Expedia Group, Inc. (NASDAQ: EXPE) — The #3 Toll Road, Re-Rating on Margins It Finally Earned

Independent equity research Report date: June 14, 2026 Price (ref): ~$225 · Market cap: ~$27B · EV: ~$29B · Net debt: ~$1B Fiscal year: December · CIK: 0001324424 · Segments: B2C, B2B, trivago

⚡ Claude’s Take

This block is the author’s own independent, subjective opinion. It is general information, not investment advice. The analysis that follows takes no position and carries no price target — that view is reserved for this clearly-labeled block alone.

Verdict: BUY / accumulate-on-weakness. Accumulation zone ~$200–235; estimated fair value ~$285–325 (≈10–11× forward adjusted EBITDA, ≈13–15× forward adjusted EPS, ≈11–13× normalized FCF), roughly 25–45% above the reference price. Not a high-conviction franchise bet like Booking — a cheaper, lower-quality, capital-return-and-margin re-rating. Conviction: medium.

Tag: “The number-three toll road, finally collecting a better toll.”

Expedia is the perennial silver-medalist of online travel — smaller, slower-growing, lower-margin and more US-exposed than Booking, with a weaker brand than Airbnb. For a decade that earned it a permanent discount, and deservedly so. What has changed is not the competitive ranking — it is the economics inside the ranking. Under Ariane Gorin (CEO since May 2024), Expedia has executed a genuine cost and platform restructuring: adjusted EBITDA margin expanded ~233 bps in 2025 to ~23.8%, B2C marketing spend is falling in absolute dollars while bookings grow, and the company is throwing off ~$3–4B of free cash flow that it is funneling, aggressively, into a shrinking share count (156M shares in 2021 → ~120M today, a fresh $5B buyback authorized May 2026) and a freshly-reinstated, already-raised dividend. The stock has nonetheless fallen ~25% from its 2025 high to ~$225, screens at the 9.6th percentile of its own decade-long P/E history, carries a strongly negative Momentum factor loading (it is the textbook out-of-favor, formerly-loved name), and trades at ~8–10× EV/EBITDA and ~10× FCF — a turn or two below Booking on every multiple despite a credible, in-the-numbers turnaround.

The framing is value-with-a-catalyst, not quality-compounder. You are not buying the best business in travel; you are buying the cheapest one at the moment its margins inflect and its buyback machine is running hardest, with the B2B (Rapid API) segment — ~$36B of bookings growing 20%+, now ~30% of the company — as the one genuinely-good asset hiding inside a mediocre consolidated story. The honest bears are right that the moat is thin (Expedia is a price-taker to Google, a follower to Booking, and its growth — room nights +6% in Q1-2026 — trails both peers), that the equity story is “margins + buybacks” rather than top-line, and that the governance is ugly (Barry Diller’s dual-class control on ~4.6% economics; three CFOs in 16 months). And the AI-disintermediation question hangs over all three OTAs equally. But at ~10× FCF with ~7% of the market cap returned every year and double-digit EPS growth available from margin and share-count math even on mid-single-digit revenue, the price already discounts the mediocrity. The single fact that flips me more bullish: B2B sustaining 20%+ growth while blended EBITDA margin holds its gains into 2027 — proof the mix shift compounds rather than dilutes. The single fact that flips me bearish: B2C room-night growth stalling toward zero while Booking and Airbnb keep compounding — the fingerprint of structural share loss (to AI, to Google, or to better-run peers) that no buyback can outrun.

1. Executive Summary

Expedia Group is the world’s second-largest online travel agency (OTA) by gross bookings, intermediating ~$119.6 billion of travel bookings in 2025 and converting that into $14.73 billion of revenue (+8%) at a blended take rate of ~12.3%. It operates three reported segments: B2C (the consumer brands — Expedia, Hotels.com, Vrbo, plus Orbitz/Travelocity/ebookers; ~64% of revenue), B2B (the Expedia Travel Group white-label and API business — Rapid, TAAP; ~33% of revenue and the growth engine), and trivago (~59%-owned, separately-listed metasearch; ~3%). It is an asset-light intermediary: it aggregates fragmented hotel and home supply and fragmented traveler demand, and takes a toll on the transaction. The economics are good but not great — ~90% gross margin, ~23.8% adjusted EBITDA margin, ~$3–4B of free cash flow, ~19% ROIC — and structurally a notch below Booking (~37% EBITDA margin) and Airbnb (~35%, ~38% FCF margin).

The investment story is a margin and capital-allocation re-rating, not a growth story. Three facts define it. First, the turnaround is real and in the numbers: under CEO Ariane Gorin, 2025 adjusted EBITDA grew ~19% on ~8% revenue, margin expanded ~233 bps, B2C marketing leverage turned positive (direct sales-and-marketing fell 7% in Q1-2026 while consumer bookings grew 10%), and the company guides another ~100–125 bps of margin expansion in 2026. Second, capital return is aggressive and accelerating: the share count has fallen ~24% since 2022, a fresh $5 billion buyback was authorized in May 2026, and the dividend (reinstated February 2025) was already raised 20%. Total shareholder yield is ~7%. Third, the price discounts the mediocrity: at ~$225 the stock trades at ~8–10× EV/EBITDA, ~20× trailing P/E (~13–15× forward adjusted EPS), and ~10× FCF — the 9.6th percentile of its own ten-year valuation range — after falling ~25% from its 2025 peak, with a deeply negative momentum factor loading marking it as an abandoned former winner.

The bull case is that double-digit EPS growth is achievable from margin expansion and buybacks alone, that the B2B franchise (20%+ growth, ~$36B bookings, new exclusive partners including Uber and Bank of Montreal) is a genuinely good business mispriced inside a mediocre wrapper, and that the AI-disintermediation fear is over-discounted (the OpenAI checkout walkback in March 2026 sent the stock +12%). The bear case is equally real: Expedia is the #3 grower (room nights +6% vs Booking +7%, Airbnb +8%; FY-bookings guide +6–8% vs Booking’s +14–16%), its moat is the thinnest of the three (it is a price-taker to Google and a follower to Booking), its margin gains lap out in 2H-2026, and its governance is poor — Barry Diller controls ~32.6% of the vote on ~4.6% of the economics, and the company has cycled through three CFOs in sixteen months. The AI threat, finally, is genuine and unquantifiable, and Expedia — more metasearch-and-marketing-dependent than Booking — is arguably the most exposed of the majors.

This memo takes no position and sets no price target (that is reserved for Claude’s Take above). It lays out the mechanism, the numbers, and the falsification tests for each side.

2. Business Overview

Expedia Group is a portfolio of online travel brands and a travel-technology platform. It earns money by facilitating reservations between travelers and travel suppliers (hotels, alternative-accommodation hosts, airlines, car-rental firms, activity providers) and taking a commission, merchant margin, or advertising fee on the transaction. As of FY2025 it reorganized its disclosure into three operating segments:

• B2C (Business-to-Consumer) — the consumer-facing brands: Brand Expedia (full-service OTA, the primary growth contributor and “one-stop shop”), Hotels.com (lodging-focused, relaunched with the “Bellboy” campaign), Vrbo (whole-home alternative accommodation, the Airbnb competitor), plus the secondary brands Orbitz, Travelocity, ebookers, Wotif, CheapTickets, Hotwire and CarRentals.com. B2C generated $9.47 billion of revenue in 2025 (+2%) on $83.9 billion of gross bookings (+3%) — a mature, low-single-digit-growth business whose value lever is margin, not volume.
• B2B (Business-to-Business) — the Expedia Travel Group platform that powers travel for third parties: airlines, banks/loyalty programs, corporate-travel managers, other OTAs, and offline agents. Its products are Rapid (a lodging-supply API — the single largest growth driver), TAAP (the Expedia Travel Affiliate Program for agents), and white-label/co-branded templates. B2B generated $4.84 billion of revenue in 2025 (+18%) on $35.7 billion of gross bookings (+20%), and has now strung together roughly eighteen consecutive quarters of strong double-digit growth. It is the highest-quality, fastest-growing, most-defensible part of Expedia.
• trivago — a hotel-metasearch site (Düsseldorf; Nasdaq: TRVG), ~59.2%-owned and consolidated with a noncontrolling interest. It sends referrals to OTAs (including Expedia’s own brands) and monetizes advertising. $417 million of third-party revenue in 2025 (+33%), but only ~$20 million of adjusted EBITDA — small, volatile, and the most AI/Google-exposed corner of the portfolio.

