In the last year of doing this newsletter, Daniel and I have developed a fondness for owning the “best of the best” — companies that are undeniably beacons of quality in the niches they dominate. Sometimes, while we appreciate that you must pay a premium for quality, we turn our heads away over valuation concerns.

There’s no exact science, though, to tell you if you’re paying a reasonable price for a great company, or overpaying.

Today’s pick, Booking Holdings, is another such quality compounder that, unsurprisingly, doesn’t necessarily trade at a bargain price tag. Yet, is it still attractive enough to earn a spot in our Intrinsic Value Portfolio of stocks?

Let’s find out.

— Shawn

Booking Holdings: The World’s Biggest Travel Company

Why I’m Presenting Booking Holdings to Daniel (and You, the Reader)

As an investor, I like to use my experiences as a customer to inform my investment opinions — companies like Reddit and Airbnb, holdings in our Intrinsic Value Portfolio, first came across our radar because of how positively these two platforms had impacted my life personally, compelling us to explore the investing thesis as well.

Just like how we don’t pay up for every top-notch business, we don’t invest in every company that we love as customers, either, like Amazon and Apple, but the idea is to embrace Peter Lynch’s wisdom and keep our eyes open to the most impactful businesses in our own lives as investment inspirations. That way, not only will we be positioned to better understand our portfolio companies, but we’ll have opinions based on our own experiences, indicating whether said companies are moving in the wrong direction.

For example, if I found myself no longer checking Airbnb first for the trips I plan, then that would be a signal to me that I’m likely not alone in that feeling, and if so, the investment thesis may be deteriorating.

It’s hard to pick up on those same intuitive signals, though, when you’re just an investor reading about a business you don’t otherwise interact with, or trying to absorb scuttlebutt insights from others. As much as we’re all susceptible to biases, having your own consumer opinion can be important in investing.

After recently using Booking.com for the first time and with Daniel helping me appreciate just how much more popular it is in Europe than in the U.S., I began thinking about how this would be an interesting company to cover. Not only would it help us understand the competitive threats against our largest portfolio holding, Airbnb, but Booking Holdings’ track record has been impressive enough to attract legendary investors like Francois Rochon, so why not learn more?

Look at these numbers: 11% revenue CAGR over the last decade, 33.3% operating margins (the same as Alphabet), nearly 30% returns on invested capital over the last five years, billions of dollars in annual share repurchases and buybacks, and a forward P/E ratio of 22x, which I find to be temptingly reasonable for a company of the described-above financial caliber.

Betting on Global Travel

Conveniently, I’m also drawn to businesses riding structural tailwinds, and travel is one of those forces. When people get wealthier, decades of data show us that they travel more and pay up for better experiences, too. Meaning, over long arcs, travel has tended to outgrow GDP, as larger swathes of the global population have hit the threshold for being able to periodically afford air travel.

Booking Holding (the parent company behind Booking.com, Kayak, OpenTable, and Agoda) has built a reputation as a neutral travel marketplace, connecting would-be travelers with available hotel rooms or flights and earning commissions for helping fill those vacancies.

That value proposition remains intact, as specific airlines and hotel brands typically lack the brand power to singularly sell all their inventory alone, leading them to lean on demand aggregators like Booking, which can connect them with a wider audience of potential customers. And customers, naturally, appreciate having the best prices for every reasonable accommodation available across the internet in one easily searchable archive.

But Booking is now expanding its surface area. Management often talks about a vision they call the “connected trip,” where one’s lodging, flights, rental car, restaurant plans, and trip insurance are seamlessly linked together, adjusting intelligently in real time to changes in your plans.

