It has been a busy few months. We've spent the last year building the Intrinsic Value Portfolio from scratch — covering company after company, deciding which ones deserved a spot and which ones, however good the business, didn't clear our bar. Along the way, our portfolio has grown to 17 positions, and I'd say we're largely happy with how it's shaped up. But there are some companies that we added under the premise of testing our conviction in them over time.

Two of those companies will be leaving our portfolio today. Another one will be promoted to a core holding. We'll also revisit a good bunch of our other portfolio companies and a few watchlist names that a lot of you have been asking about — Trade Desk and FICO, in particular. We’ll also cover a couple of companies that Kyle, our new co-host(!), is getting up to speed on.

Yes, that's the other big news. Kyle Grieve is officially joining us going forward. Many of you will know him from We Study Billionaires. For us, that means even more stock pitches. Excited to have him along!

If you want the full background and even more information than I will be able to provide in this newsletter without turning this into half a book, you should check out the podcast episode with the three of us.

— Daniel

Berkshire Weekend is coming up, so Shawn and I wanted to send a quick reminder that we’ll be hosting The Intrinsic Value Conference on Friday, May 1st, at the Hotel Indigo. The conference will start at 1:00 pm local time.

We’ll be pitching stocks, walking through our investment philosophy and current portfolio, and we’ll also leave plenty of time for live Q&A.

Members of The Intrinsic Value Community have already reserved spots, but there will be a limited number of additional seats available on a first-come, first-served basis.

If you’ve been considering applying to the community, this could be a great opportunity. We’re also hosting a private Member Dinner on Saturday evening, which is a more personal setting to connect the day after the conference.

Hope to see you there!

Shawn actually named Exor as his stock pick for 2026 on The Investor’s Podcast in January, so this felt like a natural one to walk through with our new co-host, Kyle, on the show. At 7% of the portfolio, it's one of our larger positions, and it's had a bit of a rough start to the year. That’s not entirely surprising, given that Ferrari, Exor's crown jewel, is down around 13% year to date.

The simplest way to understand Exor is as a holding company. The Agnelli family has run it for over a century. They hold stakes across a portfolio of businesses, and the most important one by a wide margin is Ferrari, where Exor controls around 20% of shares outstanding and roughly 30% of voting rights.

What makes this interesting from a valuation perspective is the discount. Exor's net asset value, the total value of everything it owns, net of debt, is around €33 billion. Exor's market cap is roughly €14 billion. That means you're buying the underlying assets at a ~60% discount to their market value. The Ferrari stake alone is worth nearly the entire Exor market cap. Everything else — CNH, Stellantis, Philips, stakes in the football club Juventus, the luxury fashion retailer Christian Louboutin, or the media company The Economist — you get for essentially free.

Now, I have to say that holding company discounts are completely normal and somewhat rational. You're forced to hold assets you may not want, and it's difficult and costly to exit large concentrated positions. But 60% feels quite extreme, especially when the primary asset is one of the most durable luxury businesses in the world.

Historically, the discount has traded closer to 20–30%, so mean reversion alone could generate very attractive returns. At the same time, Exor has been buying back shares aggressively, including a one-billion-euro reverse Dutch auction, which mechanically forces the discount to close. When you're buying back shares at a 60% discount to NAV, every euro spent creates significant value for remaining shareholders.

The big question is whether you trust John Elkann to allocate the growing cash pile well. We would argue that his track record is better than the market gives him credit for. Exor's NAV compounded at roughly 16% annually from 2009 to 2025, versus 11% for the MSCI World. Even if you would adjust for Ferrari, the return would still look decent. And I have my problems with just removing the most successful investment from a track record. As Buffett famously said, he made most of his wealth from just a handful of stocks. Successful investors buy these assets and, most importantly, hold them without getting scared out of them. That’s something Elkann did successfully.

And even the timing of the one reduction in Ferrari looks pretty smart in hindsight. When Elkann sold some Ferrari shares, that was close to the highs, from which we are now almost 40% down.

Still, the cash is a source of uncertainty until they tell us what they're going to do with it. Once that decision is made, we’d hope a lot of the NAV discount resolves itself. We're comfortable holding.

Reddit has been one of the biggest surprises in the portfolio since we first bought it at around $80 per share. The business has continued to outperform in almost every dimension, and the numbers from the most recent earnings were, once again, phenomenal.