How it makes money — the take rate. The headline operating metric is gross bookings ($119.6 billion in 2025), the total travel value reserved. Revenue ($14.73 billion) is Expedia’s cut: commissions, merchant margin, advertising, and fees — a blended take rate of ~12.3%, which has been roughly flat for three years. Critically, the take rate differs by segment: B2C ~11.3%, B2B ~13.6% — so the B2B mix shift is, counterintuitively, accretive to the blended take rate even though B2B carries a lower EBITDA margin. Revenue is recognized largely at travel/check-in, while gross bookings are booked at reservation; the lag creates a working-capital float and a timing wedge between the bookings and revenue growth rates within a year.

Revenue by model and service. By business model, 2025 revenue split ~70% merchant (Expedia is merchant-of-record, collects the traveler’s cash up front, remits to the supplier later), ~22% agency (the supplier collects; Expedia invoices a commission), and ~8% advertising/media/other. By service type, lodging is ~80% of revenue ($11.75 billion, +7%) — overwhelmingly the profit driver — with air a trivial ~3% ($407 million, −5%) (a deliberately low-margin traffic-driver), Expedia Group Advertising $758 million (+19%) (high-margin, fast-growing media), and other (car, insurance, cruise, activities) ~$1.4 billion.

Geography. Expedia is the most US-centric of the major OTAs: the US is ~59% of revenue ($8.7 billion), international ~41% ($6.0 billion) — and the international book is growing faster (+13% vs US +4%). This is the mirror image of Booking (international-led) and a structural fact: Expedia’s home turf is the most mature, most competitive OTA market on earth.

Recurring vs. non-recurring. There is no contractual recurring revenue; every booking is a fresh transaction. The behavioral recurrence comes from the ~2/3 of bookings that arrive via direct channels (app, repeat, loyalty) and the One Key unified loyalty program (cross-earn/cross-redeem across Expedia, Hotels.com and Vrbo, launched US-2023, UK-2024, expanding globally). One Key is the closest thing to a moat asset Expedia is building — but the company discloses no member counts or engagement metrics, and management would only say active members grew “mid-single-digits,” which trails bookings growth. That opacity is itself a tell.

Verdict. A clean, asset-light, transaction-toll model with one genuinely good franchise (B2B/Rapid), one large mature franchise being optimized for margin (B2C), and one small structurally-challenged satellite (trivago). The business is well understood and the model is sound; the issue is quality relative to peers, not viability. The most important structural fact is the B2B mix shift — accretive to take rate and growth, dilutive to consolidated margin — which is quietly re-shaping what Expedia is.

3. Industry Dynamics

Market size and growth. Global travel gross bookings reached ~$1.7 trillion in 2025 (Phocuswright), growing ~5–6% per year as a mature end-market. The relevant secular engine is the online subset — ~$1.07 trillion in 2025 (+8%) — which structurally grows ~8–12% per year (roughly double total-travel growth) on the persistent offline-to-online migration, concentrated in emerging Asia and Latin America. Online travel is a beneficiary of the durable post-pandemic “experiences over goods” shift in developed-market consumption. This is a good secular backdrop — but Expedia, US-centric and #3, captures less of the fastest-growing (Asian, emerging-market) pools than Booking (Agoda) or Trip.com.

Profit pools and value-chain position. The economic rent in travel concentrates in the asset-light intermediary layer. OTAs earn high-margin commissions on third-party inventory with no real estate and no fleet (hence 20–37% EBITDA margins across the group). The supply side (hotels) captures the bulk of gross dollars but is capital-intensive and fragmented. Metasearch (trivago, KAYAK, Google, Trivago) monetizes referrals at lower margin and a thinner moat — and is the layer most exposed to Google and AI (which is exactly why trivago earns only ~$20 million of EBITDA on ~$417 million of revenue). The OTA sits at the chokepoint between fragmented supply and fragmented demand. But Expedia’s chokepoint is leakier than Booking’s: where Booking’s strength is independent European/Asian hotels and a ~65%-direct demand base, Expedia is more reliant on chain hotels (more able to push direct), US metasearch, and Google-sourced paid traffic — the layer where margin leaks out.

Competitive intensity — Expedia is structurally #2 (or #3). The accommodation-OTA market is a global duopoly-plus: Booking (dominant Europe/Asia, $186B gross bookings, ~14.5% take rate, ~37% EBITDA margin) and Expedia (strongest in the US, $119.6B bookings, ~12.3% take rate, ~24% EBITDA margin), with Airbnb owning the branded whole-home category ($91B GBV), Trip.com dominant in China and rising across APAC, and a long regional tail (MakeMyTrip, Despegar). On the metrics that matter, Expedia is the laggard of the big three:

Expedia is the slowest-growing and lowest-margin of the three, on a smaller-than-Booking but larger-than-Airbnb base. That is the core competitive fact the valuation must reflect.

Where the take rate comes from, and why Expedia’s is lower. The OTA’s take rate is the equilibrium of a tug-of-war: hotels accept it because the OTA delivers incremental demand at a customer-acquisition cost below the hotel’s own; travelers accept it implicitly for selection, comparison and trust. Expedia’s ~12.3% sits ~2 points below Booking’s ~14.5% for structural reasons: a higher mix of chain hotels (which negotiate harder and push direct), a US/merchant skew, a meaningful low-take air business, and historically weaker conversion. The B2B mix shift lifts the blend (B2B take rate ~13.6%), which is one reason the blended rate has held flat despite competitive pressure.

Regulatory landscape. Europe is the binding constraint for the whole industry — the EU Digital Markets Act (which designated Booking.com a “gatekeeper,” banning price-parity clauses) and national competition actions are eroding the supply-side moat. Expedia is less directly targeted than Booking (it was not named a DMA gatekeeper), but it operates under the same parity-and-distribution pressures, occupancy-tax litigation in the US, and recurring Italian withholding-tax settlements (~$178 million in 2025, ~$107 million in 2024) that are a genuine, repeating cash cost depressing the run-rate. Alternative-accommodation regulation (STR caps) is a Vrbo-specific overhang shared with Airbnb.

The capital-cycle lens (Marathon). Online travel is the rare industry where the dominant incumbents are also the low-cost producers; capital intensity is trivial (Expedia’s capex is ~6% of revenue including capitalized software, low in absolute terms). The danger sign in Marathon’s framework is high returns attracting a flood of capital that competes returns away — and at the supply-aggregation layer that flood has not come (building a global supply graph is a low-ROIC, decade-long slog no rational entrant attempts head-on). Instead capital has flowed to adjacent layers — Airbnb in homes, Google and now AI in demand acquisition. For Expedia specifically, the capital-cycle read is mixed: it benefits from the same supply-side barrier as Booking, but it sits closer to the contested demand-acquisition layer (more metasearch, more Google dependence) where the AI capital wave is breaking. It is the major most exposed to a demand-layer disruption.

Verdict: a structurally attractive industry, in which Expedia occupies a structurally inferior seat. The secular online tailwind is intact and the intermediary profit pool is real and high-margin. But Expedia is the #3 player on growth and #2-to-#3 on margin, more US-exposed (most mature market), more Google/metasearch-dependent (most AI-exposed demand layer), and lower-take-rate than Booking. A good industry; a mediocre position within it. The investment case cannot rest on the industry doing the work — it must rest on Expedia closing the self-inflicted portion of its quality gap (margin, platform efficiency), which is precisely what management is doing.

