The Phocuswright Conference 2025 arrived at a defining inflection point for travel technology capital allocation. The dominant signal from San Diego was not merely that AI is coming to travel — it is already reshaping distribution economics, startup formation patterns, and the competitive moats that have protected incumbents for two decades. Barry Diller opened the main stage by declaring that ChatGPT's 900 million users represent the first genuine force capable of breaking Google's search monopoly, a system that Expedia and its nearest OTA competitor currently fund to the tune of more than $10 billion per year. For investors, that single observation reframes the entire supply-side economics of the travel market.
The more granular and actionable insight from conference sessions, however, is that the most immediate AI investment opportunities are not at the front of the funnel. The sessions that generated the sharpest investor attention — from the Innovation Launch startup competition to the Wall Street analyst panel — consistently located the high-conviction near-term value in post-booking infrastructure: disruption management, refund and exchange automation, corporate event servicing, and back-office workflow orchestration. These are contractual B2B revenue streams with compounding data moats, sub-three-month enterprise sales cycles, and quantifiable ROI that does not depend on consumer behavioral change. This is where capital is easiest to underwrite.
Simultaneously, a structural shift in distribution power is underway that will take years to resolve and will create both casualties and winners among public market players. Google's announcement of end-to-end agentic hotel booking inside AI Mode — partnering with Booking.com, Expedia, IHG, Marriott, and Wyndham — confirms that the OTA intermediary layer faces a genuine, if slow-moving, existential test. The Wall Street analyst panel (Evercore, Mizuho, Barclays) was broadly constructive on OTAs' near-term resilience, but flagged material risk if AI drives traffic back through Google channels where OTAs pay roughly eighty cents per dollar of acquired revenue. The structural question is not whether OTAs survive, but whether their margin structure survives — and what infrastructure players and direct-booking startups capture in the redistribution.
For founders, the most consequential theme was arguably the collapse of the build-versus-validate sequence. Phocuswright's own startup survey found 81% of travel startups are already using generative AI, with 54% reporting major positive impact and the top-cited benefit being faster prototyping. A third of startups project more than a quarter of their codebase will be AI-written within 12 to 24 months. The cost to reach a demo-worthy product has fallen so dramatically that the vintage of a travel tech startup founded in 2025 or 2026 should be expected to reach feature parity with 2022-vintage companies in a fraction of the time. Investors pricing rounds on traditional development timelines may be systematically mispricing this cohort.
Key Findings
Finding 1: Post-Booking is the Highest-Conviction AI Investment Category Right Now
Across multiple sessions, a clear consensus emerged that post-booking operations — not trip planning or booking interfaces — represent the most immediately monetizable AI opportunity in travel. The evidence is specific and multi-sourced.
Acai Travel (People's Choice Award winner, Innovation Launch 2025) reported $2 million ARR, 50% monthly growth in its two most recent months, sales cycles under three months, and customer acquisition costs below $10,000. CEO Riccardo Vittoria quantified the market at $200 billion in annual travel support and disruption costs, with a $20 billion revenue opportunity for AI providers. His three-agent architecture — supervisor, front-office, and back-office agents — delivered 60% reductions in handling time and 20–30 basis-point CSAT improvements at deploying customers.
Wenrix (runner-up, Travel Innovator of the Year) presented DeepFlow, an AI execution layer for post-booking servicing trained on over 50 billion real-world data points, achieving 93% automation for one travel company versus an industry average materially below that. CEO Amir Balaish quantified the execution gap precisely: 40% of simple ticket servicing is automated, but the remaining 60% of complex edge cases consumes 82% of agent time.
Robert Buckman of Amadeus — which holds a minority stake in Acai — explicitly named travel disruption management as the next major frontier "likely to be meaningfully solved within a few years." NDC complexity has increased call center handling times by 20–30%, while experienced human agents are retiring faster than replacements can be trained. The demographic and technical forces driving this problem are accelerating, not stabilizing.