If you miss a connecting flight, imagine having a digital travel agent faithfully rebook your hotel, reschedule your rental car pickup, and cancel your restaurant reservations as needed. Travel might still be a headache, but boy, that sounds a lot less painful than sorting out these issues yourself. I say that as someone who once spent a total of nearly 12 hours on the phone waiting to speak with “customer support” — two words that serve as a baffling mischaracterization of my actual experience in not getting any support — after my airline declared bankruptcy days ahead of a planned trip to Europe, leaving me to rebook flights at the last minute while hoping that I’d somehow be reimbursed.

When the airline goes bankrupt, though, it’s hard to feel confident that the funds exist to, in fact, refund your ticket. Anyways, we all have our own travel horror stories, so I think it’s self-evident why using AI to sync up one’s travel plans is a compelling idea.

What Booking Does

But before we get swept away by bullish AI fantasies, let’s better understand the company behind today’s stock pitch. Booking is the modern incarnation of a story that began with Priceline’s late-90s “Name Your Own Price” concept, where folks would blindly bid on how much they’d be willing to pay for a trip to, say, Paris without exactly knowing their accommodations or flight schedule. However, that idea can only take you so far (it’s more of a novelty than anything), and so, with the 2004 decision to acquire Booking.com, the company became dramatically reoriented around being an online travel agent (OTA), which it’s now redefining in the era of AI, as mentioned.

Today, Booking.com represents 90% of the group’s sales, underpinning the parent company’s rebranding from Priceline (ticker: PCLN) to Booking Holdings (ticker: BKNG). The acquisition of Booking.com, which literally changed the company’s identity, is arguably the most successful acquisition in corporate history, enabling a middling small-cap travel stock to grow into a travel giant with $170 billion of market capitalization.

Powering this growth are two distinct monetization models as an OTA: agency revenues and merchant revenues.

The agency model defined Booking’s early years (and still accounts for a third of revenues), in which Booking matches customer demand with excess hotel room supply, but leaves the traveler to complete payment in-person at the hotel during check-in, when Booking is then paid a commission by the hotel. The result is a very extended cash conversion cycle (not ideal!)

The merchant model flips the payment timing. Booking becomes the merchant of record, collecting payment for your booking upfront while keeping a fee for themselves before remitting payment to the hotel partner. The latter model has been gaining share for years (rising from a third of revenues to two-thirds over the last 12 years), and it has enhanced Booking’s business quality.

Why? The merchant model improves working capital & free cash flow (payments are received earlier), while also enabling Booking to realize its “connected flight” vision by bundling flights + stay + car + insurance into a single, standardized checkout.

For Booking, flights are a relatively new area they’ve gained traction in, but they’re strategically important, even if coming at lower margins, because flights open the door to the rest of the trip bundle. Once a traveler books a ticket, the lodging decision is an inevitability.

Geographically, Booking remains strongest in Europe, where hotel supply is fragmented, dominated by smaller boutique hotels than large corporate chains. This leaves more room for Booking to add value for independent operators who rely more on Booking’s help in attracting customers and managing the checkout process.

In terms of competition, Expedia is stronger in the U.S. but is a less profitable business because the U.S. is dominated by corporate chains that rely on Expedia more narrowly, primarily just for help with accessing more eyeballs, leaving Expedia with a lower take rate and profit margins due to a diminished value-add.

Booking, as such, is only 50% larger than Expedia in gross bookings but generates 6x more operating profit(!)

And then Airbnb dominates mindshare for alternative accommodations, which is defined as basically any stay somewhere besides a traditional hotel. Within this map, Booking’s differentiation is less about a single killer feature and more about cumulative iteration, testing how to best drive conversion (they’ve done millions of A/B tests!) and building the infrastructure that smaller hotel partners rely on but don’t want to build themselves.

The Google Problem

For so many of the companies we’ve covered, Alphabet is a thorny competitor in different ways, whether it’s with Reddit, Adobe, The Trade Desk, or Uber. Alphabet is, once again, a complicating factor here.