4. Competitive Position

Name the moat. Expedia has a real but second-tier moat, and it is honest to be precise about its source and its limits. In Greenwald’s taxonomy, the genuine advantage is economies of scale combined with (weak) customer captivity:

• Supply-side scale (real). Expedia aggregates ~3.7 million lodging properties (with ~800,000 exclusive/direct relationships per management), global air, car, and activity inventory, and a payments-and-merchandising platform. A new entrant cannot replicate this graph cheaply; it is a decade-long, low-ROIC build. This is the same barrier that protects Booking — but Expedia’s graph is smaller and less differentiated, particularly in the unique long-tail (where Airbnb wins) and in independent European/Asian hotels (where Booking wins).
• The B2B flywheel (the best moat asset). Rapid (the lodging API) turns Expedia’s supply graph into a wholesale utility that powers other companies’ travel offerings — Uber, Bank of Montreal, airlines, banks, other OTAs. This is genuinely sticky: once a partner integrates Expedia’s API into its product and earns on the margin, switching is costly and disruptive. Eighteen consecutive quarters of 20%+ growth, with exclusive wins, is the financial fingerprint of a real competitive advantage. B2B is the part of Expedia that most resembles a durable franchise.
• Brand/loyalty (weak captivity). The consumer brands have awareness but thin switching costs — travelers multi-home, comparing Expedia, Booking, Google, and the hotel’s own site freely. One Key is an attempt to manufacture captivity via cross-brand loyalty, but its undisclosed, mid-single-digit member growth suggests it is not yet a binding lock-in.

Where the moat leaks — the Google tax and the demand layer. The decisive weakness is demand acquisition. Like all OTAs, Expedia rents traffic from Google (performance marketing is the single largest cost line), and it is more dependent on paid/metasearch channels than Booking, whose ~65%-direct demand base insulates it. Expedia’s direct mix (~2/3 of bookings per management) is improving and its 2025 achievement of growing bookings while cutting marketing dollars is the single most encouraging competitive datapoint in the story — but the structural fact remains that Expedia keeps less of each demand dollar than Booking and is more exposed to Google’s algorithm, Google’s own travel functionality, and now AI search reconstituting the discovery layer.

Head-to-head. Versus Booking, Expedia loses on take rate (12.3% vs 14.5%), margin (24% vs 37%), growth (room nights +6% vs +7%; FY guide +6–8% vs +14–16%), international footprint, and direct-demand mix — it wins on nothing structural except a stronger US position and a faster-growing B2B platform. Versus Airbnb, Expedia’s Vrbo is a distant #2 in whole-home (Vrbo supply is far smaller than Airbnb’s >9 million listings and weaker in brand), though Expedia’s multi-product OTA breadth (flights, cars, packages) is something Airbnb lacks. Versus the hotel chains’ direct apps, Expedia faces the same direct-booking pressure as all OTAs, more acutely in its chain-heavy US book.

The financial test of the moat. The right standard: if a “moat” claim can’t be tied to a financial outcome that would deteriorate without it, it isn’t a moat. Expedia’s ROIC rose from ~10.6% (2023) to ~19.4% (2025) and adjusted EBITDA margin to ~24% — evidence the scale-and-platform advantage is real and that returns are well above cost of capital. But the level (24% EBITDA margin, 19% ROIC) versus Booking (37%, far higher) is the quantified measure of the moat gap: Expedia has a moat, but it is demonstrably shallower, and its returns sit a tier below the best business in the category.

Verdict: a real but second-tier competitive advantage — genuine supply-side scale and a strong, sticky B2B franchise, undercut by thin consumer captivity, heavier Google/metasearch dependence, and a structurally lower take rate and margin than Booking. The moat is widening on the B2B side and (modestly) on direct-mix/marketing efficiency, while remaining narrow on the consumer side. It is investable, but it is not Booking, and the price must reflect that.

5. Growth History and Forward Opportunities

Historical growth. Revenue recovered from the COVID trough of $5.2 billion (2020) and $8.6 billion (2021) to $11.7 billion (2022), $12.8 billion (2023), $13.7 billion (2024) and $14.7 billion (2025) — a ~7% CAGR over the last two years as the post-COVID recovery normalized into a mature growth rate. The more important story is composition and quality: B2C (the bulk) is now a low-single-digit grower (+2% in 2025), while B2B compounds at ~18–20% and trivago is small and volatile. The consolidated ~8% masks a two-speed company.

The 2025–2026 inflection — quality is improving even as the rate stays modest. Q1-2026 was a clean beat: gross bookings +13% to $35.5B, revenue +15% to $3.4B, adjusted EBITDA +83% to $542M (margin 15.8%, “highest Q1 in 15 years”), adjusted EPS $1.96 (~4× YoY). Underneath: consumer bookings +10% — the “fastest in 12 quarters” — with all three core brands growing for two consecutive quarters, Vrbo back to growth, and Brand Expedia’s vacation-rental line hitting a $1 billion annualized run-rate for the first time. This is the durable, organic improvement the bull case needs. The asterisk: room nights grew only +6% (the air-pocket from Mexico travel advisories and the Middle East conflict cost ~2 points in March), and the EPS quadrupling was flattered by margin lapping and buybacks, not a volume surge.

Forward opportunities (ranked by quality):

1. B2B / Rapid (the best). ~$36B of bookings growing 20%+, ~2/3 outside the US, with exclusive new partners (Uber as exclusive hotel partner, Bank of Montreal AIR MILES). This is a genuine flywheel — more supply → more partners → more volume → more supply — and the single most attractive growth vector in the company. The caveat: part of B2B growth depends on partners’ discretionary marketing spend, which management flagged as decelerating sequentially in Q1-2026 — a quality-of-growth watch-item.
2. Margin expansion (the near-term EPS driver). Management guides ~100–125 bps of further adjusted EBITDA margin expansion in 2026 (“high end”), on top of ~233 bps in 2025, from marketing efficiency, platform consolidation, and headcount actions. This is not “growth” in the revenue sense but is the dominant near-term earnings driver — and it laps out in 2H-2026, after which the algorithm reverts to revenue-plus-modest-leverage.
3. Advertising (EG Advertising). ~$758M (+19%), high-margin, accretive to B2C margin, with new video formats and AI-targeted inventory. A small but genuinely high-quality, fast-growing profit pool.
4. One Key loyalty + Vrbo recovery. Cross-brand loyalty and the alternative-accommodation comeback (supplier-funded promos now >1/3 of Vrbo bookings) are real but unquantified levers.
5. Tiqets / experiences (optionality). The agreed acquisition of Tiqets (activities/experiences, Amsterdam, B2B-tilted, Dec-2025) adds a high-frequency, high-margin attach category — modest, sensible bolt-on.

The honest constraint. Even with all of this, Expedia guides FY2026 gross bookings +6–8% — roughly half Booking’s +14–16% and below Airbnb’s low-teens. The company has beaten its own guide repeatedly yet refuses to raise the full-year outlook, citing macro caution. The growth algorithm is structurally mid-single-digit-to-high-single-digit on the top line; double-digit EPS growth comes from margin + buybacks, not volume.

Verdict: medium-quality growth, improving in mix and margin but capped in rate. The B2B engine is high-quality and the 2025–2026 consumer reacceleration is encouraging and organic. But the consolidated top line is structurally the slowest of the big three, and the near-term earnings growth is dominated by margin self-help and share-count reduction — both finite. This is a re-rating-and-capital-return story dressed in modest organic growth, not a secular grower.

6. Financial Quality

Revenue and margins. Revenue of $14.73 billion (2025, +8%) at a ~90% gross margin (COGS is mostly merchant/payment costs). The headline financial-quality story is operating leverage: GAAP operating margin rose from 11.9% (2023) to 12.2% (2024) to 14.7% (2025); adjusted EBITDA margin from 18.2% to 18.3% to ~20.7% (ROIC basis) / ~23.8% (company segment basis) — ~233 bps of expansion in a single year, with an incremental operating margin of ~47% in 2025. The company is converting modest revenue growth into strong profit growth, the essence of the turnaround.