Investment implication: Post-booking AI infrastructure has enterprise B2B revenue, contractual relationships, compounding proprietary data moats, and measurable ROI. It is the category with the clearest path from seed to Series A to Series B without requiring consumer behavioral shifts. Multiples should be evaluated against SaaS comparables, not volatile consumer internet benchmarks.
Finding 2: Distribution Power is Shifting — But the Endgame is Unclear, Not Inevitable
Google announced on the opening Monday of the conference that it is enabling end-to-end hotel booking inside AI Mode, with Booking.com, Expedia, IHG, Marriott, Wyndham, and Choice Hotels as launch partners. James Byers, Group Product Manager for Travel at Google Search, described a product capable of running complex multi-variable queries — flights, hotels, weather, transit — simultaneously, with 75 million daily active users on AI Mode.
The Wall Street analyst panel took a differentiated view on the risks. Mark Mahaney (Evercore) argued generative AI will act as a new lead-generation platform for OTAs rather than eliminating them, and noted that ChatGPT is currently delivering high-quality leads that OTAs are not yet being charged for. Lloyd Walmsley (Mizuho) was more cautious, noting that even if AI does not directly disintermediate OTAs, if it routes traffic back through Google, OTAs face margin deterioration at 80 cents per dollar of acquired traffic. Emma Taylor (Barclays) shifted focus to back-end AI ROI — call center automation, software velocity, fintech — as the area of most measurable near-term return.
DirectBooker, featured in the "Reinventing Hotel Booking" panel with former Google Travel VP Richard Holden and TripAdvisor founder Stephen Kaufer, is executing a specific disruption play: a low-cost API pipe connecting hotel chains directly to AI discovery platforms (ChatGPT, Gemini), bypassing OTAs. The company has signed contracts with six of the top ten hotel chains globally. Pricing is below OTA commission rates and below credit card processing fees, with hotels retaining merchant-of-record status, guest data, and loyalty point crediting. Kaufer described the current AI moment as the fourth major disruption in travel's digital history.
Cloudbeds CEO Adam Harris reported that LLM referral traffic to its hotel properties is "growing exponentially," with the majority going direct rather than through OTA intermediaries. He predicted that within five years, consumers will interact with brands directly rather than through websites.
Booking Holdings CSO Rob Ransom acknowledged the company has launched an app inside ChatGPT's store via MCP-enabled queries and was announced as a Google AI Mode booking partner, but characterized it as an "MVP learning environment" with relatively low transaction volume. He confirmed the Booking.com app now represents more than half of total bookings.
Investment implication: There is no single clear winner in the distribution shift. The highest-risk public market positions are pure metasearch players (the analyst panel discussed Booking's Kayak write-down as a possible early casualty) and OTAs heavily dependent on Google traffic. The highest-risk short-selling thesis is against companies whose moats rely entirely on search engine traffic volume. Watch DirectBooker as an early indicator of whether hotel chains will break ranks with OTAs at scale.
Finding 3: Financial Institutions Are Becoming Serious Travel Competitors
The "Everyone's Trying to Sell Travel" panel surfaced a structural competitive threat to OTAs that is already past the investment thesis stage and into execution. Chase Travel is pursuing vertical integration at scale — acquiring CX Loyalty, Frost (luxury travel), The Infatuation (dining), and Fig (payments/offers) — and targeting $12–15 billion in annual gross bookings while growing at approximately 20% CAGR. CEO Jason Wynn described it as a "vertically integrated travel operating system" leveraging Chase's existing financial trust and customer data.
Hopper Technology Solutions (HTS) — now over 90% of Hopper's revenue — has decoupled its fintech ancillary products (price freeze, cancel-for-any-reason, disruption assistance) and makes them available via API to any travel seller globally. CEO Dakota Smith cited Nubank in Brazil as a case study: instead of loyalty points, Nubank offers travelers 12-month installment plans at 0% APR in a market where average consumer APR is 35%. Hopper's ACS Assist agentic AI is processing approximately 3 million live customer service conversations per year. iSeatz CEO Kenneth Purcell estimated the US financial institution travel marketplace at $45–50 billion in total gross bookings, a fraction of overall travel but growing rapidly.
