.
- Traditional SEO alone is no longer sufficient. You need Generative Engine Optimization (GEO) to appear in AI-generated answers.
- Content that AI can cite becomes your new organic traffic. Blog posts, FAQ pages, and structured data that AI engines reference drive visibility even when users never visit your site.
This is why building a GEO content strategy is now as important for hotels and tour operators as having a booking engine.
The Big Picture: Three Scenarios for Hotel Distribution in 2026-2027
Based on this week’s developments, here is how the distribution landscape could play out:
Scenario 1: OTAs Maintain Dominance (Most Likely Short-Term)
ChatGPT’s retreat from direct bookings reinforces OTAs as the transaction layer. Hotels continue paying 15-25% commissions. AI platforms drive discovery but funnel bookings through existing intermediaries. Independent hotels see margins continue to tighten.
Probability: 60% for the next 12 months.
Scenario 2: AI Platforms Enable Direct Booking (Medium-Term)
Companies like Lighthouse and Hotels Network (which launched direct booking inside ChatGPT in March 2026) create infrastructure for hotels to accept bookings directly through AI interfaces. Early adopters capture high-intent travelers at near-zero acquisition cost.
Probability: Growing through 2026-2027.
Scenario 3: Financial AI Disrupts Everything (BCG’s Vision)
AI embedded in banking and credit card apps integrates hotel inventory via GDS APIs. The entire search-compare-book funnel collapses into a single conversational interaction inside apps travelers already use daily. OTAs, Google, and even ChatGPT become less relevant.
Probability: 2-3 year horizon, but foundations being laid now.
What Hotels Should Do Today
Regardless of which scenario plays out, the winning strategy is the same:
- Build AI visibility now. Make your property discoverable across all major AI platforms.
- Invest in rich, narrative content. Move beyond structured fields. Tell your property’s story in ways AI can understand and recommend.
- Implement technical GEO. Schema markup, llms.txt, FAQ content, and structured data are no longer optional.
- Track your AI presence. Know where you appear (and don’t) across ChatGPT, Gemini, Perplexity, Claude, and Grok.
- Stay on OTAs, but shift the ratio. Use OTAs for visibility and reviews while systematically moving bookings to direct channels.
The Week Ahead
Next week, watch for:
- Expedia’s Q1 earnings (April 8) for data on AI-driven referral volume
- Google I/O announcements expected to expand Gemini’s travel capabilities
- More Lighthouse/Hotels Network adoption data on direct AI bookings
- Cloudbeds report deep dives from industry analysts
The travel distribution landscape is being rewritten in real time. The hotels that adapt fastest will be the ones paying the least in commissions and capturing the most direct revenue.
FAQ
Is ChatGPT still relevant for hotel bookings after pulling back from direct transactions?
Absolutely. ChatGPT remains the dominant AI discovery platform with 900+ million weekly users. It has shifted from completing transactions to product discovery, meaning it still influences where travelers book. The difference is that it now routes bookings through Expedia and Booking.com rather than handling checkout directly. Your property still needs to appear in ChatGPT recommendations.
How much do OTA commissions actually cost independent hotels?
Based on the Cloudbeds 2026 report and industry data, OTA commissions typically run 15-25% per booking. Booking.com averages around 15%, while Expedia ranges from 15-22%. For a hotel doing $50,000/month through OTAs, that is $7,500-$12,500 in commissions. Shifting even 30% of those bookings to direct channels saves $2,250-$3,750 monthly.
Which AI platform should hotels prioritize for visibility?
No single platform is sufficient. ChatGPT leads with roughly 65% of AI referral traffic, but Gemini (8.65%) is growing rapidly and ties into Google’s hotel ecosystem. Perplexity (7.07%) is strong for research queries. A multi-platform approach using palmtree.ai’s AI Travel Score tracks visibility across all five major engines simultaneously.
What is GEO and why does it matter for travel businesses?
Generative Engine Optimization (GEO) is the practice of making your content visible to AI engines like ChatGPT, Gemini, and Perplexity. Unlike traditional SEO which targets Google search rankings, GEO focuses on being cited and recommended in AI-generated answers. With zero-click searches hitting 60-83%, GEO is becoming the primary way travelers discover properties.
Can small independent hotels compete with chains for AI visibility?
Yes, and in some ways they have an advantage. AI engines favor unique, detailed content over generic chain descriptions. A boutique hotel with rich narrative content about its location, experience, and character can outperform a chain property with standardized listings. The key is investing in AI-readable content and technical optimization that makes your property easy for AI to understand and recommend.
The AI travel landscape is moving fast. To know exactly where your property stands across ChatGPT, Gemini, Perplexity, Claude, and Grok, check your AI Travel Score free at palmtree.ai.