Returns on capital. ROIC improved markedly: 10.6% (2023) → 14.2% (2024) → 19.4% (2025) — comfortably above cost of capital and evidence the scale advantage is real. ROE of ~112% is arithmetically true but economically meaningless: it reflects a near-zero (in fact negative-tangible) book equity created by years of buybacks (treasury stock ~$16.8 billion, goodwill+intangibles ~$7.7 billion against ~$2.5 billion total equity including ~$1.3 billion of trivago minority interest). For Expedia, ROIC and FCF-on-EV are the meaningful return metrics; book-based ratios (P/B at the 99.5th percentile of its own history) are noise.

Cash generation — strong, but read it carefully. Operating cash flow was $3.88 billion in 2025; management cites TTM free cash flow of ~$4.1 billion (Q1-2026 basis) and FY2025 FCF of ~$3.1 billion. Two structural features inflate cash flow above net income, and both must be understood:

• The merchant/loyalty float (a genuine, durable strength). As merchant-of-record on ~70% of bookings, Expedia collects traveler cash up front and remits to suppliers later, plus holds deferred-revenue/loyalty balances (~$10.6 billion of current deferred merchant bookings and loyalty liabilities). This produces a negative cash-conversion cycle of ~−706 days and a customer-funded working-capital benefit (~$1.17 billion in 2025). Growth funds itself. This is a real, durable benefit (like Booking’s and Airbnb’s float) — but it means OCF overstates steady-state FCF in growth years and would reverse in a downturn (as it did, painfully, in 2020).
• Capitalized software and SBC. Expedia capitalizes significant software development (a real, recurring cash cost embedded in capex) and adds back $398 million of stock-based compensation (2025, down from $458 million — a genuine economic cost and ~2.7% of revenue, lower than Airbnb’s ~13%). Normalizing for the float benefit and capitalized software, underlying owner-FCF is closer to ~$2.5–3.0 billion than the headline $4.1 billion — still a ~9–11% FCF yield on the ~$27 billion market cap.

Balance sheet. Conservative. Cash and short-term investments of ~$5.7–5.8 billion against ~$6.4 billion of total debt → net debt of only ~$1 billion (~0.3× adjusted EBITDA), with a well-laddered maturity schedule (2026–2036), a $2.5 billion undrawn revolver, and a committed investment-grade rating. The 2026 maturities ($750M senior + $1B convertible) are easily covered by cash and the April-2026 $1B issuance. This is a balance sheet built for continued buybacks, not deleveraging.

Quality-of-earnings flags (honest accounting, not business deterioration):

• Recurring “one-time” items. The Italian withholding-tax settlements (~$178M in 2025, ~$107M in 2024) and serial restructuring charges (~$100M in 2025, ~$72M in 2024) are labeled non-recurring but recur, depressing GAAP run-rate. Normalizing them out, FY2025 operating earnings are higher than GAAP — a tailwind to the bull case, but also a reminder that “adjusted” numbers are doing real work.
• Past impairments. 2023 carried ~$297M goodwill + ~$129M intangible impairments; 2024 a ~$147M intangible impairment. None in 2025 — improving, but a reminder the portfolio has had value-destroying corners (legacy brands, trivago).
• GAAP vs. adjusted gap. The bridge from GAAP EPS ($9.81 diluted, 2025) to adjusted EPS (meaningfully higher) runs through SBC, restructuring, legal, and amortization. The gap is reasonable for the sector but means the “20× P/E” (trailing GAAP) overstates richness while “13–15× forward adjusted EPS” is the more representative figure.

Verdict: economics are good and improving with scale, a clear tier below Booking and Airbnb. Margins are expanding, ROIC is strong (~19%) and rising, the balance sheet is conservative, and cash generation is robust and self-funding. The honest deductions: the ~24% EBITDA margin is ~13 points below Booking’s; headline FCF is flattered by float and should be read net of capitalized software; and “adjusted” earnings rely on normalizing recurring “one-time” charges. On the metric that matters — durable, growing, capital-light cash generation — this is a solid (not elite) financial profile that is getting better.

7. Capital Allocation

Philosophy. Expedia generates far more cash than its asset-light model can reinvest, so capital allocation is the bridge from business value to shareholder value — and here Expedia scores well. The approach under Gorin (and Diller as Chairman): modest organic/M&A reinvestment, large and consistent buybacks, a reinstated-and-growing dividend, and disciplined balance-sheet management toward investment grade.

Buybacks — the dominant lever, executed aggressively. Share repurchases have run consistently above ~$1.6–2.0 billion per year: $2.03 billion (2023, 19.1M shares @ $106), $1.62 billion (2024, 12.1M @ $134), $1.66 billion (2025, 9.0M @ $185), and $700 million in Q1-2026 (3.3M @ $212). The share count has fallen from 156 million (2021) to ~120 million today — a ~24% reduction — and in May 2026 the board authorized a fresh $5 billion program. This is the engine that converts mid-single-digit revenue growth into double-digit per-share growth. Two honest caveats: (1) the average repurchase price has climbed sharply ($106 → $185 → $212), so recent buybacks are being done at higher valuations than the 2023 lows — the discipline test (“buy when cheap”) is passing less cleanly than it did; and (2) unlike Booking, Expedia has not bought back so aggressively as to drive book equity deeply negative on a net basis, but treasury stock (~$16.8 billion) shows the cumulative scale.

Dividend. Reinstated February 2025 at $0.40/quarter ($1.60/year, ~$200M) after a COVID-era suspension, and already raised 20% to $0.48/quarter in February 2026 — a signal of confidence in the FCF run-rate. Combined with buybacks, total shareholder yield is ~7% — competitive with Booking’s ~6.5% and above Airbnb’s net ~3%.

M&A — disciplined and small. Expedia has avoided large, value-destructive deals in recent years. The pending Tiqets acquisition (activities/experiences, Dec-2025) is a sensible, B2B-tilted bolt-on into a high-frequency attach category. trivago remains a ~59%-owned legacy holding (a structurally challenged metasearch asset, ~$20M EBITDA) that the company has neither fully integrated nor divested — a minor capital-allocation loose end.

Incentive alignment — mixed, with a notable gap. Executive compensation is base salary + equity only (no traditional annual cash bonus), with PSUs (50/50) vesting on annual Revenue Growth and Adjusted EBITDA Margin growth (bps) over 2025–2027. The positives: the metrics are operationally relevant and the structure is equity-heavy (CEO Gorin’s FY2025 comp was ~$17.6M, ~$15.7M in stock). The gaps: there is no relative-TSR metric, no FCF metric, and no ROIC metric — so management is incentivized to grow revenue and margin but not explicitly to create per-share value or allocate capital well, and not measured against peers. The 2025 PSUs paid out 176.9% (revenue +7.6% vs 6% target; margin +233 bps vs +50 bps target, maxed at 200%) — a rich payout in a year the stock fell, illustrating the disconnect between the comp metrics and shareholder outcomes.

Insider behavior — opaque, no evidence of conviction buying. Form 4 transaction codes could not be read from the mirrored corpus (an open item requiring a direct EDGAR pull), but the filing cadence clusters around vesting windows, consistent with routine RSU/PSU vest-and-withhold rather than discretionary open-market purchases. The proxy shows Barry Diller’s holdings static (no open-market buying), with Vanguard (13.3%) and BlackRock (8.7%) the largest institutional holders. There is no insider buying signal to corroborate the value case.