Phocuswright research presented during this session showed Expedia's B2B segment growing more than 25% year-over-year and representing approximately 30% of Expedia's over $100 billion in gross bookings. The implication is that even as B2C OTA economics face pressure, OTA B2B supply businesses are positioned as infrastructure beneficiaries of the financial institution trend.
Investment implication: The financial institution travel segment is past early stage. Founders should evaluate whether their solution addresses fintech-first travel buyers — this is a fast-growing distribution channel with high transaction values. For investors, the Hopper B2B model is the clearest template for how a consumer-originated travel tech brand can pivot to durable B2B revenue. Expedia's B2B segment growth makes it a more defensible position than headline OTA narratives suggest.
Finding 4: Agentic Commerce Infrastructure is an Underappreciated Early-Stage Opportunity
The "AI Agents, Real Transactions" panel with Visa, Stripe, and ARC (Airlines Reporting Corporation) covered the emerging protocols that will govern AI-initiated travel transactions. These infrastructure layers are live but nascent, and represent an early-stage investment category.
Stripe's Agent Commerce Protocol (ACP), co-developed with OpenAI and announced September 2025, went live with Etsy for simple products. Clara Liang of Stripe described shared payment tokens with spend limits, time limits, and revocation capabilities, layered on fraud signals from approximately 92% of all card transactions globally. Visa's Trusted Agent Protocol (TAP), announced October 2025, adds authentication (verifying agent legitimacy), tokenization, and buyer intent metadata capture. Both are designed as open standards.
ARC's Jennifer Watkins confirmed that agentic commerce was absent from airline distribution conference agendas in spring 2025 but dominated every panel by late 2025. Airlines' immediate concerns include disintermediation from the customer relationship, fraud in new payment flows, and distribution channel readiness. NDC currently processes approximately 20% of ARC transactions; full "order" standard support by airlines is targeted for 2026.
The panel identified B2B flows as the most immediate use case: OTA-to-hotel payouts, bulk flight bookings, and reconciliation are natural starting points before the more complex consumer-facing agentic booking use cases scale. Stablecoins were described as an enabler for real-time cross-border settlement, with real-world examples already in production.
Investment implication: The companies building at the payments and identity infrastructure layer of agentic travel commerce — not the application layer — will capture structural, recurring toll revenue. PassiveBolt (Global Startup Pitch Scaleup Winner), with its Keyshare platform and Gartner's forecast of 500 million standards-based digital identity credentials by end of 2026, is representative of infrastructure plays that will become prerequisite to agentic check-in and transactional flows. For founders, the practical advice from the panel was explicit: treat payments as a first-class consideration from day one, not an afterthought.
Finding 5: The "AI Alignment" Problem Creates a New Content Infrastructure Category
Bonafide.ai (PhocusWire Hot 25 Travel Startups for 2026) identified a specific and measurable commercial problem: approximately one in four LLM responses does not align with a travel brand's factual information, representing a 25% inaccuracy rate with direct booking revenue consequences. CEO Layton Han cited a concrete example — Google Gemini was unaware of Mandarin Oriental Hong Kong's premium limousine service — and reported that Bonafide's platform improves LLM alignment by up to 40%, directly correlated to a 40% improvement in booking opportunity.
This problem is structural and will worsen before it improves. LLMs are trained on periodic data snapshots; travel product information changes continuously. Hotels, airlines, and experiences operators all face the same challenge: their information in AI discovery platforms is outdated, incomplete, or simply absent. The traveler query behavior compounding this — asking granular, specific, personalized questions rather than broad keyword searches — means the accuracy gap has disproportionate commercial consequences.