Nearly every marketplace that depends on discovery has to wrestle with Google and its monopoly on search results (which is why we’re happy to also own shares in Alphabet). It doesn’t help, either, that Google has taken a keen interest in travel — for years, they’ve opted to take a more passive role, though, choosing not to act like a merchant or agency and instead simply showcasing flight/hotel data that leads customers to book via a site like Booking.com or directly at the airline’s/hotel’s site.

This makes Google’s travel portal more like a metasearch travel site, similar to Kayak, which Booking owns but is a small part of its overall business, rather than truly stepping on Booking.com’s core business. If anything, Google’s travel portal often funnels people to Booking to complete the transaction.

While the Google risk has been flagged for years, I do think that, with the rise of LLMs and AI agents, Google may find it more interesting to directly compete with Booking than it has historically, leveraging Gemini to do so.

Painpoints can also come from changes in SEO dynamics, auctions in paid search, or, as mentioned, Google’s prioritization of its own metasearch-based travel tools, leaving intermediaries like Booking paying a higher toll to access the same potential customers. Booking is, ultimately, playing an arbitrage game, paying one price for ads to attract would-be travelers, and then hopefully collecting a bigger commission from suppliers for those customers they can successfully convert.

For context, only about half of Booking’s customer traffic comes “directly,” meaning around half the business is primarily sourced via the paid marketing arbitrage described.

The challenge is that, if even less traffic arrives “for free” through the company’s brand and direct channels, and more traffic arrives through increasingly expensive paid funnels like Google, profit margins could compress as paid marketing becomes something of a treadmill.

Booking’s answer has been to push toward more direct/app visitors driven by generous loyalty programs. That is to say, getting more people to go straight to the Booking app when they’re first planning a trip, rather than typing into Google Search, where Booking may have to pay to sponsor results to rank above competitors in the search results.

Once a traveler downloads and uses the app with payment credentials saved and notifications enabled, the marginal cost to bring them back for the next trip booking falls dramatically. Even better, the app can learn to personalize the next trip from a standing start, making the direct user experience stickier.

This also shows why Booking has been keen to get its hands dirtier by acting as a merchant instead of following the more passive agency model that once defined the business. In a world where Google shapes how someone starts a search, Booking can still expand its slice of value by controlling checkout, fulfillment, and fixes when plans shift.

That doesn’t eliminate the risk of Google effectively raising its customer acquisition costs or, worse, more directly competing with Booking, but the long-term game is to ensure that a higher share of demand originates on owned terrain.

They want a larger share of revenue to come from merchant payments while post-booking attach rises, too (think: tacking on rental cars, paid excursions, etc. to your hotel & flight booking), as the Booking app becomes the default place to both plan and repair trips.

If Booking can keep shifting gravity from the open web into its own orbit, the platform can absorb volatility at the top of the funnel (Google Search) without surrendering the bottom of the funnel, where the most valuable economics lie.

Non-Google Risks

Booking has other challenges, too. At the same time that Booking is trying to catch more users in its ecosystem, hotels and airlines want to do the same. Don’t get me wrong, they’ve benefited tremendously from their arrangements with Booking over the years, but they’re also constantly trying to convince travelers to come directly to their sites/apps for future trips with special credit cards, loyalty points, and other promos.

I should mention, though, that hotels & airlines can’t directly undercut Booking in most countries. With the exception of a few countries that have banned the practice, Booking has “price parity” legal protections with supply partners, banning them from offering one price on Booking (or any other OTA) and then offering a lower price for customers who go straight to, say, Hilton or Marriott to book.

Booking can also recognize fairly quickly when this is happening and punish those suppliers by de-prioritizing them in their own search results, so while corporate hotels would love to have customers coming straight to them and save on the costs of a middleman like Booking, there are limits on the extent to which they can encourage this.

Of course, there’s the ongoing pressure from both Expedia and, to a lesser extent, Airbnb, which are both fighting for customers’ eyeballs and wallets when planning their next trip. Expedia’s dominance over hotels in North America and Airbnb’s dominance over alternative accommodations don’t leave much meat on the bone there for Booking, but with their previous acquisition of Agoda, they’ve been able to gain significant market share in Asia, expanding their growth runway.

