No PriorsTravel Through the Lens of AI with with Booking.com CEO Glenn Fogel
CHAPTERS
- 0:00 – 1:18
No moats in travel: constant innovation and deep industry complexity
Glenn opens with a blunt view that there are no permanent moats—advantages erode quickly under innovation. He argues that winning in travel requires continuous product improvement and a realistic appreciation of operational and regulatory complexity before committing capital.
- •Competitive advantages are temporary; innovation can erase them quickly
- •Long-term success comes from continually building new services and better workflows
- •Travel is more complex than it appears from the outside (partners, demand, operations)
- •Would-be disruptors should understand the industry deeply before investing
- 1:18 – 2:08
From Wharton to mainframes: early tech and finance foundations
Glenn recounts starting in Morgan Stanley’s back-office MIS work, operating mainframes and learning software fundamentals. Despite a finance degree, this technical start shaped how he later evaluated technology and businesses.
- •First roles involved hands-on data center/mainframe operations
- •Transitioned from operator to developer, gaining practical tech context
- •Realized early that initial career path wasn’t the right long-term fit
- •Early exposure to systems work later informed his leadership perspective
- 2:08 – 3:31
Law school, Wall Street, and getting fired: a formative leadership lesson
After pivoting through Harvard Law into investment banking, Glenn was laid off during a bank acquisition. He describes the experience as a lasting lesson in empathy and how to handle difficult people decisions.
- •Used law school as a route into Wall Street banking
- •Layoff experience became a personal lesson about how job loss feels
- •Shaped his views on leadership and handling terminations responsibly
- •A personal low point prompted reevaluation of life direction
- 3:31 – 5:03
Writing a novel, meeting his future spouse, and returning to markets
Unemployed, Glenn attempted to write and publish a novel and met a former Random House editor on a blind date—who became his wife. He then returned to finance via a trading role, though it ultimately wasn’t a fit.
- •Wrote a novel; publishing didn’t happen but life changed via the relationship
- •Motivation to find work led to a trading job opportunity
- •Served as head trader for Barton Biggs but didn’t enjoy the work
- •Period reinforces themes of experimentation and finding the right path
- 5:03 – 7:27
Joining Priceline at the peak—and surviving the crash
Glenn joined Priceline in early 2000 right around the Nasdaq peak and watched the company’s market cap collapse from tens of billions to a few hundred million. He stayed through near-delisting, reverse splits, and the long rebuild into a $100B+ business.
- •Delayed start to capture bonus—then joined just as the market peaked
- •Priceline’s market cap fell dramatically; stock approached delisting levels
- •Reverse split and persistence through the downturn
- •Long-term compounding: shares moved from ~6 post-split to near ~6000
- 7:27 – 10:02
AI wave vs. early internet: hype cycles, parallels, and survivorship
Sarah and Glenn compare today’s AI exuberance to the late-90s internet boom. Glenn expects disappointment and failures alongside real innovation, emphasizing that speculative cycles are recurring features of economic history.
- •AI mirrors the internet era’s optimism and subsequent backlash
- •Success/failure ratios are hard to predict; booms repeat across history
- •Speculative bubbles bring both losses and meaningful innovation
- •Magnitude feels larger today in both upside and downside implications
- 10:02 – 11:33
Exit vs. keep building: motivations, meaning, and situational decisions
On whether founders should sell, Glenn rejects universal rules and instead focuses on context, confidence, and personal motivation. He frames the decision as a life-allocation choice: what do you want to accomplish with limited time?
- •No general rule for selling; depends on facts, conviction, and options
- •Early-stage companies may welcome offers—survival matters
- •Clarify whether you’re building something meaningful or optimizing for money
- •Time is finite; decisions should align with personal purpose
- 11:33 – 15:23
Travel through the lens of AI: marketplace value and misconceptions
Glenn explains why outsiders often underestimate travel’s complexity and overestimate how easily AI can replace incumbents. He positions AI as a tool to improve Booking’s marketplace mission for both travelers and supply partners, not as an automatic disintermediator.