Casago CEO Steve Schwab mentioned investing in "deeper, long-tail property data to improve LLM discoverability" as a strategic priority. Marriott launched a natural-language search interface for Homes & Villas partly to ensure its 200,000-property inventory is surfaced accurately in AI search contexts. Cloudbeds CEO Adam Harris described LLM referral traffic as already "growing exponentially," confirming that AI platforms are actively driving booking intent that brands are either capturing or losing based entirely on data quality.
Investment implication: This is an early-stage B2B SaaS category with clear enterprise customers (hotels, airlines, experiences operators, OTAs) and quantifiable revenue impact. The core moat is a proprietary testing, gap-analysis, and content reinforcement methodology applied at scale. Competitors are likely to emerge from SEO and content marketing agency backgrounds, but travel domain specificity is a real differentiation. Investors should evaluate whether Bonafide's 40% alignment improvement claim holds across diverse property types and languages before extrapolating unit economics.
Finding 6: Hotel Operations is an Underfunded Category with Near-Term AI ROI
Levee (Global Startup Pitch Seedup Winner) CEO Al Lagunas framed hotel operations as a category where the foundational data infrastructure for AI, automation, and robotics does not yet exist — and built his wedge product accordingly: a multimodal AI room inspection agent delivered via smartphone. The pitch was specific on commercial framing: labor is the number-one line item on hotel balance sheets, supply of workers is not growing, and workforce augmentation via technology is the only viable path.
The panel "How to Compete for the Modern Hotel Guest" reinforced this with granular data: citizenM has 22 FTE handling reservations across 36 global hotels, processing 55,000 invoice requests annually with full automation and resolving 96% of Booking.com guest queries without human intervention. CEO Lennert de Jong attributed this to treating operational data architecture as a first-principles problem. Cloudbeds CEO Adam Harris described the transition from "systems of record" to "systems of action" — hotel data orchestrated in real time to power agentic AI conversations — as the foundational shift the industry needs to complete.
American Airlines VP Neil Geurin noted that AA has records on 40 million customers but faces a latency problem: parsing that data meaningfully within the sub-second consumer tolerance window is technically unsolved. His near-term practical redirect was to focus personalization on the post-booking, pre-travel window where milliseconds don't constrain relevance. Accor's Group Deputy CEO Jean-Jacques Morin identified data quality from siloed PMS/CRS systems as the core barrier to any downstream AI application.
Investment implication: Hotel tech stack modernization is a multi-year, multi-billion-dollar infrastructure replacement cycle. Amadeus's Robert Buckman described hospitality IT as "the space most in need of a full tech reboot." The enterprise sales cycle is long, but the competitive moat for vendors who embed early is high. Levee's wedge via frontline worker tools — an underserved segment that larger vendors historically ignored — is a credible bottom-up GTM strategy that has worked in other labor-intensive verticals.
Finding 7: The Startup Formation Environment is Paradoxically Attractive Despite Funding Drought
Phocuswright's global startup survey (presented by Mike Coletta) was clear: venture funding has slumped to decade lows for travel tech, yet AI tooling has dramatically compressed the cost of building. The result is a paradox — a market with fewer funded competitors but dramatically lower barriers to reaching product-market fit. Eighty-one percent of travel startups are already using generative AI; 54% report major positive impact; the primary benefit is faster prototyping.
Coletta articulated a speculative but serious scenario: an AI-only OTA launches with no human agents, promises 10-second refunds, zero fees, automatic disruption management, 1:1 loyalty optimization, and charges only 5% commission. A majority of Phocuswright conference attendees considered this scenario at least possible within 10 years. His closing BCG-derived estimate: AI-only rivals are 5–15 years away — a timeframe that has historically been sufficient for major industry disruptions in travel.
Agoda CEO Omri Morgenshtern noted that 95% of developers across Southeast Asia and India are already using AI in back-office operations. Sabre CEO Kurt Ekert reported 15% developer productivity improvement from Google's Gemini and Vertex AI platforms. Skyscanner CEO Bryan Batista cited 20%+ productivity gains from AI tools.