The bigger question mark to me, and this applies to most of the companies we’ve looked at in the last year, is to what extent OpenAI or another LLM will rewrite the consumer paradigm. Initially, people thought OpenAI, with ChatGPT, was mainly looking to disrupt Google Search, but it’s clear that the company’s ambitions are even bigger than that — they seem to be trying to disrupt the entire desktop.

As in, fundamentally changing how we transact over the internet, hoping to move much of that commerce off of websites and onto ChatGPT’s platform directly. Imagine, for example, consulting with ChatGPT for help building a personalized trip (as many folks already do), and then asking it in “agent mode”, on your behalf, to go ahead and book everything necessary for the trip, from flights to hotels and rental cars. In this not-so-distant-sounding hypothetical, ChatGPT’s AI models might go directly to the hotels & airlines to book these trips, cutting out Booking entirely.

If humans are no longer the ones making travel decisions, or even if just some percentage of travel is outsourced to AI assistants (let’s say 5-10%), that seems like a major headwind to booking that obseletes their business model. Again, you can appreciate why Booking is racing to lead in the “connected trip” area, pulling customers directly onto its platform to access AI assistants who can help with trip planning.

I would typically be inclined to say that Booking, as the company with the most travel data in the world, would be uniquely well-positioned to train the world’s best AI travel assistants, but things are changing incredibly quickly, so it’s important that we be honest about what we can and can’t know. This risk feels more abstract, but I can very tangibly envision myself jumping onto ChatGPT, brainstorming with it on a trip, and then, if I felt it could get the best prices for me, trusting a ChatGPT agent to then make the bookings on my behalf.

ChatGPT already has a “Booking” integration!

Maybe the masses won’t be frequently using this tomorrow, but within the next 5 years? It’s plausible. That idea really scares me, and Daniel has similar hesitations. As you’ll see in the above, the first result for me was to “Book (directly) with United Airlines,” bypassing Booking.

And yes, to me, it’s a bit different with our portfolio holding, Airbnb. Many of the hosts on Airbnb only list their properties on Airbnb, because they’re small-time operators and Airbnb is the best platform for them to use (so why bother listing elsewhere?), which means that, simply put, your AI travel assistant would probably still be going through Airbnb to book more unique accomodations and experiences whereas a routine trip to the midtown Hilton in NYC could easily be booked directly on Hilton’s site by an AI agent without going through Booking.

Valuation & Portfolio Decision

I think that’s all the abstract AI risk pondering I can take for now, but even for companies we like, it’s important to be intellectually honest about what worst-case worries we have. I’m not by any means saying we’ll all outsource trip planning to ChatGPT, but I also wouldn’t be surprised if a meaningful percentage of the population did…

Anyway, let’s try to think about how to value Booking. The market has long recognized Booking’s quality, and rightfully so. Over almost any time frame, the stock has consistently traded at a premium, in P/E terms, to the broader S&P 500.

Since 2020, they’ve repurchased over a fifth of their total share count, while launching a dividend for the first time in 2024:

We love a good share cannibal! There are two ways to grow earnings per share, and shrinking the denominator is an equally valid way to do so.

And while Booking, unsurprisingly, faced setbacks from Covid-era travel restrictions, operating profit margins have recovered again above 30%:

The company has announced plans to cut costs by several hundred million dollars in the coming years, as AI helps them trim customer support jobs, for example, which should enable margins to continue expanding modestly as the business’s revenues continue to grow, too. Wall Street certainly expects as much, and given the runway they have to execute on the “connected trip” vision, while travel generally outpaces GDP growth, short of a black-swan disruption from ChatGPT, it’s hard to imagine this company not continuing to grow its top line at a decent clip for years to come.

