The Agentic Commerce Shift: How Travel & Hospitality Will Win the AI Era
Travel is becoming a machine-to-machine market wrapped around human moments. The winners will be the companies that make their inventory, policies, service quality, and brand trust readable and executable by agents.
The Travel Commerce disruption
A perfect storm is reshaping travel shopping at unprecedented speed, and the rapid adoption of agentic channels by customers is creating significant uncertainty and doubt across the travel value chain. The first phase of the transition towards Agentic Commerce is already well underway for many travel brands as “AI is beginning to rewire travel discovery” (Skift Travel Outlook 2026). In 2025, 30% of travelers used AI to plan trips, and many may never visit travel websites again as Agentic Commerce addresses today's travel booking pain points of fragmentation (the average customer visits 277 websites before booking (Expedia Path to Purchase Research)) and friction (70% of travelers describe booking travel as stressful (Yougov Travel Stress Report 2025)).
Agentic Commerce shifts travel distribution from search-and-select to intent-and-execute. Instead of juggling dozens of tabs comparing flights, hotels, and policies, a traveler simply describes its intent and receives outcomes rather than options. AI agents execute bookings, orchestrate multi-vendor transactions, manage exceptions, and maintain continuity across the
journey lifecycle, making the kind of proactive, anticipatory service once reserved for dedicated concierge clients accessible to every traveler.
Travel and Hospitality companies have strong defenses in the Agentic disruption. The industry is backed by unique physical assets and increasingly focuses on selling differentiated experiences. However, the reality is that AI agents now prioritize data quality and alignment with user intent above what previously mattered to human users.
The Generative Engine Optimization Imperative
Travel brands have one short-term imperative: meet customers where they are by ensuring visibility on AI channels through Generative Engine Optimization (GEO). GEO involves improving first-party content structure, fine-tuning AI access rules to digital channels, influencing third-party content, and establishing data-driven measurement and optimization practices (likely soon extended to advertising) to maximize visibility on AI-native digital gatekeepers such as public AI chatbots. The urgency is real as organic search click-through rates continue to decline.
Evolving to a Machine-to-Machine Market wrapped around Human Moments
Visibility is only half the story. Travel commerce is evolving into a machine-to-machine market, with transaction completion the natural next step for customers already engaged on agentic channels. Instead of deep-linking to booking platforms, customers will instruct their primary life agent (whether on a consumer platform, first-party consumer or corporate travel tool, or enterprise assistant) to return prescriptive recommendations and execute.
All current agentic adoption trends point to a shift that could be faster and bigger than the internet itself (The automation curve in agentic commerce, McKinsey). The travel industry ranks high in both opportunity and complexity: Improving travel shopping is long overdue, but transactional complexity and offer volatility require robust solutions. Unlike early-funnel travel inspiration optimized through GEO, agentic travel shopping and booking requires broad real-time access to core industry systems such as offer engines and reservation systems.
Upgrading the Travel Operating System for Agentic Commerce
As part of this emerging transition, many in the industry noticed the potential of Model Context Protocol (MCP) to enable Travel Agentic Commerce. Yet more than one year after MCP’s release (an eternity in AI time) production results remain scarce because model intelligence and agentic messaging alone cannot deliver agentic commerce capabilities.
Just as User Experience enables seamless human interactions, agents require an "Agent Experience", called context engineering, to deliver accurate, trustworthy results. The gap in travel agentic commerce isn't agents themselves, but the orchestration layer around them: software that unifies data, delivers contextual knowledge, coordinates workflows, manages tools, enforces policies and permissions, and organizes customer feedback loops.
Consider what it takes to answer an open-ended query like “find a trip under $2,000 to a national park” or “show me the cheapest options to see my favorite singer during spring break” across multiple arrangements such as airlines and hotels. This requires aligning today’s fragmented systems, intermediaries, and providers that currently operate narrow request-based systems.
Robust agentic identity and permission management capabilities are needed to differentiate anonymous users from authenticated ones, and to act and share data on users’ behalf. Agentic payment systems that balance convenience, trust, and compliance are needed to enable transactability.
Builders should implement use cases iteratively to leverage innovations like MCP while approaching necessary rework step-by-step and clearly communicating to users what's possible and what's not at each stage.
Accelerating Modern Retailing in an Agentic World
For over a decade, the industry has been gradually shifting away from commoditized offers toward retail-like capabilities (bundling, dynamic pricing, rich content, and channel flexibility) that enable revenue growth, differentiation, and customer loyalty. Agentic Commerce can amplify retailing capabilities, fueling richer context across broader portfolios for more relevant offers and greater customer satisfaction. However, suppliers must ensure that trust built through Modern Retailing (such as rich content or loyalty recognition) carries into the agentic world so key offer differentiators remain constantly perceptible by end customers.
