The numbers from Celsius have been out now for a while and in a few weeks, probably around May 11, the company will release its Q1 2026 earnings. But the stock has come down substantially and that’s why I have asked David to have a look at the previous earnings.

As always, I have added and adapted to David’s output, so you can see this as a collab article.

The key numbers

Q4 2025 revenue came in at $721.6 million, up 117% year-over-year and beating analyst expectations by over 10%.

Year-over-year growth was primarily driven by the addition of Alani Nu, of course, but still, a beat by 10% is very strong.

Full-year 2025 revenue reached $2.52 billion, an 86% increase versus 2024.

That growth was largely driven by North America, where revenue increased 89.1% year-over-year. $1B came from Alani Nu. The other acquisition, Rockstar Energy, added $55.6 million, while the core Celsius brand grew 7.5% to $1.46 billion.

Source

Q4 gross profit came in at $341.8 million, up year-over-year but down sequentially versus Q2 and Q3 but that’s not unusual, as there is some seasonality for energy drinks and, therefore, for gross profits.

If you don’t want to miss all the great Fiscal AI data, get 15% off here.

Full-year gross profit increased by $587 million to $1.27 billion, up from $680 million in 2024, up 86.7%.

The stock dropped after the earnings were released, but this was also the period when everything went down. The geopolitical tensions between the United States and Iran were starting.

Multi-brand portfolio

Looking at the full-year market share, we can still see that Red Bull and Monster are on top of the market, but Ceclius remains strong in a third place. Alani Nu and Rockstar Energy maintained their Q3 positions into Q4, with market share remaining stable. Altogether, the combined portfolio controls just under 20% of the U.S. energy drink market.

Celsius remains well-positioned, expanding strategically without overextending. While it’s unlikely that Celsius will overtake Red Bull or Monster in the near term, it holds a unique advantage with its multi-brand portfolio.

Having a multi-branded portfolio has both advantages and disadvantages. Celsius must execute multiple marketing strategies, as each brand targets a distinct consumer segment.

While all products fall under the energy drink category, each brand targets a different demographic. On the other hand, the risk of becoming obsolete diminishes significantly with diversification.

Think here of the next new fad we see every year, that slowly fades into obscurity. Celsius can simply move within its product offering, ride the entire wave of new fads, and remove the old products from its lineup.

Not so for Red Bull and Monster; both have their flagship products and occasionally play with other varietuies.

Celsius has three different product lines to work with. Celsius itself is focused on a younger core of customers who are conscious about their health.

Alani Nu focuses on a female audience, with a strong emphasis on lifestyle and wellness.

And lastly, Rockstar, which focuses on a male audience, using high-energy, high-performance slogans. Psychology here is one of the extreme sports.

Having these three different lines can increase the overall marketing costs. This is something which is visible in the numbers.

Marketing and selling expenses increased by $200.6 million, driven by diverse campaigns totaling $76.6 million (notably “Live. Fit. Go”) and marketing focused on Alani Nu totaling $65 million.

Source

Celsius wants to prioritize differentiation, which, in turn, increases costs because packaging must remain fresh and maintain its uniqueness and appeal.

I don’t see its marketing costs going down anytime soon. Developing brand awareness through marketing initiatives and expanding abroad will both need a big marketing push.

Combine this with different marketing campaigns for all the products in its portfolio, and the costs can only become higher in the near future.

The chart above highlights how the top three brands are currently competing in the market.

MULO (Multi-Outlet) includes channels such as grocery stores, mass retailers, and drugstores. It also includes convenience stores. Think of 7-Eleven, gas stations, and similar retail formats. Monster and Red Bull remain the category leaders, but Celsius (across its full portfolio) is steadily gaining share, as you can see.

Competition

Recently, Celsius stock went down after Costco Wholesale introduced Kirkland Signature energy drinks.

Nothing new under the sun, we might say. there are plenty of competitors. However this time it is slightly different. Costco is a US-based retailer that sells bulk goods at lower prices. Here, it is the same: Costco sells a lot of products under its private label, Kirkland Signature.

As you can see, Kirkland Signature energy drink has the same look as Celsius: white with colored banners featuring fruit images. The flavors? Similar to Celsius’s popular flavors. But the drink is much cheaper than Celsius.

But does it all matter? We often see new energy drink brands coming onto the market. It is not an easy market, and gaining traction often takes considerable time and investment.

I think it is too early to tell whether the new offering impacts Celsius’s top line. Celsius generated about 11% of its sales within Costco in 2025, a large number, but let’s not exaggerate. Implementing a new product does not mean it is doomed even within Costco. People still have brand loyalty or they don’t want to be seen with the “cheap knock-off” version. And many people are simply not aware of the product. See this X post, for example:

Numerous competitors have come and gone. The energy drinks market is growing rapidly and becoming increasingly attractive, enticing competitors to capture a share of it.

There was another worry as well. Iran bombed aluminum smelters in the Middle East and two major sites were seriously damaged. But the Middle East as a whole (not these two) are only responsible for 9% of the global supply. So, I think this fear is exaggerated.

If you are not a paying subscriber, the article ends here for you. I hope you enjoyed it. If you want to know if I would buy Celsius now, consider upgrading. One good idea might pay back the investment.

Subscribe now

Read more

It’s Sunday and that means the Overview Of The Week again. On Tuesday, I am leaving for Omaha for the Annual General Meeting of Berkshire Hathaway.

It’s the first time without Warren Buffett, but I am still looking forward to it. To be honest, the Berkshire AGM is OK to follow live, but you can probably follow it better through the livestream. The really fantastic thing about Omaha in that week is that everyone who means anything in investing is there and there are numerous events in the days around the AGM.

I have the honor of inviting you to two of those.

The first is one I organize with Pieter Slegers and Brian Feroldi. On Saturday evening from 8.15 pm to 10.15 pm in the speakeasy Howard & Fine, inside Moe & Curly’s Pub And Grill.

Please register if you want to come.

You can do that here.

I hope to see you there!

Secondly, my friend Vitaliy Katsenelson organizes a FREE BREAKFAST in Omaha on Sunday, May 3. He allowed me to invite every single one of you. Important: if you want to go, you have to let him know. No registration, no breakfast. But it’s totally free. Don’t wait too long, because, while there is a big capacity, I think it will be full fast. I will be there as well. You can register here.

One last announcement: since I will still be in Omaha next week, there will be no Overview Of The Week next Sunday.

Articles In The Past Weeks

This is the third article this week.

The first one was about ServiceNow.

In the second one, I combined the Best Buys Now and the additions to the Forever Portfolio. You can read that article here.

Memes Of The Week

Four memes this week. The first one plays on Bitcoin’s weak performance over the last year.

Source

If you want to know why I’m a long-term investor, this is one of the reasons.

I also had to laugh with this one.

Flo shared this one about ServiceNow.

Interesting Podcasts Or Books

I listened to a very interesting Invest Like The Best podcast episode this week. The guest was Dylan Patel and if you want to have insights on the edge of AI and chips. You can listen to it here.

The markets in the past week

The Russell 2000 was up 0.36%, the S&P 500 0.55% and the Nasdaq 1.50%.

The Greed & Fear Index surged incredibly in the last weeks. It was at 9 on the 14th of April, now it’s already at 66, in Greed territory.

Quick Facts

1. An Exceptional Market?

This is a weird market. Despite Iran and the ensuing energy crisis, the indexes are at all-time highs. The fact that this is exceptional can be seen in this Bank of America chart.

Source

A portfolio of 25% in stocks, bonds, commodities and cash is up 26%. That is the highest return since 1933, when the stock market was still suffering from the results of the Great Depression. That shows that these are not normal times. On the other hand, are there “normal” times in investing?

But at the same time, the fact that there’s a fast recovery after a crisis is not that exceptional. Look at this overview.

Source

Investors often panic, but just focus on the long term, not the short term. Not that focusing on the long term is easy.

Now, just to be clear, the recovery has been remarkable. The S&P 500 is up 12.5% over the last four weeks and that was the 20th-biggest 4-week jump since 1950.

Source

2. Tech? Tech! Well, Some Tech.

The S&P 500 at all-time highs is not that crazy if you see that the earnings estimates for this year and next year keep going up. But the crazy thing is that almost all of the estimate upgrades come from three sectors: energy, materials and tech.

Source

As you can see, tech is very resilient, with EPS estimates up already 11% this year and 15.1% for next year.

But of course, tech is divided into multiple subdomains. If you look at the companies that

Source

As you can see, Micron alone is responsible for 51% of the EPS revision for the S&P 500. That’s totally nuts and it shows in what exceptional times we live. Add Broadcom and you already have 61%.

And it’s not just Micron. Semiconductors now make up 14% of the Total US stock market cap. That’s up 3x in three years and double as much as during the dotcom bubble. Crazy times.

Source

3. Charlie Munger and Howard Marks

I’m showing this X post completely, as it is important.

An overwhelming majority of people are focused on the short term, even if they think they are not.

One of the biggest investing mistakes you can make is selling winners too early. That’s why I always err on the side of holding too long, not selling too early. Selling too early can feel right over the short term but can be a huge way of missing out on the best investment of your life.

4. Earnings In The Upcoming Week

Normally, I do this at the end of the article, but this time, I put it here, as the upcoming earnings week is pretty crazy, especially on Wednesday, when Microsoft, Amazon, Meta, and Alphabet all publish their results at the same moment. Chipotle Mexican Grill, Brookfield, and KLA Corporation also report at the same time. Apple and Mastercard report on Thursday and Spotify, Visa and Robinhood report on Tuesday.

Source

What a week! In total, 172 S&P 500 companies report in the upcoming week, so I think it’s fair to say this is the most important earnings week for the market when it comes to Q1 2026 earnings.

If you are a paid subscriber already, you can just scroll past this. If you are not, please read this.

When you join Potential Multibaggers now, you get my entire investment system:

✅ Best Buys Now: every month, the 5 best stocks to buy

✅ My Proprietary Quality Score: Rating companies on 17 quality metrics nobody else uses

✅ Deep Earnings Analysis: Thorough breakdowns that reveal what really matters

✅ Valuations & Buy/Hold/Sell scales: Cut through the noise and know exactly when to act

✅ 15,000+ Word Deep Dives: Each pick gets 4-5 extensive articles so you understand exactly what you own. The first article alone (part 1!) is already 7,600 words. What others call a ‘deep dive’ is just the starting point of a part of an article for me. This is the work professionals literally pay thousands of dollars per report for!

✅ Private Community: Direct access to me and 800+ serious long-term investors.

And you can see what I buy every time.

Subscribe now!

Read more

As you know, I was on holiday last week. I was in Egypt. This is a picture from the sunset taken from the swimming pool.

I postponed the Best Buys Now until today because I will also combine it with buying shares for the Forever Portfolio. The portfolio additions will be made later today, but I’ll let you know what I will buy in this article.

But before we go to the content of this article, I want to invite you.

On Monday, I’m leaving for Omaha, with a stop in Minneapolis. I have two invitations for you.

The first one is one from my friend Vitaliy Katsenelson. He organizes a FREE BREAKFAST in Omaha on Sunday, May 3. He allowed me to invite every single one of you. Important: if you want to go, you have to let him know. No registration, no breakfast. But it’s totally free. Don’t wait too long, because, while there is a big capacity, I think it will be full fast. I will be there as well. You can register here.

Secondly, I want to invite you to a drink with me, Pieter Slegers (Compounding Quality) and Brian Feroldi. The drinks are yours to pay, but we will be there to talk to everyone who wants to talk with us.

The drink will be on Saturday evening from 8.15 pm to 10.15 pm in the speakeasy Howard & Fine, inside Moe & Curly’s Pub And Grill.

Same thing, you have to register to be there.

You can do that here.

I hope to see you there! But back to the content of this article.

This has not been an easy market for many investors. If you look at the indexes, they’re at all-time highs again. And at the same time, many of the stocks we own have completely underperformed the S&P 500, not just temporarily, but over a few years in some cases. Several of the Potential Multibaggers stocks have underperformed over the past five years.

Bank of America came out with research that showed that a 25% portfolio in stocks, bonds, commodities and cash has had its best return since 1933, when the world was still recovering from the Big Depression.

I believe this is all temporary. But the problem is that temporary can feel like a very long time. How do you think these Amazon investors felt after five years, from 1999 to 2004?

Some Multis will also remember how Shopify dropped almost 85% in 2022 and Cloudflare almost 82%.

You can get angry, but that won’t help you, to the contrary even. You will feel stressed and take the wrong decisions. What I did then is the same as what I do now: I keep looking at the underlying business, not the stock price.

We don’t know when the stock price action will turn, but we do know there is a transition going on right now with AI. And those kinds of transitions always make investors insecure. They don’t really know what will happen, and it is kind of crazy to think about what this new technology has become in just three and a half years. ChatGPT was introduced in November 2022 and here we are, not even three and a half years later, and look at how much AI has influenced many people’s lives already. At the same time, we also still see the limitations of AI. So it’s still not easy to separate noise from signal, which creates the uncertainty investors are dealing with right now.

I’m fully convinced that many companies will have problems shifting to AI. And it’s actually not that hard to see which ones. It’s the companies that don’t jump on the train, that don’t innovate enough. They will be out-competed.

To a certain extent, this is also different from the dot-com bubble. There is a real cost now, and that cost is tokens. Tokens are expensive, especially for the most advanced models, which are the ones you want if you’re on the cutting edge. That also means there is a barrier to entry here that didn’t exist with earlier waves of disruption.

Look at what happened with Claude Mythos, the new Anthropic model, which they have not released to the general public because of what they call safety reasons. About 50 of the biggest companies, tech companies, banks, have gotten access, and they had to pay up for it. They did so without hesitation. And there are many, many companies begging Anthropic on their bare knees to be let in, because they don’t care much about the cost of the tokens. They care more about staying competitive.

So it’s not as if those big companies are being disrupted from all sides. And I think that’s part of what you see right now. If you look at companies like Cloudflare or CrowdStrike, which have been let in on the new model, they will use it to have an edge on new competitors. That’s the complete opposite of what you normally expect from innovation. This is a very special situation.

But enough about general remarks. Let’s get to the Best Buys Now.

If you are a paid subscriber already, you can just scroll past this. If you are not, please read this.

When you join Potential Multibaggers now, you get my entire investment system:

✅ Best Buys Now: every month, the 5 best stocks to buy

✅ My Proprietary Quality Score: Rating companies on 17 quality metrics nobody else uses

✅ Deep Earnings Analysis: Thorough breakdowns that reveal what really matters

✅ Valuations & Buy/Hold/Sell scales: Cut through the noise and know exactly when to act

✅ 15,000+ Word Deep Dives: Each pick gets 4-5 extensive articles so you understand exactly what you own. The first article alone (part 1!) is already 7,600 words. What others call a ‘deep dive’ is just the starting point of a part of an article for me. This is the work professionals literally pay thousands of dollars per report for!

✅ Private Community: Direct access to me and 800+ serious long-term investors.

And you can see what I buy every time.

Subscribe now!

Read more

I’ve had a refreshing holiday in Egypt, and one of the things I wanted to do every now and then is go a bit broader than purely Potential Multibaggers to some stocks that are also in my portfolio. You will not get the same level of detail on these as on the Potential Multibaggers, but it will be what many others call a deep dive, haha. But these articles will be shorter and definitely not regular. Don’t think you will get a quarterly article on ServiceNow from now on.

If you don’t want to miss high-quality articles like this, subscribe now!

Subscribe now

Potential Multibaggers will still have absolute priority. I’m also working on part two of the new Potential Multibaggers pick and I hope to have that out later this week, but I need a bit of a broader scope and I think many of you will appreciate this as well. If you don’t, I’m sorry, but this is what I will do from time to time. I’ll also focus more on doing the Potential Multibaggers articles (from the picks, I mean) and the earnings articles will become a bit shorter, unless there’s a good reason to make them long (for example, a deep search if the thesis still holds).