Verdict: management has allocated capital intelligently — the clearest positive in the story — with two asterisks. Aggressive, consistent buybacks (24% share-count reduction), a growing dividend, ~7% shareholder yield, disciplined small M&A, and a conservative IG balance sheet are exactly right for a cash-rich, modest-growth business. The deductions: buybacks are being done at progressively higher prices, and the incentive structure rewards revenue/margin but not per-share value or capital-return efficiency — a governance weakness that, combined with the Diller control structure , tempers the otherwise-strong grade.

8. Changes and Headwinds — Last Two Years

Leadership overhaul. The defining change is management. Ariane Gorin became CEO in May 2024, succeeding Peter Kern, and has driven the platform-consolidation and margin-expansion program that defines the current story. Barry Diller remains Chairman and Senior Executive, controlling 100% of Class B shares — ~32.6% of the vote on ~4.6% of the economics (a dual-class structure that is a persistent governance overhang). Most concerning: three CFOs in sixteen months — Julie Whalen departed February 2025, Scott Schenkel stepped in and then announced his own departure (effective May 2026), succeeded by Derek Andersen (ex-Snap CFO). The filings state no disagreement on accounting or operations, and Andersen’s Snap pedigree signals continued cost discipline — but a revolving CFO door is a real execution-continuity and internal-controls risk worth flagging.

Strategic/operational changes. (1) Reorganization into B2C/B2B/trivago segments, surfacing the B2B growth story. (2) Platform consolidation onto a unified technology stack and rationalization toward three core consumer brands (Expedia, Hotels.com, Vrbo). (3) One Key unified loyalty rollout. (4) Marketing-efficiency program — the headline achievement of growing bookings while cutting marketing dollars. (5) Restructuring (Feb-2024 and 2025–2026 headcount actions, rehiring into AI/ML). (6) Capital-return reset — dividend reinstated and raised, fresh $5B buyback. (7) Tiqets acquisition (experiences). (8) AI/agentic product push — ChatGPT ads (Feb-2026), Claude and Alexa+ integrations, internal AI productivity.

Headwinds (real and persistent):

• AI / agentic disintermediation — the central long-term risk for all OTAs; Expedia, being more metasearch/Google-dependent, is arguably the most exposed of the majors. (Offset near-term: the OpenAI checkout walkback in March 2026 suggested the threat is delayed; the stock rose +12%.)
• #3 growth position — structurally slower than Booking and Airbnb; the equity story leans on margin and buybacks, both finite.
• Margin-expansion runway narrowing — management explicitly flagged that 2H-2026 margin gains will be “muted” as 2025 cost actions lap.
• Macro/geopolitical sensitivity — high-beta, deeply cyclical (lifetime max drawdown −82.7%); the March-2026 air-pocket (Mexico advisories, Middle East conflict) showed how quickly demand wobbles.
• Governance — Diller dual-class control; CFO churn; comp metrics misaligned with per-share value.
• Recurring tax/legal drag — Italian withholding-tax settlements, US occupancy-tax litigation.
• Google dependence — performance-marketing tax and algorithm/AI-search risk to the demand layer.

Verdict: the changes strengthen the thesis on balance, while the headwinds are persistent rather than acute. The leadership-driven margin turnaround, capital-return reset, and B2B momentum are genuine improvements widening the investable case. The headwinds — #3 position, AI risk, governance, narrowing margin runway — are real and cap how much credit to extend, but none is acutely threatening on a 12–24-month view. Net: a franchise being meaningfully improved by management, against a structural-position and AI overhang that the (cheap) price already substantially reflects.

9. Risk Analysis (Risk Matrix)

Catastrophic / total-loss risk. Low in the near-to-medium term: net-debt only ~$1 billion, investment-grade, FCF-generative, no refinancing wall, diversified across segments and geographies. A permanent impairment scenario requires either (a) agentic AI structurally severing OTAs from demand (slow-moving, currently speculative, and shared with peers), or (b) a deep, prolonged travel recession that reverses the float and crushes the cyclical earnings (the 2020 precedent — but the balance sheet is far stronger now). The realistic downside is a value-trap / multiple-compression scenario — the stock stays cheap because the #3 position and AI overhang prevent a re-rating — not a zero.

10. Valuation Discussion (Embedded Expectations)

The setup (as of ~$225/share). Market cap ~$27 billion (~120 million shares post-Q1 buyback); net debt ~$1 billion; trivago minority interest ~$1.3 billion; EV ~$29 billion. Against FY2025: revenue $14.73B, adjusted EBITDA ~$3.5B (company basis) / ~$3.05B (ROIC/GAAP-ish), GAAP operating income ~$1.87B, GAAP diluted EPS $9.81, TTM EPS ~$11.35, headline TTM FCF ~$4.1B (normalized owner-FCF ~$2.5–3.0B).

Multiples — cheap on an absolute and own-history basis:

• EV/EBITDA ~8.4× (on ~$3.5B adj.) / ~9.6× (on ~$3.05B) — versus Booking’s ~13× and Airbnb’s ~16×. A clear discount, partly deserved (lower margin/growth), partly the opportunity.
• P/E ~20× trailing GAAP / ~13–15× forward adjusted EPS — the trailing figure sits at the 9.6th percentile of Expedia’s own ten-year P/E history (AZI valuation index), i.e. near the cheapest it has ever been on earnings.
• P/FCF ~7–10× (headline vs. normalized) — a ~10–11% FCF yield.
• Shareholder yield ~7% (buybacks + dividend).
• P/S ~1.9× / EV/Sales ~2.0× — ~61st percentile of own history (the only metric not screaming cheap, because the sales multiple already reflects the lower-margin mix).
• Ignore P/B (99.5th percentile) — meaningless given buyback-depleted, negative-tangible book equity.

Embedded-expectations analysis — what is the price underwriting? At ~$29 billion EV and ~$3.5 billion adjusted EBITDA, the market is paying ~8.4× for a business it expects to grow EBITDA modestly. Reverse-engineering: if you assume Expedia simply executes its own guidance — ~6–8% bookings/revenue growth, ~100–125 bps annual margin expansion through 2026 then flattening, and continued ~$2 billion/year buybacks at a ~24% historical share-reduction pace — adjusted EPS compounds at a low-double-digit rate (mid-single-digit revenue × margin expansion × ~3–4% annual share shrink). At ~13–15× forward adjusted EPS for a low-double-digit EPS grower with a ~7% shareholder yield, the current price embeds expectations of essentially no re-rating and no acceleration — i.e. the market is pricing Expedia as a structurally-stuck #3 whose margin gains will fade and whose AI exposure will eventually bite. That is a low bar.

What the market is pricing correctly: that Expedia is the slowest-growing major; that its moat and margin are a tier below Booking; that the margin-expansion runway narrows after 2H-2026; that governance is poor; and that the AI threat is real and unquantifiable.

What the market may be pricing incorrectly: the durability and quality of the B2B franchise (a 20%-grower worth a far higher multiple than the consolidated 8×); the capital-return compounding (a 24%-share-reduction track record plus a fresh $5B authorization is a powerful per-share engine the low multiple under-credits); and the possibility that the AI fear is over-discounted (the OpenAI walkback and Google’s “route-through-OTAs” agentic design both support management’s “net positive / delayed” framing).

Scenario analysis (illustrative; not a forecast, no price target):

• Bear: B2C nights stall toward zero, margin gains lap and reverse, AI begins measurably diverting demand. Adjusted EBITDA flatlines ~$3.3–3.5B; the multiple compresses to ~7× EV/EBITDA. Equity value contracts ~20–30% — the value-trap outcome.
• Base: Expedia executes guidance — ~7% bookings growth, margin to ~25–26%, ~$2B/yr buyback. Adjusted EBITDA ~$3.8–4.0B (FY2026), multiple holds ~9–10×. With share-count shrink, per-share value compounds low-double-digits; modest re-rating toward the recent-year average.
• Bull: B2B sustains 20%+, blended margin holds its gains, AI proves net-positive, and the multiple re-rates toward ~11–12× EV/EBITDA (still a discount to Booking) on ~$4B+ EBITDA. Combined with buybacks, meaningful upside.