Investment implication: This is a vintage year for travel tech seed investment. The denominator cost of reaching a functional product has compressed; the numerator (competitive alternatives) has contracted with funding. Early-stage investors should evaluate teams on speed of AI-tool utilization, depth of domain data access, and clarity of the B2B revenue path — not on traditional development runway assumptions. The one risk: the "AI bubble" warning from Barry Diller, who drew specific parallels to the 2001 dot-com crash and noted data center and infrastructure spending runs well ahead of revenue.
Finding 8: Asia and Emerging Markets are the Overlooked Long-Term Growth Opportunity
Phocuswright's opening data confirmed Asia-Pacific at 27% of global bookings in 2024, with India projected to reach $61 billion in gross bookings by 2028 — roughly 1.5 times its 2024 level. Booking Holdings CSO Rob Ransom said Agoda is Booking's Asia-focused brand and called Asia "the highest long-term travel growth region," singling out Southeast Asia's entrepreneurial ecosystem and predicting Singapore will produce the next travel unicorn.
Yanolja, a South Korean platform serving 1.3 million suppliers and over 20 million end consumers in Korea alone, presented its three-sided marketplace model (suppliers, channel partners, consumers) and its expansion into Africa, Latin America, and North America. CEO Jeff Kim outlined a long-term agentic AI vision in which separate agents represent the traveler and the property, negotiating personalized experiences autonomously — a model he described as "neurons connecting brain to organs." Yanolja is backed by GIC (Singapore's sovereign wealth fund) and Booking Holdings.
Agoda CEO Morgenshtern explicitly argued that Asian markets require entirely different solutions for Japan, China, Indonesia, Thailand, and India — that market fragmentation is a feature, not a bug, for venture-backed specialization. He also framed OTAs as the ideal distribution layer for financial products tied to travel spend — loans, FX hedging, savings — and predicted OTA-bank convergence will accelerate globally.
Investment implication: The most underpenetrated frontier for travel tech venture capital is Southeast Asia and South Asia. Agoda's Morgenshtern effectively called out that cross-border payment friction, FX anxiety, and access to financing are more impactful unmet needs than trip planning AI in these markets. Founders targeting India, Indonesia, or the 100 million new middle-class travelers entering the Asian market should expect favorable competitive dynamics compared to saturated North American and European segments.
Strategic Implications
For Venture Capital and Private Equity Investors
The highest-conviction near-term category is post-booking AI infrastructure. The convergence of NDC complexity, retiring human agent talent, and proven enterprise deployment metrics (Acai's sub-$10K CAC, Wenrix's 93% automation benchmark) makes this a category where valuation discipline and domain expertise matter more than thematic boldness. Diligence should focus on: proprietary training data depth, airline-specific policy coverage breadth, and whether products are front-office-only or include back-office execution capability (the latter is materially harder to replicate).
Distribution disruption will produce M&A, not just organic winners. The OTA industry spent $15 billion annually on Google search. Even a 20% shift of that spend to LLM platforms restructures unit economics across the sector. The analyst panel identified B2B travel tech, corporate travel, fintech, ancillary services, and supplier-facing software as the categories with most investor appetite — all characterized by contractual, recurring revenue less exposed to consumer traffic volatility. For M&A strategists: Booking's consolidation of its B2B distribution arms (Booking.com, Agoda, Priceline) into a unified organization signals strategic focus on B2B infrastructure as a defensive moat.
Valuations on AI infrastructure are likely inflated; valuations on AI-enabled travel B2B SaaS are likely underpriced. Barry Diller's bubble warning is specific: data center and infrastructure spending is ahead of revenue. But the operational B2B SaaS plays — post-booking automation, hotel ops, corporate event management, content alignment — have revenue today, measured ROI, and compounding data advantages that traditional infrastructure plays lack.
The agentic commerce payments infrastructure is a platform-layer opportunity. Stripe's ACP and Visa's TAP are open standards, not proprietary moats. The moat belongs to whoever builds the highest-fidelity travel-specific authentication, fraud detection, and buyer intent signal layer on top of these standards. This is a Series B/C opportunity, not early stage — it requires transaction volume to train meaningfully.