But enough introduction, let’s go to the subject of this article: ServiceNow (NOW).

ServiceNow reports tomorrow after close, and the stock is trading at around $100. I’ve followed this one for a long time, and I’ve recently started a position for the first time in quite a while, as you may know.

The drawdown context

As is often the case in the market, the drop isn’t really about the results.

Q4 2025 was actually a strong quarter on most fronts. Subscription revenue was up 21% year-over-year to $3.47 billion, and total revenue grew 20.5%, which is strong for a company at this scale. Current RPO was up 25%. CRPO is the money the company already has under contract for the next 12 months but still has to recognize as revenue because it still has to deliver the services. So, if you look ahead, ServiceNow also seems to be in excellent shape.

On top of that, ServiceNow closed 244 AI deals worth more than $1 million in net new ACV (annual contract value), which is substantial already. For 2026, management guided subscription revenue to a range of $15.53 to $15.57 billion, good for 19.5-20% growth. Knowing guidance is usually conservative, that’s strong. The board also authorized an additional $5 billion share repurchase program.

A lot to like there, right?

And yet the stock dropped about 15% the day after the earnings, and while it has recovered a little, it’s still 23% below where it was before earnings.

So, if the reason was not in the results, it must be in something else. As so often in investing (this is consistently underestimated), it was in the story that has been circulating for months now.

It goes something like this: AI will disrupt everything, and since ServiceNow sells software to enterprises, ServiceNow must be a loser. I have called this the “anti-bubble” for months already. As the biggest companies that use AI are private (OpenAI, Anthropic, X), you see that there’s the opposite of what happened in the dotcom bubble. Instead of bubbles in big companies, supposed losers are shorted and sold, purely based on sentiment. Probably a lot of that money is being invested in the next round of Anthropic and/or OpenAI.

Now, I’m not saying the story about AI disrupting SaaS is completely wrong, but it’s extremely exaggerated. That means the child is often being thrown out with the bathwater. I would be very surprised to see ServiceNow end up at the losing side of this shift. And that mismatch, between a strong business and a weak narrative, is exactly why I started building a position.

The Platform Thesis

The first thing to clarify is that ServiceNow isn’t a point solution. It’s a platform, and that distinction matters much more than most people seem to realize.

A point solution solves one specific problem, so if AI can solve that problem better or cheaper, the point solution dies. That’s a real risk in SaaS right now, and for plenty of companies, it’s a genuine threat. For me, it’s an absolute filter for which stocks to buy or not. A platform is the layer everything else runs on, so it doesn’t die when a new technology arrives. The new technology runs on the platform, not the other way around.

ServiceNow knows this and it doubles down on this. It’s positioning itself as a sort of AI control tower, a central dashboard where enterprises can monitor and manage all their different AI agents from one place. So, that’s a product enterprises need because of AI, and I think that will become very valuable over the next few years.

ServiceNow isn’t only helping other companies manage AI, but it’s also monetizing AI itself, through its NowAssist, which was on track to generate over $500 million in annual contract value coming out of Q4.

The partnership strategy has been smart, too. Rather than trying to compete with OpenAI or Anthropic on foundation models, which would be a losing game, ServiceNow has integrated with both. The company is building the workflow layer that sits on top of whichever model wins, which is exactly where I want a platform to be positioned.

Then there’s the retention number, and this is where I think people underestimate the business the most. This is a chart since 2018. Do you see the AI impact? I don’t.

Retention at ServiceNow is so high because ripping it out of your system once it’s embedded is incredibly difficult. It’s often a multi-year project involving dozens of teams and workflows across the enterprise, and nobody takes that on just to save some money. You only rip out a system like that when something is badly broken, and right now, nothing is broken. Sometimes, you don’t have to overcomplicate things.

Which brings me to the other thing the market is worried about: pricing. The story is that AI agents will replace human seats, so seat-based revenue collapses, and the whole business model cracks under the weight of its own legacy. But a company with ServiceNow’s position isn’t going to just sit there and watch its revenue model die. I don’t get why people think that everything will change with AI, except for the companies adapting to it. We’ve seen that story with Google last year. The sentiment has totally shifted and I think ServiceNow could see the same thing. That’s not a prediction for tomorrow, but for the longer term.

If AI agents take off the way the bulls claim they will, that’s actually great news for ServiceNow, because usage will explode. More AI agents means more workflow transactions, which means more revenue per customer. The companies at risk of disruption are those that can’t switch to a usage-based pricing model because there’s not enough activity on their platform or, even worse, in their solution.

Quality And Valuation

The simplest way to test whether a software company is doing great is the Rule of 40, where you add revenue growth to free cash flow margin, and anything above 40 suggests you’re looking at a high-quality business.

Now, look at this.

34.5% + 20.9% = 55.4%, which is simply outstanding!

The important thing is that it has been there consistently, not just in one good quarter. Yes, revenue growth has slightly declined (but not much) but the FCF margin has increased.

These aren’t the numbers of a company with a pricing problem or a demand problem.

For a quick valuation, I use my own framework, summarized in this graph.

As ServiceNow continues to grow by around 20%, which you might argue is still high growth, let’s first look at an EV/Gross Profit basis.

As you can see, on an EV/Gross Profit basis, ServiceNow is the cheapest it has ever been since its IPO almost 14 years ago. But maybe the market thinks this company is no longer in high growth and has switched to mature growth? Fair, let’s check that as well. Again, you see the same pattern: the stock has never been this cheap.

If you don’t want to miss all the great Fiscal AI charts and other tools, get 15% off here.

Now, let’s also check the forward PEG (Price/Earnings/Growth).

If we divide the expected PE by the expected growth, the 2026 PEG is 1.23, which is on the cheap side of fair. The 2027 PEG is almost exactly 1, the exact border between cheap and fair. The 2028 PEG is just 0.77, plainly cheap.

So what you’re paying for is a company compounding revenue at roughly 20%, with 98% retention, and an AI monetization story already in place, at a multiple that sits far below its historical averages and bordering on cheap. That’s why I find it interesting right now.

Three things I’m watching tomorrow

With the earnings tomorrow, I’ll watch a few things in particular.

If you don’t want to miss high-quality articles like this, subscribe now!

Subscribe now

The first is net new ACV from AI deals. The Q4 number was 244 deals worth more than $1 million, and I want to see what Q1 looks like, particularly for the AI-specific cohort.

The second is cRPO growth, which, for me, is probably the most important forward-looking number in the entire report. It tells you what revenue is already under contract for the next 12 months, it’s much less noisy than headline revenue, and it gives you real visibility into the pipeline. Q4 grew 25%, which is very strong, and if Q1 holds anywhere near that level, the top-line guide for 2026 will probably be revised a few times upwards in the coming quarters.

The third thing I’ll be listening for is any commentary from management on the evolution of the pricing model. They’ve hinted at a move toward usage-based pricing, and I want to hear if there’s any news on that front.

I try to be honest about what would change my mind, because a thesis without a break point is just a story you tell yourself.

If cRPO growth decelerates meaningfully below 20%, that would worry me, and if NowAssist ACV goes flat or down quarter-over-quarter, that’s a real red flag. If the renewal rate were to suddenly drop by a few percentage points, that would be worrying.

No single one of these kills the case in isolation, but all of them together would, and I’d rather be clear about that up front than pretend otherwise.

The bottom line

The market is lumping all SaaS companies into one pile right now, and I think that’s both a mistake and an opportunity for long-term investors.

Some SaaS businesses genuinely will lose to AI. Point solutions doing narrow tasks, tools that wrap thin features around a user interface, anything where the core job can be done better or cheaper by a model... Don’t get me wrong, there will be SaaS companies in trouble and I stay away from many now.

But I believe ServiceNow won’t be one of them. It’s the workflow platform that enterprises have used for over a decade, with a real AI product already generating revenue, a credible platform thesis for the agent era, and a pricing model that it can adapt as usage patterns shift.

That’s why I recently started a position. Tomorrow’s earnings will provide valuable data to see if my thesis is on track.

In the meantime, keep growing!

Subscribe now

Did you check your calendar when you saw the Overview Of The Week come in?

It’s Saturday, not Sunday, but you already got the Overview Of The Week today.

The reason could already be deduced from the title. I’m going to Egypt with my family for a week. As a child of the 80s, I have to think of the great Bangles hit “Walk Like An Egyptian” every time I hear the country’s name. I know it’s maybe a bit immature and culturally not fair for a country with such a rich history, but it’s true.

Source

As a sidenote, the songwriter, Liam Sternberg, was a bit tipsy when he crossed the English Channel. The sea was choppy, and passengers and waitresses had to do moves to keep their balance. It reminded him of the postures of figures in Ancient Egyptian tomb paintings. But back to today.

The consequence of this very-last-minute holiday is that you get the Overview Of The Week on Saturday already, that you will hear me less often in the upcoming week and that there will be no Overview Of The Week next Sunday (no, not next Saturday either).

I have worked on part two of the new pick. I already have about 6,000 words, but I could cut or add words depending on my editing process. So, that’s one you might get in the upcoming week. Or, if not in the upcoming week, definitely in the week that follows. It depends on how much time I will dedicate to working there (don’t pity me, I still love what I do!) and how many hours the article still needs, which I never know exactly and chronically underestimate.

Articles In The Past Weeks

This is just the second article this week. The first one was, of course, on the new pick. You can read the full 7,600 words here.

Memes Of The Week

Three memes this week. This is the first one.

And the second one.

I always love it when people can break out of the current paradigm to mock it, like this one, for example.

Interesting Podcasts Or Books

One of the biggest influences on my investing was David Gardner. As long as he still talked about stocks himself, until 2021, I never missed an episode of his podcast. Now, I listen every now and then and this week, I did that. In this episode, David Gardner talks about little tricks that can help you in life.

The markets in the past week

For a while now, the market and tech stocks have been disconnected (more on that later), so if you are overweight tech (like I am), it’s now harder to assess the broader market by looking at your portfolio alone.

Let’s look at what the indexes have done this week.

As you can see, the markets were up substantially again, just like last week. The Nasdaq won 4.68%, the Russell 2000 3.97% the S&P 500 3.56%.

So, where does that leave us year-to-date?

So, after this week:

• The S&P 500 is down just 0.42% YTD,

• The Nasdaq is -1.46% YTD

• The Russell 2000 is up 5.99% YTD.

Remarkable. Because it doesn’t feel like that. But those are the numbers.

At the beginning of last week, the Greed & Fear Index dropped to 9 but it ended the week at 19. This week, it continued to zoom up, to 38.

That’s still in Fear territory, but the sentiment seems to be improving fast.

Quick Facts

1. Nasdaq 100 Replaces Atlassian by Sandisk

This week, the Nasdaq 100 has booted Atlassian (TEAM) from the Nasdaq 100. It is replaced by Sandisk (SNDK). There was not a single reason why Atlassian was replaced, but the fact that the stock is down the second-most over the last five years won’t have helped. Only “PainPal” did worse over the last five years.

But being removed from the index is not always as negative as it sounds. Sometimes those stocks have already had a big drop and the removal sometimes marks the bottom. On the contrary, the stocks that come in have often had a big recent jump, and that goes for Sandisk too. It’s up more than 25x over the last year.

But, as said, that doesn’t always mean it’s a good thing to “sell your losers” and “keep your winners.” In general, if the investment thesis holds, it’s often a good thing to hold the stocks that are out of favor.

Charlie Bilello gave an example of this with an index replacement. In August 2020, Exxon Mobil was removed from the Dow Jones Industrial Average, and Salesforce replaced it. Exxon returned almost 4x since then, while Salesforce is down almost 40%.

2. Private Equity In Trouble? Or Opportunity?

For about a year or so, I have said every now and then that I saw a bubble in private equity. I couldn’t put my finger on it exactly, but I saw initiatives by people with questionable integrity that wanted to ‘democratize’ private equity. If that’s the case, you know something is fishy.

Private credit was the first problem for private equity. This story is still evolving, but it’s clear some private equity funds will get hurt. The second and simultaneous problem for private equity is the massive exposure to software. Only in the last few months of this year have software deals come down a bit, but they still account for 49% of all deals.

Private market managers poured hundreds of billions of dollars into software firms over this period, betting that software-as-a-service (SaaS) business models would produce strong growth and consistent cash generation. The private equity industry has never been this exposed to a single sector. But now valuations crash and don’t forget that private credit was also mostly given, probably much more than 49%, to software companies.

I’m very sure that some of the software names out there are babies being thrown out with the bathwater, but there will be AI disruption of software companies that only have a nice interface, but no proprietary data, insights, high switching costs or network effects. That means that this is a double whammy for private equity.

Now, the market has sold off many private equity companies and there too, some opportunities could spring up soon. I’m thinking of companies like Brookfield (not that far off its high) and KKR, both part of a cream-of-the-crop group in private equity.

3. Amazon’s Hidden Chip Giant

Amazon’s stock was up about 12% this week after its CEO Andy Jassy published his annual shareholder letter and gave a few interviews.

If you don’t want to miss all the great Fiscal AI tools, get 15% off here.

Jassy shared that Amazon’s AWS AI revenue was already at a $15 billion annual run rate in Q1 2026. Evercore’s Mark Mahaney, a legend in tech investing, had forecasted somewhere between $5 and $10 billion. Three years into the AI wave, AWS’s AI run rate is nearly 260 times bigger than AWS was at the same point in its history. Not a typo. 260x.

The crazy thing is that AI revenue wasn’t even the best news. The chip business inside Amazon seems to have been underestimated for a very long time.

Amazon’s custom silicon division, which includes Graviton (’CPU’), Trainium (AI accelerator), and Nitro (networking), is now at a $20 billion annual run rate.

That doubled from $10 billion at last quarter’s earnings. On top of that, Jassy said if Amazon’s chip division were a standalone business selling to third parties like any other chipmaker, the annual run rate would be around $50 billion. That would make it comparable to Broadcom’s AI chip business.

The market assigns zero value to Amazon’s chip business because it’s buried within AWS. And yet it’s potentially a $50 billion standalone operation. That’s bigger than 82% of the Fortune 500.

Two large AWS customers asked if they could buy all of Amazon’s Graviton capacity for 2026. Amazon had to say no because, well, other customers exist.

On the Trainium side, Trainium 2 is completely sold out. Trainium 3, which just started shipping a few weeks ago, is nearly fully subscribed already. And a significant chunk of Trainium 4, which is still about 18 months away from broad availability, has already been reserved.

AWS CEO Matt Garman confirmed at re:Invent that all currently available Anthropic models were trained on Trainium. There are now 1.4 million Trainium chips deployed across all three generations, and Anthropic’s Claude runs on over a million Trainium 2 chips. Anthropic helped AWS co-design the chip from the ground up, working at the kernel level with Amazon’s Annapurna Labs team.

This all tells you something about the chip demand. For Amazon’s chips, but also chips in general. As I wrote last week, it may be time to buy Nvidia again.

Free cash flow dropped from $38 billion to $11 billion because Amazon is spending almost $200 billion in capex in 2026, mostly on AI infrastructure. That scares people. But Jassy said Amazon already has customer commitments for a substantial portion of this capex, with most of it expected to be monetized in 2027 and 2028. That includes the $100 billion-plus OpenAI commitment, on top of the Anthropic partnership.

In January, I made 5 bold predictions for 2026. I’m still confident that these two will prove to be true.

4. Anthropic: The Stock Destroyer (part 372)

At this point, Anthropic might be the single most effective stock destroyer on the planet. Every time it puts out a press release, dozens of stocks drop like a rock. The market sells first and reads the press release in full later.

It happened again this week. Anthropic announced Claude Mythos Preview, a new AI model it considered too dangerous for public release, and launched Project Glasswing, a cybersecurity initiative around it.