If we zoom out a bit, net income margins went from -37% in 2024 to 24% in 2025. In Q4 alone, margins hit 34%. Revenue growth was close to 70% for the year as well. I'm not sure I've ever seen that kind of margin transformation at a company of this scale, even after adjusting for some one-time IPO costs in the prior year that pushed margins even further down.

The debate, as always with Reddit, is around the quality of data and the potential of their ad business. Kyle made the obvious and correct point that Reddit knows very little about its individual users. Anonymity is a key part of the platform and what differentiates Reddit from platforms like Meta, TikTok, or X, but it also limits the precision of ad targeting relative to the aforementioned platforms. A meaningful portion of Reddit's traffic comes from users who aren't even logged in — they googled something, landed on a subreddit, read the thread, and then left.

But there are two things that I think the skeptics underweigh. The first is what Reddit's content actually is. It's human beings, often very knowledgeable ones, at least on the topic in question, discussing specific things in depth — investing, DIY repair, niche hobbies, medical questions, career advice, whatever.

That's exactly the kind of authentic, expert-adjacent content that is increasingly hard to find as the rest of the internet floods with AI-generated text. And it’s unlikely that changes. On most other social media platforms, more views mean more money or fame. On Reddit, that doesn’t exist, so there’s no incentive to increase posting activity through AI.

That makes Reddit's archive of 20 years of human content genuinely valuable to the companies training the next generation of AI models. The data licensing business is a direct result of that.

International expansion is another part of the growth story. Reddit is still overwhelmingly a US-centric platform, with nearly 70% of revenue coming from just five English-speaking markets. But international users are growing three times faster than US users. The gap in average revenue per user is still enormous — a US user is worth roughly $11 to Reddit, an international user about $2.30 — but closing even a fraction of that gap would add meaningful revenue at high margins.

Airbnb is our second-largest position at 11.5%, and it's a company we've spent a lot of time on. The basic pitch is well-known: a pure marketplace connecting hosts and guests for alternative accommodation, with a blended take-rate of around 13-14% (about 3% for hosts and 12% for guests) on gross booking value and strong network effects.

Looking at how the fundamentals have performed since IPO in 2020, the first question that comes to mind is why the stock hasn’t done anything since then. And, as so often, the answer is valuation. At IPO, Airbnb was trading at a P/OCF multiple in the low hundreds. Since then, the multiple has come all the way down to the high-teens to low-twenties, which is where it sits today. So the stock mainly didn't go anywhere because it was starting from an unreasonable place.

But there have been other headwinds, too. The post-pandemic travel boom created an inflated baseline that made growth look like it was decelerating when it was really normalizing in the years after.

Beyond that, you had some smaller but still impactful issues regarding hidden cleaning fees and Airbnb's overall reputation. Regulatory pressure, such as New York's crackdown or Barcelona's limits on short-term rentals, serves as a recurring reminder that Airbnb's supply depends on local political tolerance in a way that traditional hotel aggregators like Booking.com don't. Kyle told us about the pressure and backlash Airbnb faces in Canada, and I heard that from multiple Canadian members of our Community as well.

This adds a layer of uncertainty that investors obviously don’t like and, to some extent, discount. But we still believe that the structural advantages are stronger. Airbnb has the strongest direct traffic of any accommodation platform — a high proportion of guests go directly to Airbnb rather than arriving through a Google search.

That matters enormously as the advertising model for travel evolves around AI, with the traditional Google search-to-booking funnel potentially becoming less important. A company like Booking.com is famously paying a “Google tax” given how much it spends on Google ads to show up first in search results.

Supply exclusivity is another underappreciated factor. The majority of Airbnb's individual hosts don't list on VRBO or Booking.com. Managing multiple platforms isn't worth the effort for someone renting out a guest room or a vacation cottage. That means a guest searching for a specific property type in a specific location genuinely cannot always find it elsewhere.

And then you have the expansion into Experiences, which, to be fair, is still in early stages and has to prove that it can be more successful than the first attempt. The new approach embeds activity recommendations directly into the accommodation booking flow, so when you book a beach house, Airbnb surfaces surfing lessons and boat trips at exactly the moment you're in a planning mindset.

Network effects and scale play a role in this segment, too, so it’s normal that it will take some time. The more people use it, the more “Experience hosts” will join the program, and the more hosts join, the better the options, and the more users will use it. Airbnb will also learn over time what users want. If you book an expensive beach house, you might rather want to offer a yacht trip rather than a party boat to the guests.

We believe that Airbnb still has lots of runway, considering the push into new segments, but also just looking at the core business. About 70% of Airbnb’s revenue still comes from five countries. They operate in more than 200 markets, though, so there’s plenty of room to expand in the markets where they are already active.