Verdict: On embedded expectations, Expedia is priced as a permanently-stuck #3 — a low bar that its own guidance plus its buyback machine can clear. The valuation is genuinely cheap on EBITDA, FCF, and own-history P/E, and the principal risk to the cheapness is not over-payment but a value trap: a business that stays cheap because the market is right that nothing structural improves. No price target (see Claude’s Take for a subjective zone).

11. Variant Perception

Consensus belief. Expedia is the mediocre #3 OTA — a slower-growing, lower-margin Booking with worse governance and more AI exposure — that has run a successful but finite margin-cost program and is now a “show-me” story on whether growth can hold. Sell-side is mildly constructive on the cheapness (e.g., BTIG Buy, $330 PT) but the buy-side treats it as a low-quality cyclical: a former momentum name now abandoned (Glenview exited in Q1-2026), screening as a deep-value, anti-momentum, high-beta cyclical (negative Momentum and DividendYield loadings, +Value, +SmallSize, 9.6th-percentile P/E).

The strongest bull case. The price discounts the mediocrity, and you are paid ~7% a year to wait for a re-rating that the buyback machine is engineering regardless. At ~8–10× EV/EBITDA and ~10× FCF, Expedia can compound per-share value at a low-double-digit rate on mid-single-digit revenue, purely from margin and a 24%-and-counting share-count reduction. Inside the cheap consolidated multiple sits a genuinely good, sticky, 20%-growing B2B franchise (Rapid, with exclusive Uber and Bank of Montreal wins) that would command a far higher standalone multiple. The 2025–2026 consumer reacceleration (bookings +10%, all brands growing, Vrbo recovering) shows the B2C franchise is not in structural decline. And the AI fear — the main thing capping the multiple — is over-discounted: the OpenAI checkout walkback (+12% reaction) and Google’s “route-through-OTAs” agentic design suggest the OTAs are suppliers to, not victims of, the AI layer. New, cost-focused management (CEO Gorin, ex-Snap CFO Andersen) is executing.

The strongest bear case. It’s cheap for a reason, and “margins + buybacks” is what you say about a business that can’t grow. Expedia is the #3 grower on every volume metric and guides to half Booking’s bookings growth — structural share loss, not a one-off. The margin-expansion story laps out in 2H-2026, after which the EPS algorithm reverts to a slow top line plus a buyback funded at ever-higher prices. The moat is the thinnest of the majors: a price-taker to Google, a follower to Booking, with thin consumer captivity (One Key metrics undisclosed and underwhelming) and the most AI-exposed demand layer. Governance is poor — Diller’s dual-class control, comp metrics that ignore per-share value and TSR, and three CFOs in sixteen months. If agentic AI commoditizes OTAs into back-end inventory, Expedia — more metasearch/Google-dependent — gets hit first and hardest, and the “cheap” multiple proves to be a value trap.

The 3–5 assumptions that matter most:

1. Does the margin expansion stick (and compound) past 2H-2026, or does it lap out? (Bull needs durable ~25%+ EBITDA margin; bear expects a plateau/give-back.)
2. Can B2C room-night growth hold mid-single-digits, or does it decelerate toward zero vs. accelerating peers? (The cleanest tell of structural share loss.)
3. Does B2B sustain 20%+ growth at a stable margin, or does partner-marketing dependence slow it? (The quality of the one genuinely-good asset.)
4. Is AI net-positive, neutral, or disintermediating for Expedia specifically? (The binary swing on the multiple — and the one with no disclosed data yet.)
5. Does the buyback keep compounding per-share value, or does rising-price execution and a fading top line blunt it?

The factor-positioning read. Expedia is the textbook out-of-favor, formerly-loved cyclical: y1 return +32% but m6 return −18% (−35.5% ann.) (Sharpe −0.74), rs_6m −17.6, with a strongly negative Momentum loading (−0.49), negative DividendYield and Growth, and positive Value and SmallSize — a high-beta (~1.2–1.34), smaller-cap, value-tilted name the market has abandoned after a sharp 6-month drawdown. This is not a falling knife in the structural-decline sense (the business is improving, not deteriorating) but a sentiment/momentum washout in a cheap, cyclical, capital-returning stock — precisely the profile where a value/contrarian re-rating tends to pay, and precisely the profile that can also stay cheap for years if the growth narrative never turns. The tape says consensus is offsides bearishly on a name that is operationally improving; the risk is that “cheap and improving margins” is not enough to overcome “structurally #3” in the market’s eyes.

Where I come out (Variant view): consensus is modestly offsides — it is extrapolating the #3 growth position into a permanent impairment while under-crediting the margin turnaround, the B2B franchise quality, and the capital-return compounding. The disagreement is medium-conviction, not high: the bear case is genuinely sound, and the binary AI risk is unresolved. This is a cheap, improving, lower-quality business where the price already does much of the work — a value-with-a-catalyst setup, not a quality-compounder bargain.

12. Fact vs. Interpretation

13. Open Questions

1. Insider transactions (code-level): the Form 4 corpus was not machine-readable from mirrored files — has any insider made open-market purchases (code P), or is it all routine vest-and-withhold? (Requires a direct EDGAR Form 4 pull.) No buying signal found in the proxy.
2. One Key economics: absolute member count, tier mix, repeat/direct rates, and incremental margin — all undisclosed. Is loyalty a real captivity moat or a marketing line?
3. B2B standalone economics: what is Rapid’s true contribution margin and customer-concentration (how much rides on the largest partners, e.g. Uber)? How durable is the partner-marketing-funded portion of growth?
4. AI-channel data: what quantified share of traffic/bookings now originates from AI/answer-engine channels, and what is its conversion/retention vs. the load-bearing “2/3 direct” base?
5. Margin ceiling: how much of the 2025–2026 margin expansion is structural (platform/efficiency) vs. cyclical/one-off (marketing pullback that competitors could force back up)? What is the steady-state EBITDA margin?
6. trivago: keep, integrate, or divest the ~59%-owned, ~$20M-EBITDA metasearch stake?
7. CFO stability: will Andersen stay, and what does a third CFO in 16 months imply for forecasting reliability and controls?

14. What Must Be True

For the bull case to be right:

• B2C room-night growth holds at least mid-single-digits while peers grow — i.e. no structural share loss. Falsification test: two-plus consecutive quarters of B2C room-night growth decelerating toward zero (sub-3%) while Booking and Airbnb sustain mid-to-high-single-digit night growth → the bull case is broken; this is share loss, and no buyback outruns it.
• Adjusted EBITDA margin holds and compounds toward ~25%+ through 2026 and does not give back in 2027 as cost actions lap. Falsification: 2027 margin flat-to-down with no offsetting growth → the “margins + buybacks” engine has stalled.
• B2B sustains ~20% growth at a stable margin. Falsification: B2B growth decelerating to low-double-digits/high-single-digits with margin compression → the one good asset is deteriorating.
• AI proves net-neutral-to-positive (no measurable demand diversion). Falsification: disclosed or evident erosion of the direct-booking mix / rising CAC attributable to AI search.

For the bear case to be right:

• The #3 position calcifies into structural decline — B2C nights stall, the margin program laps out, and the multiple stays at ~7–8× EV/EBITDA indefinitely (value trap). Falsification of the bear: a sustained re-rating toward ~10–12× EV/EBITDA on the back of holding growth + margin, plus the buyback visibly compounding EPS double-digits.
• AI begins measurably disintermediating OTAs, hitting Expedia first/hardest given its Google/metasearch dependence. Falsification: continued evidence (à la OpenAI walkback, Google route-through-OTAs) that OTAs are suppliers to, not victims of, the AI layer, with stable direct mix.
• Governance/management instability impairs execution — another CFO change, a strategy reversal, or a value-destructive use of the balance sheet. Falsification: CFO stability under Andersen and continued disciplined capital return.