Mythos found thousands of vulnerabilities across every major operating system and browser, including a 27-year-old bug in OpenBSD, a system that’s famous for being unhackable and a bug that had been hiding in FFmpeg for 16 years and survived five million automated scans.

Sounds impressive, right? It is. But look at what happened before anyone even digested the details.

When Fortune first leaked the existence of Mythos back in March, cybersecurity stocks cratered. CrowdStrike dropped around 10%, Palo Alto 8%, Cloudflare 14%. The narrative instantly flipped from “cybersecurity spending is non-discretionary” to “AI will make all security software obsolete.” Billions in market cap gone in hours, all because of a leaked draft blog post.

Then, this week, the actual announcement came out. And guess what? CrowdStrike and Palo Alto were named as launch partners.

This is the pattern now. Anthropic says something, the market panics, software and cybersecurity names get slaughtered, and then people actually read the announcement and realize it’s more nuanced than “everything is dead.” But they have sold their stocks and then invent a rationally-sounding explanation like “more uncertainty.”

This fits into the anti-bubble I’ve been writing about. In the dotcom bubble, the biggest bubbles were in public companies. Think Yahoo!, Pets(.)com, Webvan and many others.

In contrast, this time, the biggest bubbles are in private markets: Anthropic, OpenAI, SpaceX, Stripe, ...

I’m not saying these aren’t phenomenal businesses; they clearly are. But the valuations are crazy. And the result is that money is being pulled out of public markets to chase these private names (see above in this Overview Of The Week for the percentage of software in private equity).

This puts more selling pressure on the public companies that are supposedly disrupted. Look at this chart that Michael Batnick shared:

This is about Wednesday and Thursday. The S&P 500 was never up this much with software down this much. This already shows you again that the movements are extreme.

Public shareholders get hit twice: once by the panic selling after every press release, and again by the capital outflows into private markets chasing the next hot AI round.

At the same time, forward earnings expectations for software stocks continue to go up.

Source

I’m not saying Anthropic is doing this on purpose, but it’s a hypemachine. An AI model too dangerous for public release? I don’t know about you, but damn, I want to try that out! I’m still frustrated a lot with the current models, for many use cases.

The model is clearly very powerful. It found four vulnerabilities that escaped for so long. Anthropic said these capabilities weren’t even trained intentionally. They just emerged.

But the market keeps pattern-matching “new AI capability = existing companies are dead” without doing the work. The companies inside Project Glasswing, Amazon, Apple, Microsoft, Google, Nvidia, CrowdStrike, Palo Alto, Broadcom, and about 40 others, are getting $100 million in usage credits and early access. They’ll be stronger from this, not weaker. Being at Anthropic’s table is probably better than being on its menu.

If you are a paid subscriber already, you can just scroll past this. If you are not, please read this.

When you join Potential Multibaggers now, you get my entire investment system:

✅ Best Buys Now: every month, the 5 best stocks to buy

✅ My Proprietary Quality Score: Rating companies on 17 quality metrics nobody else uses

✅ Live Portfolio Access: Every stock I own, every transaction I make, in real-time

✅ Deep Earnings Analysis: Thorough breakdowns that reveal what really matters

✅ Valuations & Buy/Hold/Sell scales: Cut through the noise and know exactly when to act

✅ 15,000+ Word Deep Dives: Each pick gets 4-5 extensive articles so you understand exactly what you own. The first article alone (part 1!) is already 7,600 words. What others call a ‘deep dive’ is just the starting point of a part of an article for me. This is the work professionals literally pay thousands of dollars per report for!

✅ Private Community: Direct access to me and 800+ serious long-term investors.

Normally, this costs $1,200 per year.

But now, you can get this for just $399!

I’ve added a $800 credit to your account.

That means you can get all of this for just $399 instead of $1200!

This pick is a company I’m personally loading up on, just like many other Multis are.

Claim your $800 discount NOW!

I’ve limited this to 25 spots at $399 instead of $1,200.

This is not a fake number! There are 25, not more.

In a bull market, they’d be gone in hours. But there are spots left now.

I'm in!

Read more

Every now and then, I feel like everything comes together and it’s the right moment to start building a position in a new company.

Sometimes it’s a recent discovery that grew really fast on me, but even then, it’s usually still months for me. But I’ve been following this one for years. To be precise, since 2019. That’s 7 long years.

And all the while, I felt like I had missed it. But that’s the great thing about the stock market: there are always new chances. Like many other growth stocks, this stock has dropped by 50%+ and this feels like a present from the market that I can’t ignore.

Growth remains high after all these years despite headwinds, and there’s visibility into the future, with more exciting innovation and an expansion of the current offerings. The company has undergone a fundamental shift in recent years that only makes it more interesting.

Before we start, I want to point out that I have an advantage over you. I’ve already done the research. Not just in the last few weeks but over a long period. That means I’m probably in a different place than you are right now when it comes to conviction. Take your time and see whether this stock is something you want in your portfolio.

Something I have been thinking about quite a bit lately is labels and how they can be misleading sometimes. Of course, we all use them to make sense of the world quickly and labels can be very useful in that sense. But they can also make you miss the essence of something and for a long time, I think that was the case with this company. I knew what was happening inside the company, I saw its new products made sense, I understood the reasoning for investing, but the label was one of the reasons that kept me from investing.

And still to this day, the label is partly misleading. But that’s OK, because it creates a competitive advantage for those who did the work: me, but also you after reading this article.

If you are a paid subscriber, just scroll past this.

If you are not, this is your opportunity!

This pick will be in my personal portfolio, just like my other picks. I put my money where my mouth is. As a subscriber, you can see my exact positions and transactions almost in real time.

No hiding. No BS. Complete transparency.

When you join Potential Multibaggers, you get my entire investment system:

✅ Best Buys Now: every month, the 5 best stocks to buy

✅ My Proprietary Quality Score: Rating companies on 17 quality metrics nobody else uses

✅ Live Portfolio Access: Every stock I own, every transaction I make, in real-time

✅ Deep Earnings Analysis: Thorough breakdowns that reveal what really matters

✅ Valuations & Buy/Hold/Sell scales: Cut through the noise and know exactly when to act

✅ 15,000+ Word Deep Dives: Each pick gets 4-5 extensive articles so you understand exactly what you own. This article alone (part 1!) is already 7,500 words. What others call a ‘deep dive’ is just the starting point of a part of an article for me. This is the work professionals literally pay thousands of dollars per report for!

✅ Private Community: Direct access to me and 800+ serious long-term investors.

Normally, this costs $1,200 per year.

But now, you can get this for just $399!

I’ve added a $100 credit to your account.

That means you can get all of this for just $399 instead of $1200!

This pick is a company I’m personally loading up on, just like many other Multis are. It could be a generational wealth builder. If you’ve been waiting for the right time to join, this is it.

Claim your $100 discount NOW!

Read more

Happy Easter if you celebrate.

I don’t always publish the Overview Of The Week on such special days, but I wanted to do it this week, especially because I have an announcement to make. And of course, as always, some thoughts about the markets. Let’s dive in.

Articles In The Past Weeks

This is just the second article this week. The first one was on Topicus. You can read it here.

A NEW PICK!

There were two reasons I published only one article this week.

The first one was private. On Wednesday evening, my good friend Michael Gielkens invited me to Maastricht to hear Sofina CEO Harold Boël speak.

(Source)

If you don’t know Sofina, it’s one of the best Belgian holding companies. It’s been run by generations by the Boël family, who made their fortune through steel.

While I was in Maastricht, I also took my bike with me and did two days of cycling, which really gave me the same feeling as a full week of vacation.

The first day, I did a relaxing and flat 50-km ride in Belgian Limburg, the second day was an 80-km ride in Dutch Limburg. There, I did some hills you might know from the Amstel Gold Race (if you like professional cycling): Cauberg, Geulhemmerberg, Bemelerberg, Looorberg, Camerig, Vaalserberg and several others that are not included in the race.

But there was a second, for you much more important reason why I only published one article this week: research.

Two weeks ago, I already wrote this:

I’ve been researching much more this week, as there are more attractive stocks with this pullback. There may be crystallizing something, but I’m not sure yet.

It’s that same stock that really convinced me in the last few weeks. I’m always careful to share anything because I want to have the option to withdraw at any step of the process. If I find something that impacts the thesis too much, I want to throw away my work. Without a doubt or without sunk cost fallacy. If I announce too much in advance, I feel too obligated to live up to my promise.

But I have done all the work now and found nothing about the new pick that worries me too much and a lot that makes me really excited. That’s why you will get the new pick this week, probably on Tuesday or Wednesday. I have the outline of the first article, but it still needs some work.

I’m very excited about this new pick. It has been on my Watchlist for a long time and because of the recent pullback, I think it’s the right time to invest in this company.

Memes Of The Week

Just one meme this week, but it’s a good one.

Interesting Podcasts Or Books

This week, I listened to the No More Investing Secrets podcast.

The episode was about Constellation Software, the SaaSmageddon, Wise and much more. You can find it here.

The markets in the past week

This week, the markets were up substantially. The Nasdaq won 4.44%, the S&P 500 3.36% and the Russell 2000 3.28%.

While the Greed & Fear Index dropped to 9 early in the week, it ended the week at 19. Still in Extreme Fear, but not as much as it was.

Quick Facts

1. Buy Stocks!

Ben Carlson wrote this on X this week:

No one knows how long a correction will last when you’re in it But buying stocks when they are down tends to be a good long-term strategy A look at what happens when you buy the stock market down 10%, 20% and 30%.

I keep repeating such stats because I know it’s necessary. I’ve seen it so many times before: people nodding when I point such things out, even saying “of course” when I show them the theory, but when it comes to action, they are still doing the exact opposite of what they theoretically know to do.

The difficult thing about investing is that we are used to immediate feedback, not delayed feedback. In that sense, investing resembles the long-term effort of building a business, training for a sports career, or anything else that takes years to build. There will be rough patches, like injuries in sports, but only with persistence, you will succeed.

This overview shows the same thing.

(Source)

As you see, in most of the years, 63% of the years to be exact (12/19), the stock market was positive from day 60 to the end of the year.

Now, to be honest (and long-time Multis know this), I don’t care about my returns for the year. I focus on the real long term, preferably decades.

2. Is It Time To Buy Nvidia Again?

This week, several people sent me this and today, Karan also shared it on Slack.

Quietly, Nvidia seems to have become very cheap again, at just 18.7x forward earnings, a multiple it has not traded at since 2018. That’s crazy, if you know that the stock is up 60% over the last year and 1,170% over the last 5 years.

The funny thing is that many people think Nvidia is expensive because it’s up so much. But you can see here, that it’s not multiple expansion, just sheer earnings power.

The Return Of The Potential Multibaggers

So, how did our stocks do this week? This is an overview of all the stocks that moved 5% or more. There were 10 stocks this week, all up more than 5%.

Do you want to know which 10 stocks and get the new pick? Then it’s time to upgrade your subscription!

You can try out Potential Multibaggers for 7 days and unsubscribe with just a few clicks if you don’t like it.

Don’t wait! Try it out now!

Read more

Some weeks ago, Topicus (TOI.V) (TOITF) reported its Q4 and FY 2025 results and this is a deep dive into those earnings and the potential AI impact on the company.

As you probably know, I have a second service called Best Anchor Stocks, which is run by Leandro. He is also a Topicus stockholder and wrote an earnings analysis for Best Anchor Stocks subscribers. I used that as a basis for this article, but I added context, commentary, and my takes. Of course, as always, I will update the Selling Rules, the Quality Score and the valuation.

Right now, the stock is down more than 50% from its all-time high, set not even a year ago.

Is that warranted or not? Let’s find out.

The Numbers

Topicus grew its revenue 20% year-over-year in both Q4 and the full year. Q4 revenue came in at €436.8 million, up from €364.9 million. Full year revenue came in at €1,552.3 million (or about $1.8B), up from €1,294.9 million. Not bad for a “small” European software compounder that most people haven’t even heard of.

Organic growth was 4% for both periods, with maintenance and other recurring organic growth at 6%. Now, 4% organic growth doesn’t make you jump out of your chair. But 6% on the recurring side, the revenue that’s sticky and predictable, is actually quite decent for a serial acquirer that constantly buys companies with sometimes flat but mostly declining organic growth. Remember, the Constellation/Topicus playbook is to buy cheap, stabilize, and slowly get that organic revenue moving upward. The 6% recurring organic growth tells you the existing portfolio is healthy.

What most people miss here is the composition of that revenue. Maintenance and other recurring revenue hit €1.097 for the full year, up from €900 million in 2024. That’s a 22% increase and represents about 71% of total revenue. This is the core of the business. This revenue stream is highly predictable, sticky revenue and margins are high.

Professional services came in at €372 million (up 14%), while hardware and other revenue jumped to €39 million from €25 million, a 59% increase that likely reflects some of the acquired businesses. But overall, it’s just 2.5% of revenue.

One of the problems with Topicus is that the numbers are not easy to read and that was the case again with these earnings. The headline earnings look terrible if you take them at face value. Net income dropped 53% for the full year, from €149.5 million to €70.1 million. On a per share basis, diluted EPS fell from €1.11 to €0.50. If this is all you look at, you’d think the company is under pressure. But it’s not.

The full-year net income decline is almost entirely driven by a €221.7 million non-cash (!) expense related to the Asseco Poland investment. When Topicus elected to record this investment at cost under the equity method of accounting, it triggered this one-time charge. It’s accounting noise, not operational problems. On top of that, the company also recorded €119.7 million in income from derivatives, fair value adjustments, and a dilution gain tied to Asseco. So, even within the Asseco-related items, it’s a messy picture that tells you essentially nothing about how the actual business is performing.

This is why you need to look at cash flows. And the cash flows are singing a beautiful song.

Operating cash flow grew 19% for the full year to €412.7 million, up from €347.6 million. In Q4 alone, CFO jumped 35% to €107.7 million.

The metric that matters most for Topicus is free cash flow available to shareholders, or FCFA2S. In short, FCFA2S is the cash flow Topicus would generate for its shareholders if it stopped doing acquisitions, after stripping out all the money that goes to minority partners and Constellation Software, the parent company.

FCFA2S grew 23% to €218.7 million for the year.

In Q4, FCFA2S was up 40% to €51.2 million. One thing I appreciate about Topicus versus Constellation is that FCFA2S is actually a clean number here. There’s no IRGA (intergroup revolving account) muddying the waters, which makes Topicus’ cash flow a much more transparent measure of what shareholders are actually earning.

Digging into the FCFA2S numbers, a few things stand out. Interest paid on other facilities nearly doubled from €5.7 million to €13.0 million in Q4, and went from €21.1 million to €24.9 million for the year. Not very high on the total, but it reflects the higher debt load from the Asseco stake and, of course, other acquisitions.

Lease obligations also grew meaningfully: from €24.6 million to €30.1 million for the year. This tells you the acquired businesses come with real physical footprints.

But the most important takeaway is that despite these increased costs, the FCFA2S still grew 23%. That’s strong.

Topicus now also receives interest and dividends from its Asseco stake. This contributed €9.5 million in 2025 and €1.4 million in Q4 alone. It’s a small but new source of cash flow that will recur and probably grow over time.

Record Capital Deployment

This is where the quarter and the year get really exciting. 2025 was by far the strongest year on record for Topicus in terms of capital deployment, and it wasn’t even close.

Between Asseco and other acquisitions, Topicus deployed €662 million of capital (net of cash acquired) during 2025. To put that in perspective, last year the number was €77 million. The year before that, €199 million. This was a big shift.

Now, a huge chunk of this was the Asseco Poland investment (€384.9 million net). But even excluding Asseco, the company completed acquisitions for total consideration of €390.4 million (including holdbacks and contingent consideration). The net cash spent on non-Asseco acquisitions was €248.3 million (€311 million in acquisitions minus €62.7 million in cash obtained with those businesses). That’s still an all-time record for Topicus.