Universal Music Group is a company that is a bit under the radar. In part, because it is not an American business and it only trades in Europe at the time. But also because streaming platforms, like Spotify, tend to get more investor attention. We have looked at Spotify, but we prefer Universal as an investment.

UMG (short for Universal Music Group) owns the rights to roughly a third of all recorded music in the world — Taylor Swift, Drake, The Beatles, Billie Eilish, Kendrick Lamar, BTS, Bad Bunny, Rihanna, Elton John. The list is absurdly long. It is the most valuable music catalog on earth.

The business model is essentially a perpetual royalty stream. Every time one of UMG's songs gets streamed on Spotify, featured in a Netflix scene, used in a commercial, or licensed for a film soundtrack, Universal gets paid.

There are two major income streams: recorded music, which is the master recording royalty, and publishing, which covers the underlying composition — the melody and lyrics. These generate separate payments, so if someone samples a Rihanna hook, Universal earns on the publishing rights entirely independently of what Spotify pays for the stream.

What makes this model exceptional from an investment standpoint is the cost structure. Once you own the rights to a song, you own a perpetual royalty stream with essentially zero marginal cost. Nothing needs to be manufactured, reprinted, or re-served. As a result, free cash flow conversion runs above 80% of operating profit.

Another structural advantage is that the catalog doesn't depreciate. More than 70% of music streams today are catalog, which means older music, not new releases. To add to that, social media has become an incredible engine for reviving old music with entirely new audiences. All it takes is one meme or new trend with a song from the 60s in the background, and it goes back to the No.1 spot on the charts.

And the competitive structure is oligopolistic in the most favorable way. Three labels — Universal, Sony, and Warner — control roughly 98%(!) of the top 1,000 singles globally. Universal alone commands about 30% of both recorded music and publishing. Streaming platforms cannot afford to lose Universal's catalog. Losing a third of all available music would be an existential problem for Spotify or Apple Music. That structural leverage means UMG benefits automatically from streaming price increases. So when Spotify raises subscription prices, Universal captures a percentage of that increase at zero cost and with zero backlash.

Just recently, a takeover offer from Bill Ackman caused the stock to jump about 30%. The offer itself seems a bit odd to me. I don’t view it as a viable option. It uses almost no cash, and the valuation of the shares (the buyout price) is based on a future valuation that Ackman suspects(!) the newly formed company (he plans a merger between Pershing Square and UMG) should have. Doesn’t sound right to me. But one way or the other, it’s a positive for the stock. Even if nothing comes from it, it serves publicity and clearly shows that Ackman sees the stock as massively undervalued.

As teased in the beginning, we also made some changes to our portfolio. It has been less about losing convictions in the companies we sold than about gaining convictions in others. Amazon below $200 seemed like too good an opportunity to pass on, and we decided to bump it up from 5% to 9% (after this week, it went up to present over 10%).

The two companies we sold to finance the Amazon increase were TransDigm and Copart.

We've always admired TransDigm's business model, and nothing about our view of the underlying business has materially changed. For anyone less familiar, TransDigm acquires small, niche aerospace component businesses that are certified for specific aircraft designs.

The attractive thing about this industry and business model is that once your part is on an aircraft design, it stays there for the entire operational life of the plane, which can easily be 30 to 50 years. Even if a competitor builds a part that can do the exact same thing with the exact same quality, it can’t replace the part that was originally chosen to be on the aircraft. So if you win that spot once, you keep it for decades.

So when Transdigm acquires companies, it looks for those that build parts that have little competition and are vital to an airplane. The more aftermarket exposure, the better.

The aftermarket — replacement parts and maintenance — is where TransDigm makes most of its margin. It accounts for roughly 30% of revenue but 75% of adjusted EBITDA. And because an unplanned grounding can cost an airline tens or hundreds of thousands of dollars a day, TransDigm's customers are willing to pay almost any price for replacement parts. So even when they seem expensive compared to their production cost, they are a rounding error compared to the overall cost of a plane or the cost of grounding a plane. That’s where the pricing power is coming from.

So why are we selling? When we established the position, we sized it at 2%, which we consider a starter position. The implicit plan was always to either scale it up to a full 5% position once we'd developed higher confidence, or exit if we couldn't get there. After sitting with it for a while, our conviction hasn't grown to the point of becoming a full position. It hasn't fallen either, but in a portfolio where every position has to compete with the others for capital, that's not good enough.

