Street Talk: Finance, investment, M&A, IPOs in travel at The #Phocuswright Conference
Summary
This Phocuswright Conference panel brings together three senior Wall Street analysts — Mark Mahaney (Evercore), Lloyd Walmsley (Mizuho Securities), and Emma Taylor (Barclays) — moderated by Jake Fuller of BTIG, to discuss the financial and strategic implications of AI, investment trends, M&A, and IPOs across the travel sector.
The conversation opens with the recurring question of whether AI will finally displace OTAs, as earlier cycles had predicted for mobile and voice search. Mahaney argues that generative AI will dramatically improve the travel research and planning experience, acting as a new lead-generation platform for OTAs rather than eliminating them. He criticizes Expedia's Romie and Priceline's Penny for being stuck in beta too long and urges faster rollout of agentic products. Walmsley agrees on the consumer experience improvement but flags a more structural concern: even if AI doesn't directly disintermediate OTAs, if it drives more traffic back through Google — where OTAs pay roughly 80 cents per dollar of acquired traffic — their economics deteriorate materially. Taylor shifts the lens to back-end AI ROI, noting that the most measurable near-term AI returns are in call center automation, software product development velocity, fintech, and advertising optimization, rather than top-of-funnel booking changes.
On Google's grip on travel distribution, the panel is broadly optimistic about diversification. Mahaney cites Meta now capturing approximately 10% of performance marketing spend (a shift that took eight years), and points to ChatGPT delivering high-quality leads that OTAs are not yet being charged for. He frames competition between ChatGPT and Google as net-positive for the industry, noting Google search has materially improved in the past two years due to competitive pressure.
The panel debates AI's two core threats to OTAs: disintermediation (AI booking direct with suppliers) versus traffic migration back to Google. Walmsley considers the direct traffic risk more real, pointing to Gemini now being embedded in Chrome browsers and able to comparison-shop live hotel listing pages. He's skeptical of large-scale direct booking via AI based on historical parallels with Google Hotel Finder and metasearch.
On metasearch, the panel discusses Booking's write-down of Kayak as a possible early casualty of generative AI's impact on travel planning. TripAdvisor's pivot away from hotel metasearch toward its Viator experiences marketplace is viewed as a smart but overdue strategic move. TripAdvisor's organic traffic is described as in "meaningful decline" partly attributed to generative AI.
For capital markets, Taylor highlights a resurgence of travel tech IPOs — including a B2B business, a corporate travel company, and Klook's public filing — as a healthy catalyst for both private funding and M&A activity. Investor appetite is strongest in B2B, corporate travel, fintech, ancillary services, and supplier-facing software — categories with contractual, recurring revenue less exposed to consumer traffic risk.
The session concludes with a debate on Airbnb. Walmsley upgraded Airbnb after Brian Chesky publicly committed to aggressively entering hotels — a signal the analysts said they had always flagged as a catalyst. Mahaney takes the opposing view, arguing Airbnb waited too long, has underinvested in innovation, and faces an uphill battle replicating Booking.com's success in blending hotels and alternative accommodations.
Key Takeaways
- •OTAs must accelerate agentic product rollouts (Romie, Penny) — remaining in beta while ChatGPT and Gemini mature is ceding top-of-funnel positioning.
- •The more immediate financial threat to OTAs from AI is not disintermediation but increased traffic dependence on Google, where 80 cents per dollar of acquired traffic is already paid back to Google.
- •Near-term AI ROI in travel is most measurable in cost-center applications — call center automation, software development, ad spend optimization — rather than direct revenue generation.
- •Meta now accounts for roughly 10% of travel performance marketing spend, a diversification that took eight years; ChatGPT leads are high-quality and not yet monetized, signaling a coming shift.
- •Booking's Kayak write-down signals metasearch's structural decline as generative AI absorbs the travel research use case that metasearch once owned.
- •Investors are most receptive to travel tech companies with proprietary data advantages, longtail supplier relationships, or enterprise/corporate access — assets that are inherently less AI-disruptible.
- •Travel tech IPOs (Klook, B2B and corporate travel names) are healthy for the ecosystem because they create new public-currency buyers and validate private market valuations.
- •B2B, corporate travel, fintech, and supplier-facing software are the investment categories attracting the most capital — B2C consumer OTA plays face ongoing skepticism.
- •Experiences/activities remain underpenetrated online with favorable take-rate dynamics; increasing public disclosure from Klook, Viator, and GetYourGuide is proving the unit economics work at scale.
- •Airbnb's hotel expansion is credible given Brian Chesky's product credentials and investor negativity (contrarian opportunity), but faces a structural challenge: any major investment cycle typically compresses valuation multiples before growth reaccelerates.
Transcript
All right. So, I'm going to get my glasses out here. That's a very tough act to follow. So, we got a lot to live up to here, guys. Um, I've been up on this stage a lot of years. Few years back, mobile was going to take over and it was going to upend the OTAAS. A few years after that, it was going to be voice-based searched and the OTAAS were done. And now, you know, throughout a lot of the the last few days here, we've talked about AI, right? And is that going to be the thing? Is this finally th...