What’s important here is that this capital deployment was roughly 3 times the full-year FCFA2S. In other words, Topicus deployed significantly more than it earned. How? By leveraging. That’s only good as long as the leverage stays manageable. And it does.

Topicus ended the year with €326.7 million in cash and total debt of €692.5 million (split between €345.3 million in revolving credit facility/current term loans and €347.2 million in non-current term and other loans). That puts net debt at roughly €366 million.

That’s a leverage ratio of about 0.9x operating cash flow. For any normal company, that’s very conservative and that also goes for a serial acquirer.

I also want to highlight the maturity structure: the non-current term loans of €347.2 million don’t come due until at least 2032.

That’s crucial because it means Topicus can continue using debt as fuel for acquisitions without worrying about near-term refinancing.

The balance sheet also reveals the scale of change this year. Total assets grew from €1.54 billion to €2.51 billion — a 63% increase in a single year. Intangible assets (the acquired software) went from €951 million to €1.20 billion. The investment in equity method associates (Asseco) is now €515 million, up from just €2.5 million. This shows that the company completely transformed in 2025.

And there’s still plenty of dry powder. Topicus has room under its revolving credit facility, generates over €400 million in annual operating cash flow, and has debt maturities far out. This is a company that could keep deploying capital aggressively if the opportunities are there.

The environment looks favorable for deals. Private equity distributions grew meaningfully in 2025, but these were concentrated in large deals. The mid-market, where Topicus hunts, remains stagnant. PE holding periods have stretched, and limited partners are pressuring funds for returns. That’s the kind of environment where a disciplined buyer like Topicus can find deals at attractive prices.

You should know that right now many VMS companies are somewhat in panic. They don’t want or can’t adapt to the new AI wave. They see valuations crumble and that’s when they are more ready than ever to sell. With so much flexibility left, Topicus can seize the moment for some attractive acquisitions.

Of course, capital deployment is always lumpy for a company like this. Remember that just two years ago, Topicus was being criticized for paying a special dividend because it couldn’t find enough deals. How the tables have turned.

AI Disruption? Look At These 3 Numbers

Let’s address the (supposed?) elephant in the room. A significant part of Topicus’ drawdown from its all-time high is tied to the broader AI fear sweeping through the software industry. The narrative is that AI-native tools will replace vertical market software. I’ve written extensively about why I think this fear is overblown for a company like Topicus, but let’s look at what the actual numbers tell us.

First, organic growth. Maintenance and recurring organic growth of 6% doesn’t scream “being disrupted by AI.” It’s right in line with the historical trend. If AI were eating Topicus’ lunch, you’d expect to see churn accelerating and organic growth collapsing. We’re not seeing that. Not yet, anyway.

Second: staff expenses as a percentage of revenue actually decreased in 2025.

For the full year, staff costs were €835 million on €1,552 million in revenue, or about 53.8%. In Q4 specifically, this ratio dropped even further to around 52%. And this happened during a year of massive capital deployment, where newly acquired businesses typically come with a bunch of employees and usually drag efficiency down.

Within staff expenses, R&D spending grew considerably faster than overall staff costs. That’s the exact pattern you’d want to see in an AI world: the company is getting more efficient on the operational side while investing more in product development. One quarter doesn’t make a trend, but it’s a data point worth tracking closely.

Third, contingent considerations grew in 2025, while impairments slightly decreased. Contingent considerations are extra payments that the sellers of an acquired company can get if the company reaches certain milestones. An impairment is a write-down of an asset (like goodwill or an investment) when its value has dropped below what you paid for it. It’s an accounting recognition that you overpaid or the asset lost value.

If AI were destroying Topicus’ portfolio, you’d expect to see the opposite: acquired companies missing their milestones, resulting in lower contingent considerations and more impairments showing up. This is another indicator that the acquired businesses are performing in line or better than expected.

Now, I don’t want to be dismissive of the AI risk. It’s real, and it’s uncertain. But there are several structural reasons why I think Topicus is better positioned than most:

The switching costs in vertical market software go far beyond price. I’ve given the example before of a clerk in a Belgian municipality who’s been using Cipal Schaubroeck for twenty years. You’re not ripping that out because some kid built something with Claude. The compliance requirements, the security concerns, the retraining costs, the risk of downtime and so much more parameters protect the VMS software.

Topicus is heavily exposed to Europe and to government entities. These are among the most conservative buyers of software on the planet. As Asseco recently noted, customers in their space are reluctant to even store data outside of Europe or use the hyperscalers. That’s a structural tailwind for an entrenched European VMS player.

And perhaps most importantly: Topicus is not just a software company. It’s a capital allocator. If the VMS landscape changes because of AI, Topicus can pivot its capital allocation to wherever the returns are best. The compounding machine doesn’t depend on any single piece of software but on the discipline of buying great businesses at good prices.

Conclusion

2025 was a good year for Topicus. The company crossed €1.5 billion in revenue for the first time, deployed a record €662 million in capital, grew FCFA2S by 23%. Still, the stock price is down by more than 50%. The AI narrative has created fear, and the Mark Leonard succession story at Constellation has added uncertainty, even though Topicus has been operating independently under Robin van Poelje for years now.

The business is performing. The capital allocation is performing. The balance sheet is in good shape with room to keep going. That’s all I needed to know to know that the sell-off is, so far, only based on sentiment and stories, not the data.

So, on to the Selling Rules, the Quality Score and the valuation to know if Topicus is a buy.

The Selling Rules

I made Selling Rules for Topicus. Here they are.

Let me go over them one by one.

1. Organic revenue growth

You want to see organic revenue growth. Topicus buys businesses with low, no or even negative revenue growth. That’s why they are much cheaper. But the company optimizes those businesses to see revenue growth again, especially recurring revenue.

At the same time, I want to look at the 3-year trailing twelve months, not just the quarter or the year. The reason is that if Topicus does a big acquisition one day, it will need time to turn that business around. And there’s another reason to do this. Let me explain that with the numbers of this year.

Organic growth was 4% in both Q4 2025 and full-year 2025, but over the last three years it was 7.7%. So, you might conclude that organic growth was weaker in 2025, but that is not necessarily the case. In 2025, Topicus had record capital deployment. If you acquire more businesses, organic revenue will be lower that year because the new businesses have more influence on the total. Conversely, if the company deploys less capital, organic revenue will be higher.

2. FCFA2S

Free Cash Flow Available To Shareholders is a temperature check for how well the business compounds. I’m very happy that Topicus reports it, as it shows the fundamentals. If the company acquires well, FCFA2S will grow at a good pace.

So, why 15% and 10%?

FCFA2S shows what the business would do without acquisitions and accounting necessities. If growth is under 15%, it means that the business is decelerating in revenue growth or that the company’s acquisitions are of poorer quality.

Under 10% growth, the thesis would be broken.

Again, this is not something you want to look at on a yearly basis, let alone quarterly. Acquisitions are made when they are attractive. Once the company would start doing them just because, it would be a red flag. The lower quality would then also be visible in FCFA2S.

3. Recurring revenue as a % of total revenue

Recurring revenue was €1.097B in 2025, 71% of total revenue. This is the structural health metric for VMS businesses. The entire Constellation/Topicus model depends on sticky, recurring revenue.

If recurring revenue as a percentage of total declines, it means either: Topicus is acquiring lower-quality businesses (more project/services revenue), or its businesses see churn in recurring contracts. Both signal a deterioration of the business.

A drop to 65% recurring revenue, therefore, would signal more services-heavy businesses, which have worse economics: lower margins, less predictable, and lower switching costs. Below 58% fundamentally changes the character of the portfolio from recurring revenue to consulting/services, and then Topicus is simply another story, and the thesis is broken.

4. Net Debt / CFO Ratio

Net debt was €742.4M at the end of 2025, and cash from operations came in at €412.7M. That means a net debt/ CFO ratio of 1.8x. This is conservative for a serial acquirer but it is significantly higher than the previous years.

Constellation Software has never taken more than 2x in debt, but Mark Leonard also said that this may change because of certain circumstances.

So, a higher ratio would not be a red flag immediately, but of course, too much is too much. If the ratio were 3x, I’d trim because of the higher risk. Ät 3.5x, I’d sell, because it would totally change the dynamics. The same goes for issuing shares to pay debt. That would undercut the whole model. This is not about issuing shares if it’s an offensive move. In other words, if Topicus sees more opportunities than it has money and it is convinced this will be temporary, it’s OK to raise some money. But if it’s to pay off debt, that’s a red flag to me.

If you are not a paid member, this is where it stops for you. If you want to learn about Topicus’ Quality Score and my current valuation, you can try out Potential Multibaggers for 7 days. You can unsubscribe in the first 7 days with just a few clicks.

Don’t wait! Try it out now!

Read more

Yesterday, I was a speaker at the VFB Conference in Antwerp. You can see a picture here that Multi Frederic sent me. Thank you very much for that pic. It’s the only one I have so far. There were multiple professional photographers, but it takes a while before I will have those pictures.

There were 2,000 investors there and I spoke with many people. Overall, the sentiment is really negative now.

I also spoke to some Multis who were there and came to me. Thank you for those great conversations. And what stood out to me is that Multis are wired differently. While the general mood was one of negativity about the state of the world and investing, Multis can see through the clouds. They are more optimistic.

Now, that doesn’t mean it’s easy for Multis to invest now. We are all humans and that leads to conflicting feelings. On the one hand, many of you have seen this (and worse) before.

Sometimes, people seem to forget that in 2022, there was a crash for tech stocks. The Nasdaq as a whole was down 36%, more than during the Covid crash. Right now, the Nasdaq is down ‘just’ 12.5%.

But the problem, both then and now, is that many stocks are down much more. Here are just two stocks from then, for example: Shopify and Cloudflare. Shopify was down more than 85% from top to bottom and Cloudflare 83%. This chart doesn’t show you the exact numbers because the stocks bottomed at a different moment, but it gives you a feeling of what it was.

You see the same thing now. While the index isn't down much, some stocks are down a lot. There’s negative market sentiment for all that are down and some have real or perceived company-specific issues.

It’s always possible that some never recover. That’s inherent to investing. You can never have only winners. The only thing we can do is keep monitoring our holdings closely, look at the fundamentals (not the price) and have patience.

I promise I will do my utmost best to help you to bridge this period.

Maybe this stat can help you. This is a midterm year and many midterm years have negative returns but all the consecutive years have positive returns since 1950, with an average return of almost 40%.

Source

Of course, this is not a guarantee. Every situation is different and correlation is not causation. But don’t forget that the short-term market movements are emotional. But what will be the impact of the Iran war? We don’t know. We can only keep investing. Time in the market will always beat timing the market.

What we see now is a multiple contraction. The market dropped from a forward PE of 23.1 to 19 times in just a few times.

Source

And do you want more reassurance? I saw this post.

Again, be careful with these kinds of predictions. The markets are unpredictable. But overall, I think the sentiment is too negative now.

And you can see that from this as well. Nomura released a chart this week and I think it’s interesting.

It shows S&P 500 implied skew versus realized skew. In plain English, the red line is how much people are paying for downside protection (puts versus calls). The grey line is how much the market has actually moved down.

Look at the right side of the chart. The red line is spiking to high levels, even though I also want to point out that the period is short. But it shows that investors want protection against a crash. Many seem to expect the worst.

But if so many are already hedged, it actually protects the index from a big drop. After all, you won’t sell if you are hedged. I’m not saying we can’t go lower. We absolutely can. Duh! But it seems like a lot is already priced in. Maybe not in the index but in the hedges.

Also, don’t forget that the stock market just pulls back every now and then. Ben Carlson made this chart with an overview since 1928.

I also laughed with his description of what a bear market is:

Something that only lasts for like a month and happens every 3-5 years

So, don’t overreact. Just stay invested. Just look at this post.

I always see people who say that if you miss the worst days, the result is similar. Of course it is, but you won’t time the worst days well without missing the best days. In that context, this sentence is crucial:

8 of the 10 worst days happened within 1 week of the best days.

Ok, one more, to close this off. This chart.

Source

So, except for the mid-70s, the bottom was always very close when the stock market dropped 10 of the last 11 trading weeks. Again, correlation and causation are not the same, but maybe we are closer to the bottom than we might realize. Judging on the sentiment I hear from investors, which is very gloomy, that might be the case.

But the best thing of all is that you don’t have to predict this right. Just stay calm, keep investing and that alone will make you stand out.

And drink coffee or tea, if you like that. :-)

Articles In The Past Weeks

This is the fourth article this week.

In the first article, we analyzed TransMedics’ results and whether the stock is a buy now.

The second article showed that AI is a benefit, not a headwind, for CrowdStrike.

In the third article, I showed you which stocks I bought for my PM Future Fund portfolio.

Memes Of The Week

It was a bit calmer on the meme front this week, with just two memes I liked enough to include here.

The second one tells you something about perception. The S&P 500 is now at the same level as in September 2025, but it feels totally different.

Interesting Podcasts Or Books

I didn’t find the time to listen to an investing podcast or start a new book this week.

The markets in the past week

I don’t think I have to tell you that most stocks were down, but surprisingly enough, the Russell 2000 was up 0.46%. The S&P 500 was down 2.12% and the Nasdaq 3.23% this week. That’s quite a lot for indexes.

You won’t be surprised that the Greed & Fear Index dropped further, to 10.

Quick Facts

1. AI is not killing engineering jobs

Look at this chart from Trueup, shared by Lenny Rachitsky this week.

67,665 open engineering jobs globally. That’s 78% up from the low of 37,982. The most open engineering roles in over three years.

But yeah, of course, software engineers will be replaced by AI, right?

What we’re seeing here is Jevons Paradox playing out in real time. When something becomes cheaper to produce, you don’t use less of it. You use more. Way more.

Think about what’s actually happening on the ground. Companies outside of tech are realizing they can now afford software projects they would never have given the green light two years ago. Marketing teams at big corporations are bringing in engineers to automate workflows. Life sciences and healthcare companies are hiring engineers to automate research pipelines. Small businesses are hiring their first engineer ever to build better digital experiences. I could go on, but you get the picture.

Projects like this were too expensive or too complicated before AI made software incrementally cheaper.

Jevons Paradox means that if something becomes cheaper, the demand increases exponentially.

That means there will be much more software than ever before. And the people who understand how to build, maintain, and fix it will be wanted, just like they have been for years.

AI agents still require a human who understands what to prompt, how to review when an agent hallucinates, how to maintain the system that was built and so on.

All the advice floating around telling young people not to go into engineering is, in my opinion, very wrong. It sounds smart gets you clicks but data says the opposite. AI is currently not killing engineering jobs, no matter what you hear.

Of course, you might be shown this graph.

In that context, the rise of job postings looks minimal But you can see that the huge peak was caused by Covid, not by AI. Many people were fired during Covid, and, of course, job openings peaked when things normalized again.

2. Recession? CFOs Don’t Think So

This week, I saw this chart from The Daily Shot.

It shows the CFO Survey, tracking optimism about the US economy versus optimism about their own company. And there’s something interesting going on.

CFO optimism about the broader economy, the black line is down a bit. That’s what you would expect, with tariffs, the situation with Iran and its consequences, and the overall sentiment.

But look at the blue line. CFOs rate the optimism about their own company still very high.

Don’t forget that CFOs are well-informed. They see the order books and the margins. So, the difference in perception is interesting.

Now, this doesn’t mean everything is fine. CFOs can be wrong too. They can’t look into the future without fail. Orders can be canceled. But right now, CFOs, who have some of the best data points out there, are still confident about their own companies.

To me, that information is more valuable than a thousand doom-and-gloom X posts.

3. The US Market Concentration In Context

There’s been a lot of drama around market concentration in the US. The Mag 7 are too big, the S&P 500 is too top-heavy, it’s dangerous, it’s a bubble, etc. You’ve probably heard it all.