Designing the Travel Data Backbone for Scale
Before the Agentic disruption, travel technology infrastructure was already strained. Standard travel industry APIs (like NDC AirShopping for airline shopping requests or PAOREQ for airline product availability requests) struggle with rising costs and scalability as growth accelerates, look-to-book ratios climb, and revenue models grow more complex . Agentic contexts amplify these challenges exponentially.
Industry leaders must adopt advanced patterns to manage this, including:
- Expand the use of precomputed data domains by building real-time event-driven systems that proactively notify agent contexts of offer changes and keep knowledge layers current across providers (like airlines and hotels).
- Unify structured and unstructured data by using structured information (prices, offer records) as the foundation, enriched with unstructured context.
- Reduce costs through predictive algorithms that lower computational complexity, or "shopping clones" that place consumers and suppliers in the same environment to minimize data movement and encourage cost optimization.
- Continuously strengthen security, data, and IP protection as agents independently handle increasingly complex and sensitive operations on behalf of users
An Agenda to Act
To succeed in this Travel Commerce transformation, the industry must act now. Each disruption wave moves faster than the last. The agentic commerce clock started in 2025.
Travel players can immediately expand their Agentic Commerce capabilities:
Generative Engine Optimization is the critical first step for reaching customers where they are. Companies can review how agents consume their public-facing data and establish processes for continuous improvement to maximize their "share of model."
High-impact initial use cases like business travel assistants or first-party branded discovery chatbots can launch quickly, providing a foundation to master technology, develop skills, and learn from customer behavior and data.
Core building blocks such as Customer 360, personalization, and omni-channel unification provide critical enablers for success in the Agentic Commerce era. IDC forecasts that by 2030, 50% of AI budgets in hospitality and travel will be allocated to personalization efforts.
Ultimately, Agentic Commerce favors companies with strong organizational, technical, and product foundations that enable rapid development of new paradigms, possess architectures that minimize data unification efforts for agents, and refine products to earn customer trust.
About the authors
Guillaume Dupont joined AWS in 2025 as a Senior Travel & Hospitality Architect, partnering with the world’s largest airlines on their digital transformation journey. His 15-year track record spans AI innovation at travel industry leaders Amadeus, PROS, and Airbus. Guillaume holds a master’s degree in Operations Research from the National Civil Aviation University of Toulouse, France, where he continues to share his knowledge as a guest lecturer.
Keith Chaney leads the management and execution of Amazon's most complex, high-impact strategic partnerships, driving AWS adoption across global industries. His work spans Amazon's entire portfolio of business units and subsidiaries, orchestrating cross-organizational initiatives. With over 15 years of experience leading technology and organizational transformations at McKinsey & Company, Google, and Booz Allen Hamilton, Keith brings deep expertise in strategic innovation and enterprise-scale change management. Keith holds an MBA from Carnegie Mellon University and a BSE in Industrial Engineering from Kettering University. He resides in Washington, DC with his wife and two daughters.
Matias Undurraga is an Enterprise Technologist and technology leader with more than 20 years of experience helping organizations use AI, data, and modern cloud platforms to drive innovation. Based in Switzerland, he advises leaders on topics such as generative AI, data strategy, modernization, and digital product development. His work focuses on making complex technology shifts practical, actionable, and tied to real business value.
Christian Wolff is a Solution Architect in the UK Cross-Industry Travel and Hospitality team, specializing in Generative AI adoption and enterprise modernization. He is a lead contributor to the UK Future of Search initiative, which provides thought leadership and guidance to customers on the changing future of optimization and discovery. He has published blogposts and sample code to demonstrate the art of the possible in Generative Engine Optimization.
Moni Panchavadi is the Global Partner Solutions Architect for Travel & Hospitality at Amazon Web Services, based in Atlanta, GA. With expertise spanning cloud and the travel and hospitality industry, Moni drives partner solution strategy and collaborates across AWS's global partner network to deliver meaningful industry solutions for AWS customers.
Massimo Morin joined Amazon in 2017, and he is now Global Head, Travel, AWS for Travel and Hospitality. He brings 25 years of experience in the travel and hospitality industry and has deep knowledge in the travel industry of retailing and digital transformation. Before joining Amazon, he worked for ITA Software as Technical Director, and at Google as Head of BD Travel, EMEA. His core expertise is digital transformation in Travel, airline pricing, distribution, revenue management, customer care, and e-commerce. He graduated in Software Engineering from the University of Venice and acquired a master’s degree in the science of Transportation / Airline Business and Management from MIT.
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article