- •Perceived ‘easy disruption’ narrative ignores industry realities
- •OpenAI’s commerce experiments shifted market perceptions but aren’t definitive
- •Booking serves two customer groups: travelers and partners (a marketplace)
- •AI can make discovery, decisioning, and service cheaper and better
- 15:23 – 17:29
Agentic travel planning: personalized assistants and the ‘domino’ problem
Glenn outlines an agent-driven future where systems understand travelers deeply and handle complex itineraries while preserving human “agency” for final decisions. He highlights that the hardest value appears when things go wrong—AI can coordinate cascading changes across an itinerary.
- •Travel planning is painful; concierge-like help is broadly desired
- •AI agents can outperform humans via memory, permutations, and speed
- •Users still want control/confirmation on complex trips
- •Big opportunity: proactive disruption handling and itinerary re-optimization
- 17:29 – 20:14
Priceline’s Penny in practice: complex itineraries and rapid iteration
Glenn shares a detailed example using Priceline’s AI agent, Penny, to plan a multi-person, multi-city Europe trip with mixed cabin preferences and mileage optimization. The demo illustrates interactive clarification, tradeoff reasoning, and end-to-end itinerary assembly.
- •Penny handles multi-constraint trips (cabins, separate travelers, routing)
- •Agent asks clarifying questions (e.g., miles by airline) and iterates
- •Combines flights, lodging, transfers, and planning logistics
- •Goal is to expand capability and reliability across more scenarios
- 20:14 – 21:39
Agents at scale: token economics, unit costs, and ROI measurement
Despite promising early signals, Glenn emphasizes that Booking’s massive scale requires careful cost control and ROI validation. He discusses token consumption, model selection, and the need to understand lifetime value and loyalty effects before pushing agents everywhere.
- •Booking operates at enormous volume (hundreds of billions in travel)
- •Key constraint: inference cost, token usage, and back-and-forth interaction loops
- •Need to quantify ROI: cost per trip planned vs. conversion and LTV uplift
- •Model choice and price/performance tradeoffs are central operational decisions
- 21:39 – 23:23
AI in customer service: faster resolutions, lower costs, human preference tradeoffs
Glenn describes tangible AI wins in customer service: reduced cost per contact and improved customer satisfaction. He cautions that some customers still want humans, so the best system blends automation with seamless escalation when needed.
- •AI reduces queues and resolves issues faster than traditional staffing models
- •Avoids handoffs and repeat holds by routing to the right capability immediately
- •Metrics cited: lower cost per contact and higher satisfaction
- •Balance needed: offer humans for customers who prefer them
- 23:23 – 26:01
Capital allocation philosophy: reinvestment, acquisitions, and returning cash
Glenn explains how Booking decides what to do with cash flow: prioritize projects with strong ROI, then acquisitions, then shareholder returns. He notes significant buybacks and dividends, arguing idle cash should be returned if not deployable productively.
- •Investment should be justified by clear ROI; otherwise consider M&A
- •If neither investment nor acquisition opportunities fit, return capital
- •Large-scale buybacks over time and ongoing dividends
- •Finance background informs discipline against hoarding cash
- 26:01 – 30:19
Scale as an asset—without complacency: inventory, partners, and regulation
Sarah raises Booking’s durable assets like alternative accommodation supply; Glenn agrees scale matters but reiterates that it’s not a moat. He highlights partner enablement and global regulatory complexity as underappreciated barriers, while warning incumbents must still innovate daily.
- •Booking’s alternative accommodations approach Airbnb’s scale globally
- •Durability comes from transactions, partner relationships, and operations depth
- •Regulation (especially outside the US) is complex and increasing
- •Scale helps absorb compliance and partner-support costs, but innovation is mandatory
- 30:19 – 41:04
Purpose, choosing wisely, and AI’s job impact: reskilling and social stability
Glenn reflects on why he continues: travel meaningfully improves lives by helping people experience the world. He then addresses AI-driven job displacement, stressing that speed of change is the challenge and that companies should upskill workers to prevent backlash and exclusion.
- •Personal mission: making travel easier improves lives and cross-cultural understanding
- •Career advice: choose wisely to avoid later-life regret
- •AI will displace roles (translation and support are concrete examples)
- •Key risk is rapid displacement outpacing new job creation; upskilling is essential