For context, look at his chart from Bloomberg Intelligence this week.

It shows countries ranked by the percentage of the market in the 10 biggest stocks. The US has only around 30%, one of the lowest concentrations. Some go to 80%.

Some concentration is completely normal. Great companies tend to keep growing. If you do well, it’s normal that your market cap keeps growing.

One of the things I see constantly in investing is people overcomplicating things. No, a slightly higher concentration in the top 10 stocks in the US doesn’t mean there’s a bubble or anything else. There’s too much noise out there.

4. Software Is Not Dead, Part 347

I know I have repeated this message in several forms multiple times already. But if I feel that everyone else says it differently, I think it’s good to keep repeating this to have a counterbalance to the noise.

Thoma Bravo just released their LP meeting slides. For those who don’t know, Thoma Bravo is the world’s largest software-focused private equity firm.

And what are they seeing? Fundamentals have improved while valuations have compressed. Look at this chart.

Source

The Rule of Profile (free cash flow margin plus 2x revenue growth) has been trending up since mid-2024, hitting around 34-35%. At the same time, EV/FCF multiples have dropped like a rock because “AI will kill all software.”

Thoma Bravo’s view is simple and aligns with mine: all software companies are created equal. Business durability matters more than sentiment and sentiment creates opportunity.

Thomas Bravo has more data, more due diligence, and more skin in the game than any influencer tweeting about how AI will replace all SaaS. Or even multiple times more than hedge funds trading on this theme.

The market is not distinguishing between software that will be disrupted and software that will adapt and thrive. It’s selling everything. That’s how you get opportunities.

Do you want to continue reading?

Do you want to focus on the fundamentals of growth stocks?

Take a subscription to Potential Multibaggers.

Go to this page.

Subscribe to the annual plan.

Read more

Today, I added to multiple of my positions.

As you know, I add to my portfolio every two weeks and this was the second addition of March.

Let’s have a look at what I bought.

If you want to continue reading this article, go to this page.

Subscribe to the annual plan.

Read more

I promised you that there would be a second article from Zack this week and here it is. As always, I will update the Selling Rules, the Quality Score and then evaluate the current valuation.

Hi Multis, Zack here, with CrowdStrike’s earnings analysis.

To kick off the Q4 FY2026 earnings call, CEO George Kurtz highlighted seven points that perfectly encapsulated the quarter. I thought to myself, why reinvent the wheel? So to start, I’ll let Mr. Kurtz take it away with those seven highlights, with my own commentary interspersed:

One, all-time record net new ARR of $331 million for the quarter, which grew 47% year-over-year, coming in well ahead of our expectations. For the year, we delivered $1.01 billion in net new ARR, up 25% year-over-year. Our first year delivering over $1 billion of net new ARR.

Two, ending ARR of $5.25 billion, crossing the $5 billion milestone, which accelerated to 24% growth year-over-year.

Both ARR and Net New ARR continue to signal sales acceleration and market share gains.

For SaaS companies, annual recurring revenue (’ARR’) is often a better indicator of performance than revenue because it provides a clearer view into current and future obligations.

That said, Revenue came in at $1.31B, an increase of 23% year-over-year, beating estimates by 1%.

Gross margins continued to stay steady: GAAP subscription gross margin at 79% (up from 77% in Q4 last year), and Non-GAAP subscription gross margin was 81% (up from 80% last year).

Back to George Kurtz:

Three, record free cash flow of $376 million for the quarter, or 29% of revenue. For the year, we delivered record free cash flow of $1.24 billion or 26% of revenue.

CrowdStrike continues to impress with growing free cash flow and margin. This led to cash and cash equivalents growing to $5.23B ending FY2026:

If you want Fiscal AI at a 15% discount, you can do that here.

CrowdStrike has not taken any debt since 2021 and the rock-solid balance sheet provides stability and optionality for future acquisitions or buybacks.

On the topic of share buybacks, CrowdStrike recently repurchased $50.6M in shares, its first move since announcing its $1B program last year.

Moving on to Kurtz’s comments on operating income:

Four, all-time record operating income of $326 million for the quarter, or 25% of revenue.This is the third consecutive quarter of record operating income. For the year, we delivered $1.05 billion of operating income, exceeding the billion-dollar operating income milestone for the first time.

It’s worth noting that this refers to non-GAAP operating income. On a GAAP basis, there was a slight loss from operations of $6.9M.

While CrowdStrike is still technically at a GAAP operating loss, the gap (pun not intended) is primarily driven by Stock-Based Compensation (’SBC’).

Even when accounting for SBC, CrowdStrike is on the verge of being GAAP operating profitable again.

In fact, after adding in other income and tax provisions, the company achieved GAAP net income of $39M or $0.15 in EPS. This marks CrowdStrike’s first positive quarterly net income since the July 2024 outage.

Speaking of the 2024 outage, this was also the final quarter impacted by the Customer Care Program (’CCP’) credits. It turns out the CCP had a major silver lining despite the short-term revenue headwinds as Kurtz noted:

As we lapse the one-year mark from the end of our highly successful CCP program, we have seen that accounts that took CCP deals have gross and net retention rates higher than the company average, have shown a strong trend of early renewal, and have already expanded more than twice the total $80 million of ARR value we provided.

Talk about turning lemons into lemonade!

Onto Kurtz’s fifth point:

Five, record net new ARR from cloud, next-gen identity, and next-gen SIEM collectively. Ending ARR for these solutions collectively grew more than 45% year-over-year. Amidst today’s AI backdrop, our endpoint business accelerated for the second consecutive quarter.

New products are the bread and butter of CrowdStrike’s land-and-expand model. 45% year-over-year growth signals a massive appetite for the broader platform.

Here’s the performance breakdown followed by a quick summary of each module:

Source: CrowdStrike

Cloud Security: Protects cloud infrastructure and AI workloads. As AI usage increases, CrowdStrike’s revenue will grow along with it.

Next-Gen SIEM: Collects and analyzes event data to respond to threats in real-time. Growth in this newer module continues, indicating increased usage and market-share gains at the expense of legacy providers.

Next-Gen Identity: Protects credentials for both human and AI “employees.” Since 80% of breaches now involve compromised identities rather than traditional viruses, this layer is mission-critical.

While it isn’t illustrated in the graphic above, there have been two consecutive quarters of acceleration for the more established endpoint product. This is despite a false narrative that endpoint security is becoming commoditized.

The reason for the reacceleration? AI adoption.

With the rise of Model Context Protocol (’MCP’) servers, localized LLMs, and AI coding tools, the endpoint is becoming a massive attack surface as evidenced by George Kurtz:

As of Q4, our sensors detected more than 1,800 distinct AI applications running on enterprise devices, representing nearly 160 million unique application instances across our customer base.

With the proliferation of AI agents and processes running on local devices, basic antivirus protection is no longer good enough and more advanced endpoint protection is critical for modern companies.

Other Product Updates

CrowdStrike officially moved to acquire SGNL (zero trust solution for AI agents) and Seraphic Security (enhanced web browser security), two additional tuck-in acquisitions to expand its AI-focused modules.

They also announced the general availability of AI Detection and Response (AI-DR), which has been quite a success so far, growing 5x from last quarter.

Source: CrowdStrike

AI-DR acts as a modern “firewall” for AI. Just as a traditional firewall stops malicious files from entering your system, AI-DR prevents sensitive or “bad” information from being sent to AI models using prompt filtering and governance guardrails to ensure agents stay within their authorized boundaries.

With new module growth, it’s no surprise adoption rates have remained strong:

And continued product adoption goes hand-in-hand with Kurtz’s next point:

Six, dollar-based net retention of 115% and gross retention of 97%, showcasing best-in-class durability and stickiness.

Customers aren’t leaving, they are staying, and spending more.

DBNRR (dollar-based net retention rate) is 115%, which means the total revenue from the specific group of customers who were with the company a year ago grew by 15% over the last 12 months, including all customers who left or reduced their spending.

After a downtick in DBNRR for FY2025, we see re-acceleration heading into FY2027:

And to cap it off with the last achievement:

Seven, we delivered $1.69 billion in ending ARR from accounts that have adopted the Falcon Flex subscription model, growing more than 120% year-over-year, turbocharging our land and expand motion.

Falcon Flex subscriptions continue to be a huge success in growing customer product adoption and revenue and it’s becoming a bigger piece of the ARR pie.

Falcon Flex allows customers to toggle between modules within a single commitment without renegotiating contracts, leading to more seamless procurement and spending.

Critics worried this might artificially inflate adoption metrics with temporary usage, but “reflex” (renewal) data from George Kurtz debunks that:

The proof of Falcon adoption success is in the reflex. Customers are using what they buy and expanding their Flex commitments. More than 380 Flex accounts have already reflexed, representing more than 23% of the Flex customer base, up from 5% in Q1. The average ARR lift after a reflex is 26%, happening on average within seven months.

Our multiple time reflexes on average have an ARR lift of an additional 48% from their initial Flex subscription.

In summary, Falcon Flex unlocks never-seen-before adoption for customers. Flex is now how we go to market.

It’s clear that Falcon Flex isn’t just providing a short-term bump in revenue; customers are continually increasing their spend commitments over time.

For FY2027 guidance, management shared the following:

• ARR guide of $5.5B for Q1 FY2027, $6.5B for the full-year FY2027 (24% growth year-over-year)

• Revenue guide of $1.36B for Q1 FY2027, $5.91B for the full-year FY2027 (23% growth year-over-year)

• Non-GAAP income from operations guide of $309M for Q1 FY2027, $1.4B for the full year FY2027 (33% growth year-over-year)

• Free Cash Flow Margin of 33% in Q1 FY2027 and at least 30% for the full FY2027.

Overall, guidance is solidly on track. Historically, CrowdStrike “beats and raises,” which suggests we could see even higher margins and top-line acceleration as the year progresses.

Other News

Microsoft Partnership

CrowdStrike announced a significant partnership with Microsoft, allowing organizations to purchase the Falcon platform with Microsoft Azure consumption credits. This is significant because Microsoft’s E5 solution is one of CrowdStrike’s biggest competitors.

For Microsoft to offer their biggest cybersecurity competitor a space in their marketplace is further validation that some customers prefer CrowdStrike as a “best-of-breed” vendor over Microsoft, which is often chosen simply because of convenience or cost consolidation to a single vendor.

Amazon Web Services Growth

In related news, CrowdStrike announced it has generated nearly $1.5 billion in total contract value on the AWS Marketplace, up 50% year-over-year. Success on AWS could potentially foreshadow significant contributions from the latest Microsoft partnership as well.

These cloud market partnerships are an efficient way to generate sales for CrowdStrike, as they allow the company to leverage Microsoft and Amazon’s sales to cross-sell their products at a lower commission fee than the larger overhead of managing a direct salesforce.

Memorandum of Understanding (’MOU’) with Saudi ARAMCO

CrowdStrike signed an MOU with the Saudi-owned ARAMCO. While it’s just an initial proposal, it elevates CrowdStrike from just a software vendor for enterprises to a trusted national security partner for mission-critical assets.This could set the stage for future multi-million dollar deals in sovereign markets leading to long-term contracts driven by sticky infrastructure-level integration.

Perplexity Comet

CrowdStrike and Perplexity also announced a partnership. This is what Mark wrote about it in the Overview Of The Week:

Earlier this week, CrowdStrike issued a press release announcing a strategic partnership to integrate the CrowdStrike Falcon platform with Perplexity’s Comet Enterprise.

If you’re not familiar with Comet Enterprise, it is Perplexity’s enterprise-focused AI browser that allows employees to research, analyze information, and interact with AI directly within a browser interface while giving organizations administrative controls over security, data usage, and governance.

Source

As companies increasingly incorporate AI tools into everyday workflows, CrowdStrike is positioning Falcon to secure these emerging environments through partnerships like this one. The collaboration also highlights the value of CrowdStrike’s recent acquisition of Seraphic, a browser security company, which provided the browser-layer runtime protection needed to integrate Falcon directly into AI-driven browsing platforms such as Comet Enterprise.

The AI “Threat”

A modern-day analysis would not be complete without addressing the AI “threat” that hangs over each SaaS company.

Last month, Claude announced the release of Claude Code Security:

Source: X

Like clockwork, CrowdStrike sold off 7% by the end of the day.

A couple of days later, George Kurtz fired back his own rebuttal:

Source: X

Pretty ironic, right? Even Claude agrees it cannot create CrowdStrike.

To press further, Anthropic (Claude) is a customer of CrowdStrike:

Source: Stockscurrent

If the capability were there, Anthropic would have built it in-house but clearly, that isn’t the case. Let’s take a deeper look at why:

Claude Code analyzes the code that it created itself to find vulnerabilities, functioning much like an automated application security testing (’AST’) tool. It flags issues and suggests fixes. This is helpful, and it will likely become an efficient, low-cost alternative for developers looking to ship cleaner, more secure code.

One market fear is that if code ships with zero vulnerabilities, the need for an expensive platform like CrowdStrike vanishes. But remember: 80% of all breaches are not malware-based. So even if Claude fixes every code vulnerability, a stolen identity or a human error (like clicking a phishing link) can bypass secure code instantly.

In contrast, CrowdStrike monitors in real-time, living deep within the operating system to block malware execution and track hackers moving through the network. While Claude proactively analyzes and asks for approval, CrowdStrike reacts in milliseconds to stop active threats.

George Kurtz highlights this distinction perfectly:

Here’s what you have to remember, is that what customers want is real-time prevention. You have to be in line, you have to be able to get the data in milliseconds, and you have to make a decision. That’s not the case with an LLM.

In short: Claude may allow developers to build more securely, but CrowdStrike is still needed to operate securely.

The Long-Term AI Moat

But let’s take it a step further. What if AI eventually builds a competing security platform?

CrowdStrike maintains a few distinct advantages over an alternate AI-built platform or other competitors for that matter:

1. CrowdStrike is a Net Data Creator. The platform generates real-time proprietary telemetry from approximately 2 trillion objects and over 1 trillion security events daily. These create unique patterns and security rules that don’t exist anywhere else on the internet. So any new security platform that may be spun up, no matter how advanced, will be behind the ball due to its lack of learned real-time customer data insights.

2. This usage data enables a technique called “Cyber-Reinforced Learning from Human Feedback” (’RLHF’). Breaches in real time produce expert-labeled data as a by-product of operations. This closed-loop system creates a flywheel: more events occur, which are validated by their experts, lead to platform improvements that no other competitor has, lead to more product adoption and additional events.
   

3. Lastly, reliable Cybersecurity hinges on deterministic outcomes. You can’t simply be right 90% of the time or suffer from occasional hallucinations. Stopping adversaries requires both efficacy and precision; you don’t get a second chance if it fails. Would an organization risk its infrastructure security to save money on less-tested alternative solutions? Cybersecurity is non-negotiable.

AI as a Tailwind

To flip the script, rather than a threat, AI and the “agentic workforce” represent a massive new market for CrowdStrike.

AI adoption means a new attack surface that must be secured. As it grows, CrowdStrike becomes an even more necessary layer of protection for the AI stack. We are already seeing this reflected in the growth of the modules (Identity, Endpoint, AI-DR, etc.) mentioned earlier in this article. CrowdStrike is set to benefit from this AI revolution, not be replaced by it.

Conclusion

With that, we conclude with another record-breaking quarter for CrowdStrike.

The company has successfully navigated the 2024 outage, turned its recovery program into a growth engine, and is now crossing the threshold into meaningful GAAP profitability. Module adoption and growth prove that AI is a tailwind for the business, not a threat, and with the $5 billion ARR milestone in the rearview mirror, the path toward its $10B ARR goal by 2030 looks to be on track. Remember that the goal was initially for $10B in revenue in 2032.

Thank you for reading! If you enjoyed this article, please give me a follow @ZackStacksKap. I often provide additional commentary on this and other Potential Multibagger stocks and would love to hear from you on X.