TransDigm is a complex company trading at a reasonable but not cheap price. The use of debt and special dividends, the heavy leverage, and a private-equity-style capital structure require a level of conviction we haven't been able to build with certainty. And when you look at it alongside something like Amazon, we decided the opportunity cost of holding it is too high.

Copart is yet another business we have a lot of respect for. It operates in a duopoly alongside IAA in the salvage vehicle auction market, and for a long time, Copart was the clearly dominant operator. Insurance companies, fleet owners, and dealers submit totaled or damaged vehicles to Copart's auction platform, where a global buyer base of dismantlers, rebuilders, and dealers bids for them. There are strong network effects in this business: more sellers attract more buyers, more buyers attract more sellers, and the flywheel reinforces itself over time.

But we ran into the same structural issue as with TransDigm. It's a 2% position, sitting in a part of the portfolio that's supposed to be reserved for what you might call moonshots — companies that are earlier in their lifecycle with asymmetric upside. Copart doesn't fit that profile. It's an excellent, mature business, but mature businesses need to be held at full position sizes or not at all. At least, that’s how we think about it.

In my opinion, Copart is attractive at its current valuation, but it also faces some headwinds. IAA has become much better operationally since the acquisition, and high levels of uninsured cars are hurting the business as well. This is not why we sold, but it’s part of the opportunity cost calculation.

I don't want to go too deep into the weeds on Amazon and Mercado Libre today. I've just published a dedicated episode with Clay covering both theses in full. You can listen to it here.

In short, Amazon sits at the intersection of cloud infrastructure, AI, chip investments, and logistics automation, and I think the compounding effect across all of those areas has been significantly underappreciated by the market recently.

AWS cloud demand continues to outpace supply, and the $200 billion investment has made the market nervous about near-term overcapacity. But thinking five years ahead here, I can hardly imagine demand not growing into that capacity. And through AWS Bedrock, Amazon is positioning itself as the neutral infrastructure layer for enterprise AI adoption, giving customers access to hundreds of foundation models, including Anthropic's Claude, where Amazon holds a substantial stake.

So, instead of taking part in the expensive and highly competitive LLM field, Amazon benefits regardless of who wins that race.

Another less-discussed part of the thesis is how AI strengthens the entire Amazon flywheel, not just AWS. In logistics, the next generation of robotic systems can handle the full picking-and-packing workflow. Amazon's fulfillment cost base runs around $90 billion annually. Even a 10–15% reduction from automation would add $9-$14 billion to the bottom line without selling a single additional item.

On the customer side, AI improves search and product discovery, makes ad targeting more precise, and allows Amazon to better personalize the shopping experience across hundreds of millions of users. That directly feeds the advertising business, which already generates over $70 billion in revenue annually and is one of the highest-margin segments in the entire company.

Margins in the core business have already doubled from 5% to 11% since 2021. I think that's just the beginning, and the true earnings power is still well ahead of what the current price reflects.

Some companies have long been requesting an update, so here it is!

Trade Desk has been one of the most requested watchlist companies from you over the past few months, and it's not hard to see why. The stock peaked at around $140.

Today it's trading somewhere in the $20–25 range. That's an 85% decline from the all-time high. For a company that many considered one of the highest-quality names in the entire market just eighteen months ago, that kind of move demands an explanation.

For those less familiar, Trade Desk is the leading independent demand-side platform, or DSP. When brands want to buy advertising across the open internet — streaming TV, podcasts, sports apps, news sites — they typically work through a marketing agency, and that agency uses a DSP to automate and optimize the buying process. Trade Desk sits on the buy side only.

They own no ad inventory, so their incentives are fully aligned with the advertiser's. It’s about getting the best outcome for the campaign, not the best margin for the platform. For every $100 that flows through Trade Desk's system, they keep about $20. That’s a pretty high take rate.

The independence and neutrality that come from not owning inventory are structural advantages of Trade Desk. Google, Meta, and Amazon all own enormous amounts of ad inventory and run their own platforms, so when you buy through their DSPs, you're buying from a broker with a financial conflict of interest.

Trade Desk is the neutral alternative, which is why advertisers should have a structural incentive to keep it alive. Without a viable alternative to the walled gardens, Meta, Google, and Amazon’s pricing power over advertising budgets would increase substantially.

But why did the stock decline so much if Trade Desk has such an important function in the advertising market?

