I am loading up on Trivago - the hotel metasearch business. Here is the data and thesis.

REDDIT.COMOct 26, 2:03 PM UTC

Key insights

  • Trivago (TRVG) is showing signs of a turnaround with six consecutive quarters of double-digit revenue growth and a raised FY2026 Adjusted EBITDA guidance to ~€30M. The company's market structure is improving as major OTAs Booking (BKNG) and Expedia (EXPE) decrease their share in the metasearch auction, while Trivago's own partners increase. This shift, coupled with widening price dispersion in the market, is seen as a positive development for Trivago's business model. The company is trading at an attractive valuation relative to its forward EBITDA.
I am loading up on Trivago - the hotel metasearch business. Here is the data and thesis.

TLDR: Debt-free, ~$33M of 2026 Adjusted EBITDA guided, six straight quarters of double-digit revenue growth, ~$299M market cap (~$162M enterprise value). Net out the ~$137M of cash — under 5x forward Adjusted EBITDA — for the operating business.

The scar tissue. €1.03B revenue in 2017 at a $7B EV. Booking + Expedia — then 80% of the auction — coordinated a metasearch pullback and squeezed the toll booth. Revenue fell every year to 2023. It's been reversing since.

The numbers. Q2 2026 revenue +21% YoY to €168.4M; first positive Q2 Adjusted EBITDA since 2023. FY2026 guidance raised to ~€30M Adjusted EBITDA from €15.8M in 2025 and €20M initial guide. Target: 10% Adjusted EBITDA margin by 2028. Headcount locked at ~600. Conversion up 58–64% since 2023. Logged-in members >30% of referral revenue. GenAI traffic: <1% of revenue. Google-referred revenue fell 34% since 2023 and the top line still grew.

Healthier marketplace. Booking + Expedia went from 80% of the auction to 65%. "All Others" went 20% → 35%. 140+ partners on CPA, >30% of referral revenue. More bidders, tighter auction. They also bought Holisto for $34.3M — staged, ~90% below its 2022 SPAC ask. That bought native merchant checkout and wholesale bedbank stitching. Holisto powers Book & Go (native OTA): +609% YoY, 16 markets.

Why the market structure is turning in its favor. Booking's merchant model went from ~19% of volume in 2016 to ~72% today, heading to 85–90% by 2031. When the OTA is merchant of record it controls the checkout price — and it can rebate commission, FX spread, and FinTech margin per channel, per device, per geo. Same-room price spread widened from 2–5% in 2016 to 8–22% today, and it's going to 12–30%. 75% of global searches already show an OTA undercutting the hotel's own site. Price dispersion is the product. The more chaotic the market, the more valuable the toll booth.

Why agents don't kill this

  • Hotels can't just sell cheap direct. Channel contracts, corporate rate obligations, and rate-parity policing cap the public discount ceiling at ~11.7% before it becomes value-destructive — and undercutting an OTA gets you demoted algorithmically within days. So hotels dump inventory through wholesale bedbanks and gated channels: logged-in OTA rates, opaque wholesale, Book & Go. Those rates are invisible to a public scraper by design. An agent browsing public URLs sees the sanitized price. * 7.0M+ properties in Trivago's database. ~1.2–1.5M traditional hotels plus 3.8M+ alternative units, deduplicated across Booking, Agoda, Trip.com, Goibibo, Airbnb, Vrbo, and PMS feeds. No agent is normalizing that. Supplier consolidation is valuable — standardized, deduplicated, real-time inventory across 190+ countries.

Bear case. TV saturation in 30 markets caps ROAS expansion. Agentic booking is the real 3–5 year watch item, could be disruptive or just an additional channel on top of existing ones.

Continue reading on REDDIT.COM

Related Articles