For Travel Tech Founders
Build for the execution gap, not the conversation layer. The most defensible products shown at Phocuswright 2025 were the ones that automated action, not just answered questions. Chatbots are ubiquitous. Systems that actually execute a refund, manage a disruption, or change a flight reservation without human intervention are rare and enterprise-valued.
Your data strategy is your company. Agoda runs "thousands of experiments daily." Levee captures inspection data that no other system has. Bonafide builds a proprietary knowledge base of granular brand information that LLMs don't have. The companies that will be acquired or become category leaders are those that built data collection into the core product loop from day one — not as a downstream analytics afterthought.
The MCP (Model Context Protocol) layer is a practical go-to-market decision right now. Amadeus, BoomPop, Yanolja, and citizenM all referenced MCP as an active architectural choice for connecting directly to AI agents. BoomPop CEO Healey Cypher specifically cited MCP-based architecture as a technical moat enabling "real-time cascading updates across reservations, websites, and personal calendars from a single command." Founders building in hospitality, corporate travel, or distribution should treat MCP integration as a near-term product priority, not a roadmap item.
Asia-first is a viable venture strategy. The market fragmentation that makes Asia hard for incumbents to address at scale is exactly the condition that rewards specialized, locally-built startups. Agoda's Morgenshtern outlined specific unmet needs — cross-border payments, FX anxiety, access to financing — that are more acute and less served than the AI trip planning use case receiving most Western founder attention.
Action Items (Prioritized)
Tier 1 — Act Now
Map the post-booking AI competitive landscape. Build a watch list of Acai, Wenrix/DeepFlow, and comparable companies (Dentrixx, Oversee, DNG as named by Acai). Evaluate the differentiation between front-office only vs. full back-office execution. The category will consolidate; early thesis development creates diligence advantage.
Evaluate MCP-native travel startups. The Model Context Protocol is emerging as a foundational architectural standard endorsed by Amadeus (aligned with Anthropic), Booking.com (ChatGPT app), BoomPop, and Yanolja. Founders building MCP-first deserve premium evaluation; investors should screen existing portfolios for MCP readiness.
Engage Bonafide.ai and the LLM alignment category. The 25% LLM inaccuracy rate for travel brands is a quantified commercial problem at the enterprise level. Validate the 40% alignment improvement claim across multiple customer references. If it holds, this is a category with fast enterprise sales cycles and limited direct competition today.
Monitor the Google AI Mode booking partnership cohort. Google's launch partners (Booking.com, Expedia, IHG, Marriott, Wyndham, Choice Hotels) are the canary for agentic transaction economics. Watch for public disclosures on conversion rates, commission structures, and merchant-of-record arrangements. This data will be the leading indicator of OTA margin compression or resilience.
Tier 2 — Develop Over Next 90 Days
Build a Southeast Asia/South Asia deal pipeline. India's $61 billion projected gross bookings by 2028, combined with Agoda's explicit endorsement of Singapore as the next unicorn origin market, signals a window before Western travel tech capital fully prices in Asian market growth. Focus on cross-border payments, alternative financing, and distribution infrastructure for fragmented markets.
Develop a hotel operations thesis. Levee, Cloudbeds, Yanolja, and Amadeus's hospitality IT reboot initiative all point to a multi-year replacement cycle in hotel tech. Identify the specific stack layer (PMS, operations, guest communication, data orchestration) with the highest near-term replacement urgency and evaluate the competitive map in that segment.
Review corporate event management as a category. BoomPop (Travel Innovator of the Year, $100M annual event spend on platform, 450 brands, NPS of 85) is operating in a segment — 7,500 corporate off-sites planned daily — that existing TMCs have systematically underserved. The MCP-native architecture, subscription marketplace model, and bilateral value proposition (buyers and hotel sales) differentiate it from both legacy event management and corporate travel platforms.