Expanding margins, sales growth, buybacks, and dividends are all, well, a good formula for shareholder returns, so in theory, Booking checks all the boxes from a modeling perspective and should generate an IRR through 2030 that clears our 12% annual hurdle rate even with an assumed contraction in the valuation multiple as the business matures (from a market premium to a slight discount to the S&P’s current. multiple).

So, are Daniel and I investing in Booking?

Well, competition is the source of nearly all trouble for investors, and we’re a bit torn thinking about the competitive pressures on Booking. On the one hand, the numbers say they have a moat, but on the other, their positioning feels…a little flimsy. Google can put pressure on its customer acquisition costs or outright try to displace Booking altogether, while Airbnb and Expedia continue to be formidable competitors, seeking to attract a larger share of travel spending to their own ecosystems.

Meanwhile, airlines and hotels have their own incentives to drive customers to bypass Booking, and that’s to say nothing of the more fundamental risk posed by AI agents that we already discussed. At a 20% premium to an already historically rich S&P 500, it’s hard to feel as though we’re getting much of a margin of safety at current prices, even if, mostly, we think the company will likely continue to be a very good investment.

On top of that, we’ve learned to look carefully at how the people who run a company are incentivized, and with Booking, that incentive structure is, unfortunately, not very well aligned with shareholders, in our opinion.

Booking’s long-term equity awards for management are split between roughly 60% performance stock units (PSUs) and 40% restricted stock units (RSUs), with the RSUs simply vesting over a period of time, rather than being linked to performance.

To us, this is the executive equivalent of a participation award, where management is paid simply for not going anywhere! This comes at very real, direct, and indirect costs to the owners of the underlying business.

Worse, the performance side is also misaligned, granting awards based on three metrics, including revenue growth and adjusted EBITDA targets. We recommend reading The Outsiders by William Thorndike for anyone who doesn’t yet understand why that’s problematic. Long story short, prioritizing revenue growth doesn’t necessarily mean the growth is profitable. It’s possible to grow revenues while actually shrinking profits.

The same is true for adjusted EBITDA, which isn’t even a standardized metric — it’s an accounting fantasy that companies all define differently (that use it), and we find it to be unsatisfactory for similar reasons.

While the third metric the board considers is based on shareholder returns (good!), this focus is further diluted by incorporating Booking’s relative performance against a peer group, instead of the objective total return delivered. The problem with the relative comparisons of Booking’s stock versus other similar companies is that it’s up to the board to define the “similar” companies in its peer group, which they’ve watered down with very weak competitors. Instead of just using, say, Expedia and Airbnb as a peer group, they also factor in the performance of cruise lines, hotels, and airlines, which are fundamentally different businesses, much more capital-intensive, and generally have much worse track records of creating shareholder value.

In other words, the relative comps management is graded against are layups —they’re easy.

Beyond the troubling implications this has for what management will implicitly choose to focus on, to us, it indicates a weak board that’s unlikely to keep the CEO in check and advocate for shareholders, which is even more troubling.

If there were only competitive risks to consider, we might be able to justify buying Booking, but these executive comp issues push us over the line into passing on the company, for now. I’d need to feel more comfortable with the durability of Booking’s competitive advantages to purchase at current prices.

For those curious, you can learn more about Booking in our podcast here, access our base case valuation model here, and as always, updates on the rest of our Intrinsic Value Portfolio are below.

Weekly Update: The Intrinsic Value Portfolio

Notes

Daniel and I had the chance to do our monthly portfolio review call together this past week, and we made some changes:

• Position Exit From Ulta: We decided to sell out of our position in Ulta entirely at around $530 per share. It’s hard not to be a bit nostalgic because Ulta was the first stock we ever added to our Portfolio, but we felt the valuation had reached a point where, relative to other opportunities, we couldn’t justify continuing to hold onto it. Ironically, we still love Ulta as a business, but at north of $500 per share, the risk/reward is much less attractive than it was when we began adding to the position below $400 per share — for those keeping track at home, that’s a more than 25% increase. We expected those kinds of returns, but over a longer period, honestly, so we felt as though the stock had begun to materially “pull forward” future returns. Ulta is now the #1-ranked stock on our watch list (which you can see here), and if we can snap up shares again closer to $400, we’d be happy to add it back to the Portfolio.
  