With that, I pass it over to Kris for the Selling Rules, the Quality Score and valuation.

You are missing the most valuable information!

If you want to continue reading this article, go to this page.

Subscribe to the annual plan.

Read more

This week, you can expect two articles from Zack. This is the first one. Of course, as always, I will update the Selling Rules, the Quality Score and look at the valuation. But first, the word is Zack’s.

Hi friends, Zack here, covering TransMedics’ (TMDX) Q4 2025 earnings.

In the current “SaaSpocalypse” environment, few Potential Multibagger stocks have been spared a significant selloff. TransMedics, however, has remained pretty resilient over the past few months. It dropped with the total market, but all in all, not too much.

The reason is simple: its competitive advantage isn’t in software but in the physical organ-preservation device the company has pioneered. When you factor in the extensive FDA approvals required to reach this stage, you have a company that is remarkably “AI-proof.” Knock on wood. Hopefully, our friend Claude isn’t reading this…

How was the performance to conclude FY2025? Let’s find out.

The Financial Numbers

• Overall Revenue: $160.8M, a 32% increase year-over-year, beating estimates by 4%.

• Net Product Revenue: grew 34% year-over-year, driven by organ system sales.

• Service Revenue: grew 39% year-over-year, reflecting the continued scaling of their logistics and aviation wing.

• International Revenue: As you may recall from last quarter, TransMedics began its international expansion into Europe via Italy. International revenue grew 24% YoY, and though it still only represents 3% of total revenue, management expects more meaningful volume in the second half of 2026. Since Europe accounts for roughly 45% of global transplant volumes, the OUS (outside the U.S.) opportunity remains massive.

• Gross Profit: $93M, a 29% increase year-over-year.

• Gross Margin: 58% versus 59% last year. Slight compression in margin was due to higher clinical service expenses as they expanded their organ care program along with increased freight costs with their growing aviation and logistics fleet.

• Operating Income: $21M versus $8M last year.

• Operating Margin: 13% versus 7% last year. Strong signs of operational leverage are being exhibited here, with expenses growing more slowly than revenue.

• Diluted Earnings Per Share: $2.62 versus $0.19 last year. Note this includes a significant one-time tax benefit; without it, diluted EPS would be $0.53, still almost triple that of last year.

Zooming out on FY 2025, we see a consistent trend of rising revenue growth and increasing operating margin:

A bit of a chart crime in that it shows growth slowing to 37% but I think most investors would take 30%+ growth in a heartbeat. FY2025 operating margin jumped significantly from 8% to 18%, indicating TransMedics is well on its way to the stated long-term goal of 30% operating margin by 2028.

This was also the first year finishing with positive cash flow, leading to an increase in the company’s cash position without taking on additional debt:

If you want Fiscal AI at a 15% discount, you can do that here.

In FY2025, TransMedics has successfully transitioned from the “startup” cash-burn phase to a “self-funding growth” phase, which is excellent to see.

Guidance

For 2026, TransMedics expects roughly 22.5% growth at the midpoint. While this looks like a stark deceleration, management has a history of guiding conservatism or call it serious sandbagging if you like. At this time last year, they also guided for 22.5% and ultimately finished the year at 37%. I have high confidence that they will beat this guidance again.

Organ Transplant Numbers

In FY2025, TransMedics completed 5,139 U.S. OCS transplants, a 38% increase year-over-year. This accounted for 26% of the total U.S. transplants, up from 20% in 2024.

CEO Waleed Hassanein made a powerful point on the call:

We strongly believe that the OCS NOP once again played a key role in driving overall liver and heart market growth due to the increased use of DCD and DBD donors in the U.S.

Since 2022, U.S. national transplant volumes for liver, heart and lung grew at a rate of 25%, including OCS NOP transplant volume. Without OCS volume, national volumes for the same organs would have declined by approximately 1% over the same period.

TransMedics is taking market share by expanding the total addressable market by making more organs viable. But in this context, that sounds cold, as the reality is that TransMedics helps save more people’s lives.

TransMedics U.S. market share by organ continues to confirm this trend:

Source: TransMedics Q4 2025 conference call transcript

The company also shared revenue by organ for the quarter:

Source: TransMedics

Putting these data points together, the liver is doing very well with both revenue growth and market share. However, the heart is growing at a slower pace and lung is declining (although at low volume).

The active trials for the heart and lungs aim to address the shortage of available organs for transplantation. Before we jump into how that works, I think it’s helpful to understand the two types of donor states:

1. Donation after Brain Death (’DBD’)
   In DBD cases, the brain is no longer functional, but organs are maintained within the donor’s body because the heart is still beating and blood is circulating. This makes it much easier to keep the organ in good condition.

   Because this is the “ideal” scenario, doctors often default to the lowest-cost preservation method: static cold storage (essentially a high-tech cooler with ice). In their eyes, the incremental improvement of an advanced system doesn’t always justify the extra cost (over $100,000) or the perceived risk of changing a “working” process.

2. Donation after Circulatory Death (’DCD’)

   DCD occurs when the heart has stopped and blood is no longer circulating oxygen. In this case, organ preservation is extremely time-sensitive; if they aren’t preserved quickly, they deteriorate and must be discarded.

   Historically, DCD organs were discarded at a much higher rate because cold storage couldn’t keep them viable. Transplant centers are much more open to using TransMedics here because, without it, the organ would likely be unusable.

Tying it all together, TransMedics’ Organ Care System (’OCS’) extends the life of these organs, allowing them to capture a large percentage of the DCD market.

As TransMedics builds more valuable clinical data, they hope to prove that the OCS drives better outcomes in all organs (both DCD and DBD) and also saves costs in the long run due to fewer complications and logistical costs. Convincing stakeholders that OCS is the new “standard of care” is the primary goal for these trials, which we will review next.

OCS Heart ENHANCE Trial

Source: TransMedics

The heart trial is split into two parts.

• Part A is designed to showcase how OCS can safely exceed the conventional time and distance limitations by maintaining a DBD heart past 4 hours.

• Part B is to demonstrate the superiority of the OCS system over conventional cold storage within 4 hours.

Founder and CEO Waleed Hassanein noted that progress in Part B hit a snag: the makers of a competing static cold storage box are refusing to allow their technology to be randomized against the OCS. He did express confidence that they could bypass this roadblock, but wasn’t really clear on how that would be done.

While this roadblock seems like a negative, I actually view it as a positive as well. Why would a competitor try to stop the trial from happening? Most likely because they know the OCS device is superior and should the data come to light, that would eat into their market share.

OCS Lung DENOVO Trial

Source: TransMedics

This program is aimed at revitalizing the U.S. lung transplant market as a whole. In general, lung volumes have been small. The main issue with lungs is that they are susceptible to edema (fluid buildup) when put them on cold storage. This trial will try to demonstrate that by utilizing OCS, lung transplants have a higher survival rate and fewer complications.

The last piece that was mentioned was not necessarily a trial but a publication.

The Liver Registry Publication is an upcoming peer-reviewed paper that will detail real-world clinical data from TransMedic’s more than 9,000 OCS liver transplants. Through this publication, they hope to demonstrate “unequivocal, drop the mic, statistical superiority,” as stated by Waleed Hassanein. Although liver already makes up a large proportion of market share, this data would be the final stamp of evidence to overcome any lingering hesitations, validating the economic and clinical value compared with competing organ preservation methods.

This quote is from another conference last December but I think it demonstrates Waleed Hassanein’s confidence in their product:

OCS machine perfusion based on 9,000 livers we have done, we are convinced that it should be malpractice if the livers are not put on the OCS. I mean, we are talking about significant survival difference, significant improvement in post-transplant complication rate.

Bold words. So basically, he states that, with this evidence, OCS should be the standard of care in any liver transplant scenario.

TransMedics’ roadmap for continued market share expansion is highly dependent on the success of these trials, which are currently expected to take 12 to 18 months to enroll. The liver publication should be out by the 2nd half of this year.

OCS Kidney Updates

The team also shared updates on the OCS Kidney device, which is currently in development. With over 20,000 kidney transplants performed annually in the U.S., the kidney market represents a huge opportunity to expand TransMedics’ TAM. Over 8,000 kidneys are thrown away each year due to low viability from cold storage. This represents “low-hanging fruit” for TransMedics to capitalize on as they apply their technology to save these otherwise wasted organs.

Source: TransMedics

The team plans to unveil the OCS Kidney device in early 2026 and initiate a clinical trial by early 2027, with the goal of proving statistical superiority over traditional cold storage.

Competition

Competition was a hot topic in the past, but talk has cooled significantly as TransMedics continues to grow its volume and market share. That sentiment was reflected in Waleed Hassanein’s blunt response to competition during the call:

Listen, we’re very cognizant of everything that moves in the field of organ transplant. We are not seeing competitive dynamic impacting our ability to execute in 2026 and beyond, and I will leave it at that.

He didn’t waste his breath. This should inspire confidence that there isn’t much standing in their way.

Further evidence of weak competition can be seen when looking at Xvivo, one of their closest competitors in terms of scale. This was shared on X:

As you can read from his last sentence, it’s a different mix of transplant organs and they serve in Europe, but if you read the rest of the X post you can see Xvivo is clearly struggling while TransMedics thrives.

In fact, this revenue chart tells the whole story:

There may have been signs of potential competition leading into 2022, but since then, it has become clear who the “top dog” is.

It’s also worth noting that another competitor, OrganOx, was acquired last year and no longer provides public financial data. Based on the most recent data, it appears their revenue is much closer to Xvivo than to TransMedics, and their technology only covers the liver and lacks the integrated logistics (aviation/transport) that make TransMedics so unique.

Government Policy Alignment

TransMedics is also moving toward becoming a more integrated partner with the U.S. government. As Waleed Hassanein noted:

Specifically, we are exploring if TransMedics can be a more integrated contributor to the national transplant ecosystem in the U.S.

The government is currently working to modernize the national transplant program by increasing organ utilization and upgrading infrastructure. If this partnership comes to fruition, it will embed TransMedics even further into the national infrastructure, creating a massive moat around their business.

Conclusion

TransMedics continues to impress with consistent executional excellence. On top of the usual KPIs, key things to watch are the clinical trial results and the timeline for the new OCS systems. These milestones are crucial to reaching their goal of 10,000 annual transplants by 2028 and 20,000 to 30,000 by 2030.

If you enjoyed this article, please give me a follow @ZackStacksKap. I often provide additional commentary on this and other Potential Multibagger stocks and would love to hear from you on X. Thank you for reading!

With that, I pass it over to Kris for the Selling Rules, the Potential Multibaggers Quality Score and the valuation.

Hi Multis

Kris here. I think we can agree that Zack did a great job with the analysis. Now, we update the Selling Rules, the Quality Score and the valuation to assess if the stock is a buy at this point.

The Selling Rules

I don’t just follow up on the Selling Rules I have made in the past. I also update them regularly and that’s why I added a new Selling Rule.

The new Selling Rule is that I want the medical trials to go right. If TransMedics can’t expand OCS indications, revenue growth will almost certainly start to slow.

The Potential Multibaggers Quality Score

If you want to continue reading this article, go to this page.

Subscribe to the annual plan.

Read more

How are you doing? I hope you are not too stressed out about the markets. This week, I got a reminder that the markets are not the most important thing in my life.

Our daughter Romy (she’s 11) broke her little toe from jumping on a trampoline.

Of course, that’s nothing to really worry about, but every Multi parent will know that seeing your child in pain really affects you deeply and if it were possible, you would take over the pain without any hesitation.

Zack, who writes these great articles for you, is also very sensitive and he felt I was a bit less connected this week.

So, yes, I am OK. And I was indeed a bit less active this week. Next to spending a bit more time with our daughter, I’ve been working a bit more behind the scenes for long-term initiatives in the last few weeks. The results may or may not become visible towards the end of the year or early next year. We’ll see. I’ll update you asap if something starts to get more concrete. IF because it’s not certain, as many things in life and investing.

On top of that, I’ve been researching much more this week, as there are more attractive stocks with this pullback. There may be crystallizing something, but I’m not sure yet.

Having said that, let’s dive into the Overview Of The Week.

Articles In The Past Weeks

This is the third article this week.

In the first one, I gave you the five stocks I consider the Best Buys Now. Of course, because the whole market is down, these stocks are down as well and that’s why I see them as even more attractive. As always, I put my money where my mouth is and I bought these stocks, together with a few more.

The second one was a deep dive on Mercado Libre. I really invested a lot of time into the article. You can read it here.

Memes Of The Week

Scary times mean the number of memes shoots up, and we can appreciate them more.

This is the first one. I made it myself, based on this post.

Multi Julien did a big investment to beat the index this year.

And this one shows that AI is not always as productive as you would want it to be.

And Multi Julien, who lives in Belgium, where Meli is a brand of honey products, posted this one as well.

Interesting Podcasts Or Books

This week, I watched Drew Cohen’s deep dive on Mercado Libre. You can find it here.

The markets in the past week

While the Russell 2000 dropped the most the past few weeks, it now dropped the least, with 1.68%. The S&P 500 dropped by 1.90% and the Nasdaq by 2.07%.

Despite the alarm messages you hear and see everywhere, the S&P 500 is down just 6.77%. The Nasdaq is down 9.65% and the Russell 2000 is now down more than 10%. That’s a correction already.

That means there could be much more downside. Or not, of course. You never know in the markets. The stock market can drop hard and fast, 2020-style, or slower, 2021-2022-style. I don’t know which one is more painful. More about this later.

You won’t be surprised that the Greed & Fear Index dropped further, to 15.

Quick Facts

1. Bear Market Ahead?

I hear a lot of negativity from all sides, even with the main market indexes not even in a correction (a 10% drop) yet.

Sentiment is now at its lowest point since Liberation Day. But you don’t need graphs to see that all around you, sentiment is negative. You see posts like this one everywhere.

Well, to start with, this is utter nonsense. What if I picked the bottom of the 2008-2009 crash as my starting point and showed you these returns?

Of course, you’d call me out for cherry picking the bottom, right? At least, that’s what I hope. But picking the absolute top is not a problem all of a sudden? It’s as much cherry-picking as choosing the very bottom as your starting point.

Secondly, it assumes that you invested on one single day, the very top, and poured in all the money you had. You never invested a single cent in the markets anymore. Yes, in those circumstances, you’d have to wait for 13 years. But if you invested like “normal” people do, you would probably not throw in all the money you had on a single day. If you did, you could probably use the long waiting period to learn about how to invest better, and hopefully, you have some money to average down.

But back to today. Sure, there are reasons to believe that the current environment could trigger a bear market. I don’t have to name them, I think: Iran, expensive oil and gas, potential inflation or stagflation, a private credit crunch and probably some others.

But what if we looked at the data first? This is based on 105 5% pullbacks since World War II. Only 24.8% turned into a correction (so down 10%+) and 12.4% into a bear market (down 20%+).

Source

That means that 62.5% of all 5% drops are just that, 5% drops. But investors panic each time as if there will be a huge bear market. As you can see from the stats, chances are not that big, but of course, it’s not impossible.

Never forget that things can always change quickly. As fast as the conflict started, it can stop. Sure, there will still be some residual damage, but what I mean is that nothing is set in stone. There are no anchor points that are certain and that means it could change tomorrow.

Four years ago, Warren Buffett was asked about a potential World War. This was his answer:

Well, if you tell me all of that’s going to happen, I will still be buying the stock. You’re going to invest your money in something over time. The one thing you could be quite sure of is that if we went into some very major war, the value of money would go down.

I mean, that’s happened in virtually every war that I that I’m aware of. So the last thing you’d want to do is hold money during a war. And you might want to own a farm, you might want to own an apartment house, you might want to own securities.

I mean, during World War Two, you know, the stock market advanced. The stock market advances over time. American businesses are going to be worth more money. Now, dollars are going to be worth less. So that money won’t buy you quite as much, but you’re, you’re going to be a lot better off owning productive assets over the next 50 years, then you will be owning pieces of paper.