The single cleanest swing factor: the trajectory of B2C room-night growth relative to Booking and Airbnb. If Expedia holds pace, the cheapness is opportunity and the buyback compounds; if it decelerates toward zero while peers grow, the cheapness is a value trap and no amount of margin self-help or share repurchase saves the thesis.

15. Source Appendix

See the separate EXPE_source_appendix.md (Appendix B of the combined report) for the full, categorized source list with URLs and access dates. Primary sources: Expedia Group FY2025 Form 10-K (filed 2026-02-13), Q1-2026 Form 10-Q (filed 2026-05-08), 2026 DEF 14A (filed 2026-04-29), 8-K filings (2025–2026), Q4-2025 and Q1-2026 earnings-call transcripts; quantitative data from ROIC.ai, AZI valuation index, and the FactorsToday factor model; peer context from published results and filings of peers Booking Holdings and Airbnb; third-party share data from Phocuswright and Guesty OTA compilations; recent news from Skift, Businesswire, Investing.com, PYMNTS, and Hospitality Net.

APPENDIX A — Standard Diligence Questionnaire

Expedia Group, Inc. (NASDAQ: EXPE) — as of June 14, 2026

Supplemental to the research memo. Fact / Interpretation / Assumption labels applied where it matters.

General

What thoughtful questions have other investors asked about this company?

• Is the 2025–2026 margin expansion structural or a finite cost program that laps out in 2H-2026? (The central earnings-durability question.)
• Is Expedia structurally losing share to Booking and Airbnb, or just growing a slower-but-stable book? (Room nights +6% vs +7%/+8%.)
• How exposed is the OTA model — and Expedia specifically — to agentic-AI disintermediation, given its heavier Google/metasearch dependence?
• Is the B2B (Rapid) franchise worth far more than the consolidated multiple implies, and could/should it be surfaced?
• Does the buyback compound per-share value, or is it being executed at progressively richer prices into a fading top line?
• How much should the Diller dual-class control structure and the three-CFOs-in-16-months churn discount the multiple?

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: mid-cycle and recovering, not peak. Margins are at a multi-year high (~24% adj. EBITDA, expanding) driven by internal cost actions, but volume growth (room nights +6%) is modest and the cycle is normalizing post-COVID, not booming. The lifetime max drawdown of −82.7% (2020) is a reminder of how cyclical the business is at the trough.

Driven by the external environment or internal actions? Both, but the earnings improvement is overwhelmingly internal (Gorin’s margin/platform program: marketing efficiency, headcount, platform consolidation). The revenue line is environment-driven (travel demand). Fact: B2C marketing dollars fell while bookings grew — an internal-efficiency achievement.

How stable are revenues? Moderately. No contractual recurring revenue, but ~2/3 direct/repeat demand and the merchant-float model provide behavioral stability. Highly exposed to discretionary travel spend and exogenous shocks (the March-2026 Mexico/Middle East air-pocket cost ~2 points in a quarter).

Outlook for products/services? Lodging (~80% of revenue) is the durable core; B2B/Rapid is the high-growth engine; advertising is a small high-margin grower; air is a deliberately low-margin traffic driver; trivago is structurally challenged.

How big will this market be? Growing. Global travel ~$1.7T, online subset ~$1.07T (+8%), online penetration migrating from mid-50s% upward. Domestic and international; Expedia is US-skewed (~59% of revenue) with international growing faster (+13%).

Business Quality & Competitive Moat

Is the industry getting more or less competitive? More, at the demand-acquisition layer (Google, AI, hotel-direct), and the supply-side moat is being eroded by EU regulation. Less concentrated entry risk at the supply-aggregation layer (a decade-long, low-ROIC build).

How profitable is the business? Fact: ROIC ~19.4% (2025, up from 10.6% in 2023) — well above cost of capital. Adj. EBITDA margin ~24%, gross margin ~90%. ROE (~112%) is meaningless (buyback-depleted equity). A clear tier below Booking (~37% margin).

How profitable is the industry — how many competitors, what barriers to entry? A global duopoly-plus (Booking, Expedia) + Airbnb (homes) + Trip.com (Asia) + Google/metasearch. High barriers at the supply-graph layer; low barriers (and a Google tax) at demand acquisition. Industry returns are high for the scaled winners.

Can the business be easily understood? Yes — an asset-light transaction toll on travel bookings, three segments, one dominant revenue line (lodging).

Can it be undermined by foreign low-cost labor? Not directly (digital marketplace). The relevant analog is AI labor/automation lowering costs (a tailwind Expedia is capturing) and potentially commoditizing the OTA layer (a threat).

Do brands matter? Yes, but Expedia’s are second-tier — Expedia/Hotels.com/Vrbo have awareness but thin switching costs; Vrbo trails Airbnb badly in whole-home brand. The brand is weaker than Airbnb’s and the direct-demand mix weaker than Booking’s.

What is the nature of competition? Demand-acquisition (Google ads, brand, loyalty) and supply breadth/price. Expedia competes hardest in the mature US market.

Customers’ switching costs? Low for consumers (multi-homing is the norm). Higher for B2B partners (API integration creates real stickiness) — the best moat asset.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The B2B partner relationships, the supply graph, the brands, and the One Key data asset are intangible and largely unrecognized (or carried at impaired historical cost). The merchant float is a real, off-the-income-statement funding source.

Off-balance-sheet liabilities? Standard operating leases and a large deferred-merchant-booking/loyalty liability (~$10.6B current, on balance sheet). No unusual hidden leverage identified. Recurring tax/legal exposures (Italian withholding tax, US occupancy tax) are partially reserved.

How conservative is the accounting? Moderate. Flags: serial “one-time” charges (restructuring, Italian tax) that recur and depress GAAP run-rate; meaningful gap between GAAP and adjusted EPS (SBC, restructuring, amortization); significant software capitalization. None egregious, but “adjusted” numbers do real work.

How CapEx-hungry? Low in absolute terms but not trivial — capitalized software development is a recurring real cost (~6% of revenue incl. software). Far lighter than the supply side (hotels), heavier than a pure marketplace.

Capital Allocation & Management

How much FCF, and how is it used? Fact: ~$3–4B FCF (headline TTM ~$4.1B; normalized owner-FCF ~$2.5–3.0B). Used predominantly for buybacks (~$1.6–2.0B/yr; 24% share-count reduction since 2022; fresh $5B authorization May-2026) and a dividend (reinstated Feb-2025, raised 20% to $0.48/qtr). Total shareholder yield ~7%.

Significant acquisitions recently? Only the small, sensible Tiqets (experiences, Dec-2025, B2B-tilted). No large/value-destructive M&A — a positive vs. the company’s own history.

Buying back shares? Yes, aggressively and consistently (the dominant capital-return lever). Caveat: average repurchase price has risen sharply ($106 → $212).

Issuing large amounts of new shares to insiders? SBC ~$398M (~2.7% of revenue, modest and declining), more than offset by buybacks (net share count falls).

Compensation policy of directors/management? Base + equity only (no cash bonus); PSUs vest on Revenue Growth + Adj. EBITDA Margin growth (50/50). Interpretation: misaligned — no relative-TSR, FCF, or ROIC metric; 2025 PSUs paid 176.9% in a year the stock fell.

Motivations of management? New CEO (Gorin) executing a credible margin/platform turnaround. Governance overhang: Barry Diller’s dual-class control (~32.6% vote on ~4.6% economics) and three CFOs in 16 months. No insider open-market buying signal found.

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? No — a US common-stock C-corp (NASDAQ: EXPE), dual-class (public Class A/common + Diller’s Class B). Standard 1099 dividend treatment.

Dividend policy? Reinstated Feb-2025 at $0.40/qtr; raised to $0.48/qtr Feb-2026 (~0.8–0.9% yield). Secondary to buybacks.

How profitable is the business? See above — ROIC ~19%, adj. EBITDA margin ~24%, solid but a tier below peers.