• Increased Uber to 7%: Uber is the perfect illustration of how we think about sizing investments. Initially, Uber was just a 2% stake for us, but as we have gotten more comfortable with the thesis, we have scaled the position from 2% —> 4%+ —> 7%, including some unrealized gains. Daniel and I have come to better understand the bear case arguments, from Amazon potentially acquiring Lyft to Waymo possibly undercutting Uber in core cities, providing a clearer view of the full picture that actually makes us ~more optimistic~ about the company, and thus, we’re using this larger position sizing to reflect that conviction. You can find our previous write-up on Uber here.
  

• Increased Adobe to 8%: Adobe has been the biggest headwind to our Portfolio’s returns, but we don’t see the decision to invest in the company as a mistake. As shares have fallen toward $320 and lower, we see a buying opportunity that keeps getting more attractive. As such, we’ve continued to dollar-cost average and raise our position from 6.8% to 8%. Adobe made some headlines yesterday with its $1.9 billion acquisition of Semrush.
  

• Initiated a New 2.5% Position in Crocs: We first covered Crocs back in August, and well, clearly, we didn’t forget about it! Yes, we don’t always add companies to the Portfolio when we first cover them, but we do mean it when we say we’ll continue to keep an eye on them. In this case, with Crocs ($CROX) offering a nearly 20% free cash flow yield, we couldn’t help ourselves. The shoes are ugly, but the valuation isn’t. We believe the stock is being excessively punished for Crocs’s erroneous acquisition of the HeyDude brand, which has been in decline for two years now or so. The valuation has now reached a point, though, where, even if you assumed HeyDude would nearly go to zero (which is the opposite of what management has guided for), then an attractive return could still likely be earned going forward simply from the durability of the main Crocs brand. Wanting to be mindful of our overall exposure to retail brands, with Lululemon and Nike already in the Portfolio, and due to some concerns still about just how durable the Crocs brand is, we decided to make this a smaller holding for the time being. If our conviction in it builds, as happened with Uber, we’ll add to the position. If not, we’ll keep it untouched or, perhaps, exit the position as needed. Much of the lower weighting is due to the investment being disproportionately riskier than our other core holdings, which have received 5-12% weightings, including companies like Adobe, Alphabet, Airbnb, and Berkshire Hathaway.

Quote of the Day

"This time, like all times, is a very good one, if we know what to do with it.”

— Ralph Waldo Emerson

What Else We’re Into

📺 WATCH: A recent interview with Joel Greenblatt on whether you can really beat the market with simple investing strategies

🎧 LISTEN: How Home Depot’s founders built a $300 billion company from the ground up

📖 READ: Berkshire Hathaway makes a nearly $5 billion bet on Alphabet ($GOOGL)

You can also read our archive of past Intrinsic Value breakdowns, in case you’ve missed any, here — we’ve covered companies ranging from Alphabet to Airbnb, AutoZone, Nintendo, John Deere, Coupang, and more!

Your Thoughts

See you next time!

Enjoy reading this newsletter? Forward it to a friend.

Was this newsletter forwarded to you? Sign up here.

Join the waitlist for our Intrinsic Value Community of investors

Shawn & Daniel use Fiscal.ai for every company they research — use their referral link to get started with a 15% discount!

Use the promo code STOCKS15 at checkout for 15% off our popular course “How To Get Started With Stocks.”

Read our full archive of Intrinsic Value Breakdowns here

Keep an eye on your inbox for our newsletters on Sundays. If you have any feedback for us, simply respond to this email or message [email protected].