Request a dedicated briefing on Sabre's Concierge IQ. The announcement of a conversational commerce layer launching in full production with Virgin Australia — covering research, search, booking, and post-trip service within a single conversational interface deployable via WhatsApp or agency channels — is a GDS incumbent's most serious agentic product announcement to date. Evaluate whether this is a genuine moat or a features announcement.
Tier 3 — Monitor Closely
Track DirectBooker's hotel chain uptake. Six of the top ten global hotel chains have signed contracts. If this becomes eight or ten within 12 months, it is the strongest signal yet that hotel direct booking to AI platforms is moving from experiment to strategy at the enterprise level, with direct implications for OTA market share models.
Watch the U.S. inbound travel policy environment. Geoff Freeman (U.S. Travel Association) confirmed the U.S. will be the only country globally to see net decline in international arrivals in 2025, with a 24% collapse from Canada alone. The proposed $250 visa integrity fee remains a live threat. Travel tech companies with significant U.S.-inbound revenue exposure should model the downside case explicitly.
Sessions to Watch
| Session | Why It Matters for Investors and Founders |
|---|---|
| Acai Travel — WINNER People's Choice Award, Phocuswright Innovation Launch 2025 (Riccardo Vittoria) | The clearest investment-grade signal of the conference: $2M ARR, 50% monthly growth, sub-$10K CAC, enterprise B2B deployment. The benchmark for evaluating all post-booking AI pitches. |
| Street Talk: Finance, Investment, M&A, IPOs in Travel (Mahaney/Evercore, Walmsley/Mizuho, Taylor/Barclays, Fuller/BTIG) | The most direct investor-facing session at the conference. Three Wall Street analysts debate OTA margin risk, metasearch decline, Airbnb's hotel bet, and the IPO pipeline. Essential context for public market positioning. |
| Reinventing Hotel Booking with AI — Holden, Kaufer and Vakil (moderated by Hemmeter/Thayer Investment Partners) | Former Google Travel VP, TripAdvisor founder, and DirectBooker CEO debate the fourth major disruption in travel's digital history and the specific business model of the hotel direct booking play. |
| AI Agents, Real Transactions — The New Travel Economy (Visa, Stripe, ARC) | The infrastructure session VCs typically skip but shouldn't. ACP and TAP are the plumbing of agentic commerce. Understanding the protocols being standardized now determines which application-layer bets will have distribution. |
| Everyone's Trying to Sell Travel — The New Travel Seller Arena (Chase Travel, Hopper, iSeatz) | The financial institution travel market is a $45–50B US segment growing at 20%+ CAGR. The Hopper B2B pivot model and Nubank 0% APR installment case study are the clearest templates for how non-OTA companies are capturing travel spending. |
| What Can Travel Learn from AI-Native Startups? (Mike Coletta, Phocuswright Research) | The most analytically useful startup market data presented at the conference, including the lean startup inversion thesis and the AI-only OTA disruption scenario. Essential framing for any founder pitching or investor evaluating early-stage travel tech. |
| "Life? Travel? Count on it!" — Barry Diller returns after 20 years (Lorraine Sileo, Phocuswright Research) | Diller's near-term AI bubble warning and his specific $10B+ OTA Google spend framing are the macro context every travel tech investor needs. His "inversion" strategy for IAC's media brands is also a repeatable thesis for content-driven travel companies. |
| Competing in Travel's Next Era with Sabre's CEO Kurt Ekert (Linda Fox, PhocusWire) | Sabre's cloud migration (99% compute in cloud), Concierge IQ product launch with Virgin Australia, and "petabyte-scale intelligent caching" as a competitive moat claim all require evaluation. The GDS category is either a value trap or a turnaround — this session provides the primary data. |
*Brief prepared from The Phocuswright Conference 2025, San Diego. All data points and attributed statements are sourced directly from session transcripts. This brief is intended for sophisticated investors and operators conducting independent due diligence.*