I’m preaching against everything you hear now. I know. But don’t forget that 99% of the information on the internet and in the media comes from traders, not investors. Everyone has their reasons to talk about the short term. Brokers want you to trade, the media want you to be hooked when things go wrong, using your strong fear emotions for interaction.

And don’t forget that many influencers just want to be seen as smart. Negativity makes you seem smarter. And you can repost a post later if a bear market comes. You add something like: “You see? I told you.” If the bear market never comes, you never talk about the post again, of course. That’s how most do it online.

I want to be the antidote and I want to show you that there’s still a sun behind the current clouds and it will always return. Sure, when it returns is impossible to know, but that it will is almost certain, unless the world is devastated and then we have bigger problems.

Often, when we least expect it, the market can go up.

Maybe we have already seen the bottom? I saw this chart today.

Negativity around events is always the biggest in the first few weeks. Then it’s seen as the new reality or a temporary hiccup.

That doesn’t mean I don’t see the potential for long-term disruption from what’s going on in the world. It means that we have always pulled through.

I don’t want to downplay what we are going through. If you look at what happened in the seventies, it’s pretty scary. A 5-month oil embargo in 1973 resulted in almost a decade of stagflation. That’s definitely a risk now as well.

Sure, the US has its own oil production now but it imports a ton of things from Asia. Probably, the device you are reading on is assembled in China. They have much higher oil prices, so the product becomes more expensive, and transport costs as well. That could cause inflation to spike if the Strait of Hormuz is not opened fast.

There’s also the bull whip effect. I made this years ago and this is the normal situation.

But it also goes the other way around. Even a small disturbance in the supply side, can cause big effects for the consumer.

Natural gas is essential for fertilizer production, which is an important input for food prices, as we all experienced in 2022. Helium plays a critical role in healthcare and chip manufacturing. Ukraine was already a bottleneck for neon, a critical material for chip lithography. This adds to the supply difficulties. And think about shipping and insurance costs, and so much more. If the problem keeps lingering, it could potentially cause a big shock. The longer it takes, the bigger the impact. Of course, Iran knows this.

That’s also the reason why President Trump has set an ultimatum for Iran to open the Strait of Hormuz. We will see how that ends.

2. Tech = Cheap?

Of course, I know. Many stocks are down much more from their highs and that can be painful. Especially tech is hit hard.

For a long time, the Mag 7 has dominated the markets, but now, they are all down more than the S&P 500 and the Nasdaq.

Microsoft, usually one of the less volatile stocks, is now down 29%.

Sure, you hear about the big investments companies must make for building out data centers and other AI-related costs and that could put pressure on the margins of some. But these companies have huge piles of cash and probably, they are spending it well to make sure they will have huge piles of cash in the future as well.

Many of the things you hear are stories and they are not all based on reality. I go further and say that most are not based on reality. Remember when Google was going out of business just about a year ago? Or Meta a few years ago? Tesal has probably been proclaimed dead over a dozen times and the story around Nvidia was initially that it had no moat. Now that it has proven it has, the story is that the moat will end (so less pricing power) or growth will drop or whatever other story.

But look at this graph:

As you can see, all of them except Amazon are expected to still be profitable on an FCF basis in 2026. Be critical when you hear these stories. Often, they are just someone’s very superficial opinion stated with too much certainty.

Is tech really cheap now? I would argue that many tech stocks are. But that doesn’t mean that they can’t get any cheaper, of course.

Do you want to know how to profit from the crisis as a long-term investor?

Join Potential Multibaggers now.

Go to this page.

Subscribe to the annual plan.

Read more

Mercado Libre (MELI) reported its Q4 2025 earnings at the ned of February. The company announced more investments and that made the stock drop like a bomb. It is now down more than 35% from its top.

So, let’s look at the numbers first.

The Numbers

• Revenue: $8.76B (+44.6% YoY), beating estimates by $300M

• EPS: $11.03, missing estimates by $0.41

• Operating margin: 10.1% (down from 13.5% a year ago)

• Full year revenue growth: 39%

• Full year income from operations growth: 22%

Marketplace:

• GMV: $20B (+37% YoY FX-neutral in Brazil, +35% in Mexico)

• Items sold growth: 45% YoY in Brazil (accelerating from 42% in Q3 and 26% in Q2)

• Unique active buyers: 83.2M (+24% YoY)

Mercado Pago:

• Monthly active users: 77.9M (~28% YoY)

• Credit portfolio: $12.5B (+90% YoY)

• Assets under management: $19B (+78% YoY)

• Nearly 3 million new credit cards were issued in Q4 alone

Regional Breakdown:

• Brazil: 54% of revenue

• Mexico: 22% of revenue

• Argentina: 18% of revenue

• Other: 4% of revenue

Advertising:

• Revenue grew 67% YoY (FX-neutral)

And here’s an overview from Mercado Libre’s earnings call slides:

Investing For Dominance

Let me repeat something I’ve said in my previous MELI articles. Every single time Mercado Libre has deliberately compressed margins to invest in the long-term opportunity, it has paid off. Every. Single. Time.

And each time, analysts doubted the company. Same story, over and over again. Sometimes, the markets are funny in their own way. Management was very transparent about the investments, which I always like.

In the shareholder letter, management quantified the margin impact of the investments (note: CBT = cross-border trade, 1P= 1st-party selling, where MELI is the merchant)

We estimate the combined impact of our strategic investments, which include the lower free shipping threshold in Brazil, CBT, 1P and the credit card, was equivalent to 5-6 percentage points of operating margin in Q4’25.

So, without the deliberate investments MELI made, the operating margin would have been around 15-16%. That’s strong, and it would have probably meant a beat on earnings, too.

CFO Martin de Los Santos added on the conference call that the company won’t hesitate to keep investing:

Our main focus is on capturing the large opportunities in front of us in commerce, fintech and advertising. And we will not hesitate to invest in order to capture those opportunities as we have done in the past, even if that puts some short-term margin pressure. We’re not trying to optimize short-term margin.

Now, if you’re a trader, you hate hearing this but as a long-term investor, this is music to my ears. And importantly, these aren’t vague promises. The results are already visible. Even with all these investments, full-year income from operations still grew 22%. The company makes the right choice to grow faster at the expense of higher short-term margins.

So, yes, analyst calls like this one are correct but myopic, only looking at the year ahead.

Source: Seeking Alpha

That was the reason Mercado Libre dropped more after earnings. Thank you to all traders and short-term focused analysts.

Commerce: Brazil Is On Fire

The free shipping initiative in Brazil that we discussed extensively last quarter continues to deliver beyond expectations. As a reminder, Mercado Libre lowered the free shipping threshold (from R$79 to R$19), and the results are remarkable.

Items sold in Brazil accelerated from 26% year-over-year in Q2, to 42% in Q3, to 45% in Q4. Overall, it was 43.2% (the number disappears for one reason or another when I download the chart).

Do you want such premium insights? Fiscal has you covered! Grab a 15% discount with this link.

The unique active buyers grew by 24% YoY, the highest number of new buyers ever.

New CEO Ariel Szarfsztejn was proud:

New buyers that have come to MercadoLibre since June when we launched the new value prop are buying more items across a larger number of categories with higher retention rates compared to cohorts of new buyers prior to that change.

You can see this here.

It’s very impressive to see this. Usually, when an e-commerce business grows quickly, the average number of items sold per user drops. That MELI can make both grow so strong is another testament to the strength of this company.

I think this is a point many miss. So, anyone saying that the margin compression is important is not looking at Mercado Libre as an investment but as a trade. As an investor, you think like a business owner, and as a business owner, I’d be delighted with these results.

The logistics network is absorbing all of this beautifully. Unit shipping costs in Brazil declined 11% during the quarter. More volume is a leverage, as fixed costs go down as a percentage of delivery costs.

Mercado Libre introduced a slow-ship network to fill idle capacity, which I like a lot. When you have trucks and warehouses not in use at any given time, you can ship lower-value items during periods of lower traffic at almost no incremental cost. It’s a brilliant solution and it’s one of the things that competitors without such a big delivery system simply cannot copy, unless they invest heavily for years. Mercado Libre has already done those investments.

Mexico is performing well too, with 35% GMV growth. Cross-border trade saw a 74% increase in FX-neutral GMV. Mercado Libre now operates fulfillment centers in both the U.S. and China to serve its Latin American customers. This is a direct copy of what Amazon does worldwide. Here’s the regional breakdown.

An update on first-party (’1P’) that flew a bit under the radar. 1P is when Mercado Libre is the seller. Amazon started mostly with 1P and then expanded to 3P. Mercado Libre did it the other way around.

CFO Martin de Los Santos confirmed that 1P is now profitable before allocating central and indirect costs. That means it’s still not fuly profitable, but it’s early and this is a meaningful milestone. Scale will do the rest.

Remember, 1P exists as a sort of customer service to fill gaps in selection and pricing where third-party sellers underperform. It grew by about 80% FX-neutral in 2025 and is clearly gaining traction, especially in categories such as supermarkets. It’s an addition to the core 3P model, definitely not a replacement and it’s a good thing to see that there is a path to profitability.

Mercado Libre also changed its shipping fee structure for merchants in Brazil in January, moving to a model that charges based on actual dimensions and weight rather than flat rates. It’s still too early to tell what the financial impact will be, but it shows the company continues to fine-tune its business consistently.

This is just about a third of this article. Do you want to know whether I see MELI as a great opportunity now or not?

Do you want the whole article and know what I bought with my own money earlier this week?

Go to this page.

Subscribe to the annual plan.

Read more

Yesterday, it was the 15th of March and experienced Multis know what that means: it’s Best Buys Now time!

On top of that, I’ll show you what I bought for my Forever Portfolio, my real money I invest in the stocks I write about. Every two weeks, I add to the account, sun or rain. I’ve been adding more recently because so many of our favorite stocks have sold off a lot.

But first, the Best Buys Now. As you know, I always put my money where my mouth is, so I’ll buy all of these stocks.

Best Buys Now

If you want to know what I consider the Best Buys Now and if you want to see what I bought today, go to this page.

Subscribe to the annual plan.

Read more

Sometimes, it’s good to remember what investing is about. It’s not about the daily noise. I felt inspired yesterday evening, and I wrote this short story. It may help you when you feel stupid sometimes.

A Tale Of Two Brothers

Thomas and Lucas were 26 and 24 and they were in tears. Their father had just died. But he left them something. His two sons both inherited $150,000 and some words scribbled on a legal pad paper: “Buy stocks of companies that make things you understand. Be patient.”

Thomas took the words to heart and he bought Amazon first, as he saw that his wife placed several orders per week. Over the years, he bought several more companies that he understood.

His brother Lucas had always been brighter than Thomas, excelling in school, college, and university, and he had a good job at Google.

He made $40,000 on his first trade. At Thanksgiving that year, he told Thomas that patience was stupid and that he would help him make much more money. Thomas was tempted but decided to stay the course, remembering their father’s words. Not following that advice felt like not listening to his father’s last wish.

Lucas had a gift for reading the market, or at least that’s what he thought. So, the next year, Lucas started with leverage, and he made even more money. That Thanksgiving, he told Thomas he should have listened to him because he would have doubled his money.

But then, things slowly started to turn for Lucas. A trade went wrong but he was sure he was right. He had a special gift, right? So, he doubled down. A margin call came and he sold his winners. Each decision made perfect senseto Lucas. He just knew he was right.

Then the market dropped 30% and Lucas really started panicking, seeing his portfolio lose much more than the market because of his leverage.

Thomas did not do much. When a stock fell 65%, he read the annual report and decided to add a bit to his position. For the rest, he just didn’t look at his portfolio that often.

His inaction looked stupid to his brother Lucas. He told Thomas that he would lose all his money if he didn’t buy protection. Thomas considered it but did nothing.

Skip forward to thirty years later. Thomas’ portfolio is worth over $4.5 million. Unlike his brother Lucas, he never made a single trade that would impress anyone at a cocktail party. But he could retire rich at the age of 56.

Lucas was still working. He had $139,000 left of the $150,000. After 30 years. He finally understood, much too late, that trading was designed to keep him playing, not to get rich.

Lucas had sounded smart, but his brother Thomas had compounded the money at 12% per year and that turned $150,000 to $4.5 million. He had felt superior to Thomas all that time. He followed the market much closer than his brother; he was active, and he knew things better than most other investors. If he was honest, he had been looking down on Thomas for his inaction and underperformance when things went great.

But in the end, he realized that his brother and father had been right. Patience is the secret ingredient to successful investing. Not intelligence, action or sophistication, but patience. He had sold his stocks when he should have bought.

I hope this story helps you protect yourself from the people who sound so smart and only share their successes.

Articles In The Past Weeks

This is the third article this week.

In the first one, we looked at the Global-e results.

The second one was about the most volatile stock we have ever held: Hims & Hers. You can read that article here.

Memes Of The Week

Multi Karan shared this one this week about Fintwit (FinX?). It’s funny because it’s true so many claim to be specialists in everything.

I found this one very funny.

Multi Market Sense posted this one.

If this one is true, it looks good for the market.

This shows again that you can find patterns wherever you look. But it’s not because you see them that they mean something.

Interesting Podcasts Or Books

For the third week in a row, I listened to an episode of The Intrinsic Value Podcast. This time, it was an episode about Duolingo.

The markets in the past week

As you can see, the Russell 2000 dropped the most again, with 1.79%. The S&P fell 1.6% and the Nasdaq 1.26%.

Even if it may not feel like this, the S&P 500 is not even down 5% from its all-time highs set earlier this year. The Nasdaq is down 7.34% and the Russell 20000 8.78%.

Those are very moderate drops and that can be viewed in different ways. The positive reasoning is that the indices are remarkably resilient, and that may mean something. The negative reading is that there is a lot more room to drop. You can choose your side, but whatever you prefer, your opinion won’t change how the market behaves.

The Greed & Fear Index dropped again. It was on the border of Fear and Extreme Fear last week and it dropped from Fear to Extreme Fear. Strange, with the S&P 500, on which this is based, is not even down 5% from its all-time high.

Quick Facts

1. Growth & Value Valuations

This week, Meb Faber posted this very interesting chart.

It shows that expensive stocks have become more expensive and deep value stocks have become even deeper value.

At the same time, if you look carefully, you see that expensive stocks became more expensive already in 2016. There was a reset in 2022, as we experienced, but then valuations of the most expensive stocks started to rise again. Deep Value has already become cheaper since 2010.

Of course, the way this is measured also matters. As you can see, it’s a blend of Price to Sales, Price to Gross Profit, and Price to Economic Book. I think you know I love Price-to-Gross Profit, but I’m not sure it’s really that applicable to value stocks.

Even with the caveats, this remains an interesting chart.

2. The 5% Drop In Perspective

And more about the 5% S&P 500 drop. This is a very fascinating overview from Charlie Bilello.

The message is short: this too shall pass. You can point out that there has not been a big drop since 2008-2009, but a big recession like back then only happens about twice a century or so.

During this period, there were some steeper drops. And don’t forget that this is the S&P 500. The Nasdaq dropped quite a bit more (almost?) each time.

3. Nasdaq bends the rules for SpaceX

By Kris

Multi SS linked to a very interesting article earlier today. It’s written by Keubiko, of whom I had never heard before, but the argument is convincing.

Nasdaq is trying to change the Nasdaq-100 rules to make it possible for a giant IPO like SpaceX to get into the index fast and at a rich valuation. The consequence for passive investors: they could end up buying massive amounts of stock at inflated prices while insiders get an easy exit.

As we all know, index funds have attracted trillions of dollars in investments now. Overall, that’s a good thing, but that now means indexes can move markets. And that’s what SpaceX wants. And it’s clear Nasdaq wants SpaceX, which is expected to be the biggest IPO ever, to list on the Nasdaq, not on the New York Stock Exchange. It’s even willing to bend the rules to obtain that goal.

