The Cloudbeds 2026 State of Independent Hotels Report documents a number that deserves serious attention: OTA share of total bookings in its independent-property dataset rose from 61.3% in 2024 to 63.4% in 2025. Portugal’s OTA share reached 79.7%. The report draws on 90 million bookings across tens of thousands of independent properties in 180 countries.
That is a distribution result. It does not, by itself, explain where the guest relationship began.
A property can invest in its booking engine, CRM, and loyalty program while remaining dependent on another party to introduce its next guest. The report’s aggregate figures do not establish which investments individual hotels made or whether those investments failed. They do raise a question channel share alone cannot answer: is the hotel building its own route to demand, or improving conversion after someone else has made the introduction?
The Industry’s Default Response
When data like this surfaces, the hospitality industry reaches for a familiar set of responses. Invest in the booking engine. Improve rate transparency. Build out the loyalty program. Increase direct booking incentives. Upgrade the CRM. Run better email campaigns.
These are not bad investments. They can improve conversion, guest experience, and repeat business. But a property whose constraint is demand origin needs to address that constraint directly.
A booking engine converts demand. CRM and loyalty can develop relationships with known guests. Email can introduce a property to a qualified new audience or reactivate an existing one. The role depends on the audience, the relationship, and the permission outcome. The medium alone does not determine whether the work is upstream or downstream.
When a hotel relies on intermediaries for introduction, conversion optimization alone does not resolve that dependence. Improving the path to purchase matters. So does establishing a path to the traveler before the next purchase is at stake.
What Channel Share Cannot Measure
The Cloudbeds distribution figures measure what share of total bookings closed through OTA and non-OTA channels. That is a useful performance indicator and a starting point for diagnosis. It does not identify the origin or permission history of each guest relationship.
A traveler can discover a hotel through an OTA and book directly. Another can know the property already and still book through an OTA. The transaction channel and the relationship’s origin are different variables. One cannot be inferred reliably from the other.
When a traveler encounters a property inside an OTA interface, the platform controls that comparison environment: the options displayed, the price presentation, and the route to purchase. The hotel may receive the reservation without receiving the same view of the traveler’s search and comparison behavior.
The hotel can still earn a direct, permissioned relationship before, during, or after the stay. Doing so creates a route back to the guest. It does not retroactively change where the original relationship formed, and the booking record itself is not marketing permission.
AGR’s structural diagnosis is that a hotel can improve direct conversion while leaving its dependence on intermediary introduction intact. Cloudbeds’ channel-share figures establish the distribution outcome; they do not prove that this mechanism explains every property’s result or that campaign gains inevitably disappear.
The information asymmetry underlying this argument is examined in The Lemons Problem: How Asymmetric Information Destroyed Luxury Hotel Demand, which applies Akerlof’s information-economics model to luxury hotel demand. It is a structural interpretation of the market, not a causal finding established by the Cloudbeds dataset.
The Compounding Problem Underneath the Data
An intermediary that serves many properties can observe demand across a wider set of searches and transactions than an individual hotel can see in its own reservation history. That cross-property perspective is an information advantage. A hotel’s stay records, however detailed, do not reconstruct every comparison made elsewhere.
The hotel also has information the platform may not have: the guest’s experience on property, preferences shared directly, and the quality of the relationship developed during the stay. The issue is not that hotels receive no intelligence. It is whether they connect what they know to valid permission and a reliable route to future engagement.
Repeated reliance on intermediaries can sustain that asymmetry when the hotel does not build its own introduction and relationship-development capability. Each booking then supports current revenue without necessarily improving the hotel’s ability to reach the next guest independently. That is the compounding problem AGR asks leadership to examine.
The report also identifies a changing discovery environment. Citing Phocuswright, Cloudbeds reports that the share of U.S. travelers using traditional search engines for trip planning fell from 51% in 2024 to 36% in 2025. It separately cites Cloudbeds research in which OTAs supplied more than half of the citations in AI-generated hotel recommendations. Those are discovery and citation findings, not measurements of where permissioned guest relationships formed.
In AGR’s July 29, 2026 Index fieldwork, which recorded 824 recommendation slots across six markets, two Michelin Guide list pages recurred as cited sources across ChatGPT’s ten Los Angeles answers. That is a documented example of third-party editorial material appearing in the answers travelers may use to compare hotels.