Is net income diverging from cash from operations? Yes, structurally — OCF (~$3.88B) far exceeds net income (~$1.29B), driven by the merchant/loyalty float, D&A, and SBC add-backs. Favorable (cash-rich) but the float component would reverse in a downturn; read FCF net of capitalized software.

Risks & Downside

What factors would cause the stock to decline? B2C room-night growth stalling toward zero (structural share loss); margin expansion lapping/reversing after 2H-2026; a macro/geopolitical travel shock; measurable AI demand diversion; a value-trap de-rating; further governance/CFO instability.

Risk of a catastrophic loss? Low near-to-medium term: net debt only ~$1B, investment-grade, FCF-generative, no refinancing wall. The realistic downside is multiple compression + growth disappointment (a value trap, ~20–30% drawdown), not a zero.

Chance of a total loss? Negligible on a multi-year horizon barring a structural, prolonged collapse of the OTA model (agentic AI fully disintermediating travel — slow-moving and shared with peers) combined with a severe travel depression.

Recent News & Events

Has the business environment changed recently? Yes: (1) AI/agentic commerce is reshaping the demand layer (OpenAI checkout walkback Mar-2026 eased near-term fears, +12% stock reaction); (2) a March-2026 demand air-pocket (Mexico advisories, Middle East conflict) then April recovery; (3) management cost/platform turnaround driving margin expansion; (4) capital-return reset (fresh $5B buyback, dividend raised).

Significant acquisitions? Tiqets (experiences, Dec-2025).

Change in accounting policies? No material change identified; reorganized segment reporting into B2C/B2B/trivago.

Recent changes — new markets, facilities, management? New CEO (Gorin, May-2024); third CFO in 16 months (Andersen, ex-Snap, May-2026); One Key loyalty global rollout; new exclusive B2B partners (Uber, Bank of Montreal); platform consolidation and restructuring.

APPENDIX B — Source Appendix

Expedia Group, Inc. (NASDAQ: EXPE) — as of June 14, 2026

Primary sources first. Quantitative figures reconciled to filings; third-party aggregators (ROIC.ai, AZI, FactorsToday, Guesty) used for computed ratios, factor data, and cross-checks. All accessed June 2026.

1. Primary — SEC filings (EDGAR, CIK 0001324424)

• Form 10-K, FY2025 (filed 2026-02-13; expe-20251231.htm) — segment revenue/EBITDA (B2C/B2B/trivago), gross bookings, take rates, revenue by service type, geographic mix, debt schedule, risk factors, one-time items. https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001324424&type=10-K
• Form 10-Q, Q1-2026 (filed 2026-05-08; expe-20260331.htm) — Q1 segment results, room nights/ADR, buyback activity, debt changes.
• DEF 14A / proxy, 2026 (filed 2026-04-29) — executive compensation (PSU metrics, payouts), beneficial ownership, dual-class structure, board composition.
• Form 8-K filings (2025–2026) — CFO transition (Derek Andersen, 2026-04-23); $1B 5.5% notes due 2036 (2026-04-10); $2.5B revolver (2026-03-30); Q4-2025 and Q1-2026 earnings (2026-02-12, 2026-05-07); Feb-2025 $1B 5.4% notes; annual-meeting results.
• Q1-2026 earnings release (8-K, Item 2.02) — https://www.sec.gov/Archives/edgar/data/0001324424/000132442426000031/earningsrelease-q12026.htm
• Prior 10-Ks (FY2021–FY2024) and 10-Qs — multi-year trend reconciliation.
• Open item: Form 3/4/5 (insider) filings — listed in the EDGAR index but not machine-read for this report; transaction codes (P vs S/A/M/F) require a direct EDGAR pull.

2. Primary — Earnings-call transcripts (via ROIC.ai MCP)

• Q1-2026 earnings call (2026-05-07) — CEO Ariane Gorin, CFO Scott Schenkel. Guidance, demand commentary, B2B, AI/agentic strategy, capital allocation.
• Q4/FY-2025 earnings call (2026-02-12) — full-year results, FY2026 guidance, dividend increase, AI framing.

3. Quantitative data sources (third-party; reconciled to filings)

• ROIC.ai MCP — income statement, balance sheet, cash flow, profitability ratios (ROIC/ROE/margins), enterprise value, valuation multiples (FY2020–FY2025). Third-party aggregated; EDGAR primary where they differ.
• AZI valuation index (scripts/azi.sh fundamentals EXPE) — own-history percentile ranks: P/E 9.56th pct, P/S 61st pct, P/B 99.5th pct (ignored — negative tangible book), composite 56.7th; TTM EPS $11.35; price $224.89 (2026-06-12).
• FactorsToday factor model (factorstoday.com/api) — /stock-loadings/EXPE (Momentum −0.49, Value +0.12, SmallSize +0.49, Beta +1.21), /leaderboard/EXPE (y1 +32%, m6 −18% (−35.5% ann.), max drawdown −82.7%), /stock-info/EXPE (beta 1.34, rs_6m −17.6), /related-stocks/EXPE (BKNG, RCL, NCLH, cruise/travel cluster).

4. Peer / industry context (public peer data & industry sources)

• Booking Holdings (BKNG) — public filings / peer comps: $186B GBV, 14.5% take rate, ~37% EBITDA margin, ~13× EV/EBITDA, ~6.5% shareholder yield.
• Airbnb (ABNB) — public filings / peer comps: $91.3B GBV, 13.4% take rate, ~35% adj. EBITDA / ~38% FCF margin, ~16× EV/adj-EBITDA, ~33× P/E.
• Phocuswright — global/online travel market sizing (~$1.7T total, ~$1.07T online, +8%).
• Guesty OTA quarterly insights — Q1-2026 room-night comparison (Booking.com 319M +7%, Airbnb 143.1M +8%, Expedia 107.7M +6%); FY2026 growth-guide comparison. https://www.guesty.com/blog/ota-quarterly-insights/

5. News & qualitative (publisher-attributed)

• Investing.com (2026-05-07) — “Expedia Q1 2026 slides: 83% EBITDA surge drives earnings beat.” https://www.investing.com/news/company-news/expedia-q1-2026-slides-83-ebitda-surge-drives-earnings-beat-93CH-4670788
• Businesswire (2026-04-23) — “Expedia Group Appoints Derek Andersen as Chief Financial Officer.” https://www.businesswire.com/news/home/20260423837765/en/
• Skift (2026-03-20) — OTAs, AI discovery & transactions (OpenAI checkout walkback). https://skift.com/2026/03/20/otas-ai-discovery-transactions/
• Skift (2025-12-10) — “Expedia agrees to buy experiences platform Tiqets.” https://skift.com/2025/12/10/expedia-agrees-to-buy-experiences-platform-tiqets-to-bolster-its-b2b-offering/
• PYMNTS (2026) — “Expedia embraces agentic commerce to fight platform disruption.” https://www.pymnts.com/travel-payments/2026/expedia-embraces-agentic-commerce-to-fight-platform-disruption/
• Hospitality Net — “Nobody gets bypassed: agentic commerce… Google’s partner list routes straight through [the OTAs].” https://www.hospitalitynet.org/opinion/4132643/
• AZI news feed (scripts/azi.sh news EXPE) — BTIG reiterated Buy, $330 PT (2026-06-09); Glenview exited EXPE in Q1-2026 (13F).

6. Analytical frameworks

• Competition Demystified (Greenwald & Kahn) — moat taxonomy (economies of scale + customer captivity), ROIC/market-share tests, EPV.
• Capital Returns (Marathon Asset Management) — supply-side capital-cycle analysis applied to the OTA demand-acquisition vs. supply-aggregation layers.

Note: ROIC.ai, AZI, FactorsToday, and Guesty are third-party aggregated data, not primary. For all material figures, the SEC filings are authoritative and were used to reconcile. Management commentary from transcripts is treated as hypothesis, validated against filings.