The decline started when Trade Desk missed its own revenue guidance for the first time in 33 consecutive quarters. Q4 2024 revenue came in at $741 million — $15 million below the low end of their own guidance. The culprit was the transition to Kokai, their next-generation AI platform, which created more friction in client workflows than management anticipated. The stock fell 30% in a single day.

Then it got sequentially worse through 2025. Revenue growth decelerated from 25% in Q1 to around 14% by Q4. The most recent guidance implied only around 10% growth and a year-over-year decline in adjusted EBITDA. Then came a Publicis memo advising clients to avoid Trade Desk over fee-structure concerns, followed by similar moves from WPP and Dentsu regarding the OpenPath platform.

Agency relationships represent a large share of the Trade Desk's spending base, so any disruption to that relationship can have meaningful impacts on the business.

But there’s not only bad news. The business still has genuine tailwinds. Connected TV is one of the most important ad budget migrations of this decade — more money is still being spent on linear cable TV than on streaming, despite a massive shift in actual viewing habits. That transition will take years, but it will certainly happen. Trade Desk's Unified ID 2.0 has established itself as the leading privacy-safe alternative to cookies for audience targeting in streaming. If UID2 becomes the industry standard infrastructure layer for CTV advertising, Trade Desk sits at the center of something very large.

But we keep coming back to a question of the moat. The neutrality advantage makes sense to us in theory, but the agency pushback suggests that at a 20% take rate, Trade Desk may be charging more than the structural value of that neutrality justifies. And unlike Copart, where only a handful of large insurance companies need to coordinate to prevent a monopoly, the advertising market has thousands of participants who each act on their own short-term incentives, and for many of them, that means spending with Amazon, Meta, and Google because that's where the returns are.

Back when Shawn pitched the stock, we ended up putting it on our too-hard pile, and we can’t say that has changed. We lack the industry insight and technical understanding of ad tech needed to answer the moat question.

FICO is the other watchlist name many of you have been asking about recently. FICO is obviously an exceptional business — a near-monopoly in credit scoring that has been raising prices aggressively for close to a decade without losing meaningful volume. Getting a FICO score cost $0.60 in 2018. By 2025, it cost around $5. The most recent increase took it to $10. With these price hikes and almost no additional cost to produce the score, it’s not surprising that operating margins are extraordinary.

The reason we didn't buy it at its peak is the same reason we haven't bought it yet: valuation. At its high, FICO was trading at over 100 times earnings. And even after a 60% drawdown from the highs, FICO is still trading at around 30 times forward earnings.

The business has also run into its first meaningful regulatory challenge. For decades, FICO scores were the only model accepted for government-sponsored mortgages — the loans that flow through Fannie Mae and Freddie Mac, which account for nearly half of all new home loans in the US.

That regulatory mandate was the bedrock of the moat. In mid-2025, the FHFA officially approved VantageScore 4.0 as an alternative for those government-backed loans. And as of January 2026, the system moved to lender choice, so banks can now pick FICO or VantageScore when originating a government-backed mortgage. That broke FICO's exclusive mandate for the first time in the company's history.

In practice, we think the near-term competitive threat from VantageScore won’t have a material impact on FICO, though. The mortgage market runs on decades of underwriting models built around FICO. Investors in mortgage-backed securities know what a 720 FICO score means across different economic cycles. VantageScore doesn't have that data. We don’t expect lenders to switch en masse because of a regulatory change. They are more likely to use VantageScore as a negotiating lever against FICO's pricing, which is probably the most significant consequence of all of this.

I guess our main problem comes down to the fact that you have to pay a premium price still for a company where most of the growth story depends on raising prices in a politically sensitive market. You're always one bad headline or one determined regulator away from a significant change in the earnings trajectory. We'd rather own companies where the moat is so broad and so organic — like Alphabet or Amazon — that you can sleep soundly through almost any competitive or regulatory scenario. We'll keep it on the watchlist for now.

For more Deep Dives and Portfolio Updates, you can listen to our podcast here.

More updates on our Intrinsic Value Portfolio below 👇

“When most people say they want to be a millionaire, what they might actually mean is ‘I'd like to spend a million dollars.’ And that is literally the opposite of being a millionaire.”

— Morgan Housel

📺 WATCH: The new Richer, Wiser, Happier Episode with William Green and Matthew McLennan

🎧 LISTEN: Steve Eisman interviewing Software Analyst Rob Oliver on the Software Selloff

📖 READ: J.P Morgan’s Article on Anthropic’s Mythos

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!

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.”

Follow us on Twitter.

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].