Cloudbeds recommends structuring property information so AI systems can interpret and surface it accurately, with consistent content across the wider digital environment. That is useful work. Accurate AI visibility and a hotel-controlled guest relationship still answer different commercial questions. AGR addresses human-mediated pre-transaction demand origin through Owned Demand Infrastructure and the public source environment relevant to AI representation through Knowledge Formation Optimization (KFO). KFO’s results are assessed through observable AI answers.
What Actually Changes the Dynamic
The strategic move is to give the hotel a route to qualified travelers that can become a voluntary, permissioned relationship. Editorial presence, advisor introductions, paid media, and email-led guest acquisition can contribute. The test is where the relationship forms, what permission the traveler grants, and whether the hotel controls the path back. Visibility or an introduction alone does not establish ownership.
AGR’s 250-room luxury hotel case study documents a six-month engagement compared with the same period a year earlier, at a flat $750 ADR. OTA share fell from 61.7% to 56.89%, while direct-controlled room revenue rose by $1,342,040. Cloudbeds reports booking share; this case calculates its channel split from room revenue at the stated flat ADR.
Existing-guest suppression followed by MD5-hashed email matchback confirmed 251 bookings and 627 direct room nights from campaign recipients absent from the hotel’s supplied pre-campaign guest file. That is the booking-level evidence. A prior guest using a different email can escape suppression, and matchback can miss forwarded offers, telephone bookings, or bookings under another email. The 627 room nights are a measured floor on attributable campaign activity.
The wider channel attribution rests on additional evidence. AGR reports that the engagement was the only new demand-generation variable, with marketing spend, advertising channels, allocation, and ADR held constant year over year. On those stated controls, the matchback results, and the corresponding channel movement, AGR attributes the broader shift to the program. That does not mean every shifted room night was individually matched or that recipients never compared an OTA.
The improved channel mix equated to $223,385 in avoided OTA commission, calculated on the post-period revenue base against the original OTA share. Actual commission expense was $161,663 lower than in the prior-year period. Those are different measures of the economic improvement. AGR’s distribution-costs CFO FAQ explains the wider acquisition-cost and margin questions using published sources, including Kalibri Labs and USALI.
The strategic lesson is to examine what remains after the campaign or transaction. Booking-engine improvements, CRM, and guest-experience investments can create lasting value. The demand-origin question is whether the investment also builds a permissioned relationship and a path to future demand the hotel can govern.
A hotel that funds OTA placement, rate competition, and booking conversion but leaves permissioned introduction unbuilt can become a better converter without becoming a more independent acquirer. The allocation changes when qualified audience access, permissioned identity capture, and hotel-controlled introduction environments become explicit investment priorities. AGR examines that strategic choice in Why Independent Luxury Hotels Are Competing on the Wrong Things, using Kim and Mauborgne’s Strategy Canvas as a diagnostic lens.
Owned Demand Infrastructure (ODI) is the framework that governs the pre-transaction demand origin layer: the layer that determines where a guest relationship first forms across hotels, resorts, and cruise lines, how traveler identity is captured before booking, and how a guest relationship becomes a first-party asset rather than an intermediated transaction.
ODI concludes at Condition 4, Identity Emergence, when a voluntary, permissioned relationship is established. Execution, conversion, attribution, and retention sit outside its formal Layers. Those operating responsibilities sit within the separate AGR Hotel Demand System. ODI and KFO are parallel frameworks with distinct scopes.
Email can serve Demand Introduction when it reaches a qualified audience beyond the hotel’s existing guest file, or Guest Relationship Development when it supports an established relationship. The audience, permission, and purpose determine its role. AGR’s complete guide to email marketing for hotels explains those applications and their measurement.
The Diagnostic Question
For independent hotel leadership, direct share remains an important measure. Add the question it cannot answer: where did the relationship with those guests originate, and who controls the next contact?
If introduction is largely intermediated and the hotel does not secure its own permissioned relationship, better conversion leaves part of the dependency intact. Paid media is not automatically evidence of that problem. A paid introduction that produces valid permission in an environment the hotel governs can contribute to owned demand origin.
The 63.4% benchmark makes the distribution question urgent. It does not determine the diagnosis for every hotel. Leadership needs to examine origin, permission, conversion, and future relationship control together.
Where the constraint is demand origin, the fix has to address demand origin. Better downstream execution cannot stand in for that work.
If your property needs a clearer view of where demand control is weakest before refining strategy, AGR’s fixed-fee Demand Analytics surfaces OTA dependency, margin leakage, and the highest-priority structural issues in your current distribution model.