One proposed rule would let very large IPOs enter the index after just 15 trading days. Another would let low-float stocks get index weights way of a bigger part of the total shares, not just those that can be traded.

But if there’s a tiny float and forced buying from index funds, that could cause the stock to skyrocket even before passive money gets involved. Market funds don’t buy based on a valuation but accept whatever the market gives them.

That’s how Nasdaq wants to get a better pitch to win big IPOs and the stock gets early index status, boosting the demand. Passive investors will be the victim.

The Return Of The Potential Multibaggers

Do you want to continue reading?

Do you want to focus on the fundamentals of growth stocks?

Take a subscription to Potential Multibaggers.

Go to this page.

Subscribe to the annual plan.

Read more

I’ll start with a Peter Lynch quote. I’ll come back to it later in the article, but it’s important to keep in mind when you look at Hims & Hers.

Never a dull moment with Hims & Hers (HIMS). That’s a prime example of an understatement.

First, a partnership with Novo Nordisk, then a legal battle, and now a partnership with Novo Nordisk again. And that is just one of the multiple developments at Hims & Hers.

If you want to hold this stock, you need an iron-clad stomach. Many Potential Multibaggers have that, but this is by far the most volatile stock I’ve ever covered.

Depending on how you count, this is the fourth or even fifth 30%+ drop in the last year.

Let’s first look at the earnings results that were published at the end of February but we hadn’t had time to cover yet.

The Numbers

Revenue for Q4 2025 came in at $618M, up 28% YoY, in line with consensus. EPS came in at $0.08, a beat by $0.03.

For the full year, revenue was $2.35B, up 59% YoY.

Adjusted EBITDA for 2025 was $318M, up almost 80% YoY, at a 14% full-year margin. Net income was $128M. Operating cash flow hit $300M.

Subscribers grew to over 2.5 million. Monthly revenue per average subscriber hit $83 in Q4, up 11% YoY. Annualize that and you’re approaching $1,000 per customer per year. When it comes to health spending, people simply don’t hesitate the way they do in most consumer categories.

The number of personalized solutions tells an even more important story. At the end of 2025, approximately 65% of subscribers, about 1.6 million people, were on personalized treatments. At the end of 2023, that number was only 600,000. Nearly a million net new personalized subscribers in two years. That number shows what Hims & Hers wants to focus on.

The Stock Crash

On February 5, 2026, Hims & Hers announced it would begin offering a compounded semaglutide pill starting at $49 per month, with subsequent months at $99 per month. Novo Nordisk’s FDA-approved Wegovy pill had launched barely a month earlier, on January 5, with a starting cash price of $149 per month.

On February 7, 2026, Hims & Hers then announced it would no longer offer the compounded semaglutide pill. Only 48 hours, but what chaos.

The problem was that Hims & Hers was caught between a rock and a hard place. Because of strong patents, they couldn’t use Novo Nordisk’s SNAC delivery system for the pill to work in the gut. But if they used another mechanism, which they did, they actually needed FDA approval. And the FDA was swift to react to that aspect. With good reasons this time, if you ask me.

FDA Commissioner Marty Makary said the agency will “take swift action against companies mass-marketing illegal copycat drugs,” and general counsel Mike Stuart of the Department for Health and Human Services said he referred the matter to the Department of Justice for investigation.

Novo Nordisk filed a patent lawsuit and the 10-K filing disclosed that the SEC has opened an inquiry into Hims’ marketing, disclosures, and regulatory compliance around compounded GLP-1 products. But of course, with the new partnership, Novo Nordisk dropped the lawsuit.

Still, the FDA, DOJ, and SEC are simultaneously investigating Hims & Hers. Andrew Dudum’s response on the call was short:

I don’t think we can share too much on anything ongoing, but continue to welcome their conversations.

The existential risk posed by this compounding battle has largely been mitigated by the new partnership, but the regulatory overhang hasn’t fully disappeared. But I don’t think the outcome will be much worse than a fine, if anything.

The Q1 Guidance”Miss”

Q1 2026 guidance is $600M to $625M in revenue, which is just 2% to 7% YoY growth, and quite a big miss versus the consensus of $652M. But the details are important and the numbers have to be adjusted.

If you want to read this article and take full advantage of Potential Multibaggers.

Go to this page.

Subscribe to the annual plan.

Read more

David wrote an article about Global-e (GLBE), but I added quite a bit, so you can see this article as a joint effort, although let me be clear about that, David did by far the most work.

Of course, as always, after the earnings analysis, I will look at Global-e’s Selling Rules, the PM Quality Score and the valuation to decide if Global-e is a buy now.

Hi Multis

David here.

Global-e’s stock recently experienced some weakness, in line with broader market trends, but moved about 15% to $36 following the earnings release. It’s down a little bit to $34.68 at the moment of writing.

So, let’s jump in.

The Numbers

Revenue

Revenue in Q4 was $336.7 million, an increase of 28% year-over-year. The two revenue streams, service fees and fulfillment services, contributed $160.9 million and $175.7 million, respectively. Service fees increased 37% year over year, while fulfillment services grew 20.6% year over year.

One nuance here: the fulfillment take rate came in slightly below expectations this quarter. That’s not a sign of weakness, but because average order values were higher than expected. That means more GMV per transaction, which was good for GMV. So the “miss” on fulfillment take rate is a good miss.

For the full year, revenue mix remained stable with 47% ($451.2 million) from service fees and 53% ($510.9 million) from fulfillment fees. Both segments delivered approximately 27-28% annual growth.

Global-e guided for its first $1 billion+ revenue year in 2026. For any company, that’s a significant threshold.

Gross Profit

Gross profit totaled $154.8 million, compared with $118.7 million in the same quarter last year and up from $99.6 million in Q3 2025. That’s a 46% profit margin.

Gross Merchandise Volume (the total dollar amount of everything sold on the platform) increased 37.8% year-over-year to $2.361 billion, bringing the total GMV for 2025 to $6.569 billion, representing 35% annual growth.

A number that puts the GMV in perspective: Global-e’s total annual GMV for 2020 was $774 million. The company did $1 billion in GMV in November alone.

Global-e delivered strong operating results across key metrics, supporting a solid financial foundation for continued expansion.

The result exceeded company guidance across key metrics, and investor sentiment appears particularly positive regarding the company’s forward guidance.

Current Results and Future Guidance

Global-e updated its full-year outlook last quarter to the following:

The improved outlook for GMV, revenue, and adjusted EBITDA was discussed in the previous analysis. Management provided several drivers supporting the higher guidance, which was useful for evaluating how these expectations translated into actual results.

The table below compares company guidance with actual reported results.

This was the first full year of GAAP profitability, with a GAAP EPS of $0.39. The amortization of the Shopify warrant, a recurring drag on GAAP results for years, is now essentially gone. That removes a structural GAAP headwind and makes the reported numbers cleaner going forward.

Cash generation remains strong, with $216.2 million in free cash flow generated in Q4.

A portion of this cash is expected to be returned to shareholders. As highlighted previously, Global-e authorized a $200 million share repurchase program.

As of the end of Q4 2025, approximately $72 million in shares had been repurchased, representing about 1.8 million shares. The company indicated that repurchases are expected to continue under the existing authorization.

Management confirmed on the call that the company has continued buying in Q1 2026, with $128 million of capacity still remaining on the program.

Alongside year-end results, Global-e also released its 2026 outlook:

The company expects GMV growth of 28.6% to 34%, revenue growth of 25.8% to 32%, and adjusted EBITDA growth of 30% to 43%.

The adjusted EBITDA guidance, $259M to $284M, at a 22% margin at the midpoint, would give Global-e a Rule of 50 for the full year. That means revenue growth plus EBITDA margin exceeds 50. Usually, The Rule of 40 is seen as a benchmark for great companies..

Also important: Q1 2026 guidance calls for GMV growth of 38.8% at the midpoint. That’s an acceleration.

The outlook indicates that management expects the growth momentum achieved in 2025 to continue.

Growth

Global-e delivered growth across all key metrics in 2025, and the projected growth rates are not extreme.

The strong GMV demand is currently increasing and is even ahead of the most optimistic projections. This strong performance reflects both new merchant additions and growth from existing merchants, with results exceeding historical averages. Management continues to indicate a strong merchant pipeline.

The net-dollar-retention rate for 2025 came in at 122%, an improvement over 2024. Gross dollar retention was 96%. These numbers tell you two things: existing merchants are growing on the platform, and very few are leaving. The 122% NDR means that even without signing a single new merchant, revenue would still grow by 22% from the existing base alone. Very strong.

And that will probably continue. President and co-founder Nir Debbi said on the call that it “looks stronger today than it did in the same period in 2025.” 2025 was already a record year for growth in new merchant contributions, so that statement deserves attention.

Another tailwind can be found in foreign exchange (’FX’). As Global-e is in the middle of international transactions, the FX part of its business cannot be underestimated. Global-e collects revenue globally, but reports in USD.

With a weakening dollar, it can report higher revenue and profit in USD.

The table above illustrates Global-e’s geographic revenue mix. Approximately half of the revenue is generated in the United States, while the remainder is derived from international markets. As a result, a meaningful portion of revenue is affected by foreign exchange movements.

The partnership with Shopify continues to deliver strong results. The collaboration supports continued expansion of the merchant base and geographic coverage.

More specifically: Managed Markets 2.0 is now live. The critical upgrade in this version is full integration with Shopify Payments, replacing the separate payment rails used in the previous version. This means merchants can now run their international store exactly the same as their domestic store; they have the same processes, the same financial flows, and no learning curve. CEO, Chair of the Board and co-founder Amir Schlachet called it the biggest change. Management expects this to have an even bigger effect in the back half of 2026.

And one more thing, Global-e has already integrated the Universal Commerce Protocol (’UCP’), co-developed by Shopify and Google. As agentic commerce evolves, the complexity of cross-border infrastructure only increases and that is exactly what Global-e is built to handle. The company is more than ready.

Tariffs as a lead-generation mechanism

And then... we have to talk about tariffs. It was by far the most discussed topic on the call and one of the things I (Kris) am most asked about in the context of Global-e.

When the tariffs were introduced in 2025, in the short term, this created some pressure on US volumes. But as Amir Schlachet said on earlier calls, the tariff complexity had a positive effect on the pipeline in the midterm and that is now showing up in the Q4 numbers and 2026 guidance.

The EU de minimis removal is the next wave. Until now, small packages into Europe could pass customs without an import tariff. That’s about to end. When that comes into effect, it adds another layer of trade compliance for merchants, which is another boost for Global-e. Companies that previously shipped small parcels into Europe without customs handling will now need what Global-e does.

Global-e also launched its duty drawback offering for U.S.-based merchants in Q4. It allows merchants to reclaim import duties on internationally exported goods, and reclaim certain tariffs on returned goods. Initial adoption among the first group of merchants is described as “very high,” and it’s now being rolled out to all eligible U.S. clients. It’s embedded in the 2026 guidance, though management took a conservative approach given it’s still a new offering.

Saying to merchants: “Hey, if you just take this subscription, you will earn it back multiple times through claimbacks, where we do all the complicated stuff for you” is a pretty good selling point.

Or, in other words, the tariffs are an outstanding lead-generation mechanism for Global-e. Back to you, David!

Costs

One aspect I always look at is the costs. Not only the costs to generate revenue, but also the other costs that a company chooses to make, such as sales and marketing, and general and administrative costs.

The table below presents these costs as a percentage of gross profit. The left side compares Q4 2024 with Q4 2025, while the right side compares full-year 2024 with full-year 2025.

Cost of revenue remained stable at approximately 54% in both 2024 and 2025, with limited variation on a quarterly basis. These costs are typically more difficult to optimize and include items such as payment processing fees, shipping, and logistics.

Operating expenses declined relative to gross profit compared with both the prior year and the comparable quarter, contributing to improved profitability.

More numbers:

• R&D dropped from 9.2% to 8.5% of revenue in Q4.

• Sales and marketing fell from 11.3% to 9.4%.

• G&A from 4.1% to 3.2%.

You see that everything moves in the right direction. Management is guiding for this to continue.

AI and Other Improvements

Global-e is increasingly using artificial intelligence to improve growth and operating efficiency.

One example is a large language model used in customer support, where chatbot interactions are reducing the need for human intervention.

Another example is an internally developed AI tool designed to improve sales efficiency. The tool analyzes customer characteristics, product attributes, and geographic priorities to estimate conversion potential before sales outreach. This helps improve demo targeting and conversion rates while supporting more efficient sales and marketing spending.

But the AI story at Global-e goes considerably deeper than these two examples.

Vibe coding is already having a measurable impact. Entire features and even entire subsystems are designed, developed, tested, and deployed through AI-assisted coding. The efficiency gain is already visible in the lower R&D spend as a percentage of revenue in 2025.

Global-e also launched a full-site localization offering. It’s an in-house LLM-based service that translates entire merchant websites into multiple languages, including text embedded in graphics, a common problem for merchants.

Amir Schlachet made a good point on the call about this. Scale and AI are not separate advantages but they compound each other. Data fuels AI models and Global-e has unique, proprietary transaction data across 200+ markets that no competitor can even dream of.

The bigger Global-e gets, the better the models get. That’s a self-improving moat over time.

Conclusion

Global-e posted a strong quarter and 2026 looks to be strong as well. Does that make it a BUY?

Let’s find out with the Selling Rules, the PM Quality Score and the valuation.

If you want the full advantage of Potential Multibaggers.

Go to this page.

Subscribe to the annual plan.

Read more

It’s Sunday again and that means time for your weekly dose of news in the Overview Of The Week!

I’ve noticed that some Multis get a bit nervous from these markets and I understand that. But the equanimity I talked about last week can really make you stand out from the crowd.

As they say, staying calm when everyone is losing their mind, you will win in the end.

As you probably know, I have a background in linguistics and literature. I have never shared a poem before, but this time, I think it’s very useful. It was written in 1910 by Rudyard Kipling and it’s a father talking to his son. But I think it’s a useful poem for investors as well.

Try to be this and be an example for others.

Articles In The Past Weeks

This is the third article this week. The two previous ones were about two stocks that dropped hard after they announced their earnings and I put a lot of effort in them. I was very happy to see that was appreciated a lot.

In the first article this week, I dove deep into Duolingo’s earnings.

The second article was also a deep dive, this time into Sea Limited’s earnings.

Memes Of The Week

Multi SkQ shared this in our meme channel:

I also liked this one.

Interesting Podcasts Or Books

Last week, I listened to the Intrinsic Value Podcast episode about Constellation Software.

In that episode, I heard they also have one about Chapters Group. I listened to it this week. You can find it here.

The latest episode is about Duolingo.

The markets in the past week

What did the market do this week? I think most of you will know that stocks generally dropped quite a bit.

As you can see, the Russell 2000 dropped the most, by a whopping 4.07%. The S&P dropped 2.02% and, surprisingly, the Nasdaq dropped the least of the three, but it was still down 1.24%.

For months, I have been saying that the market felt like 2021, when growth stocks were already down 50%, 60% or more and the indexes remained at or near their top. Is this the point of surrender for the indexes? You can’t know, of course.

The Greed & Fear Index dropped. It was already in Fear but close to Neutral. It is now close to Extreme Fear.

Quick Facts

1. Most Protection Ever

This week, I saw this chart.

This chart shows that in the last 16 years, there has never been a period with more protection bought than now.

Historically, that has been a good counterindicator. In other words, if so many investors buy puts, the market usually goes up.

On the other hand, if this were a percentage of invested capital, it would probably look a bit different.

All in all, it’s like with all data points. It could be true in the short or long term, but I wouldn’t invest much money betting on it becoming true.

2. Higher Oil: Bearish for Stocks?

The most discussed theme of the weekend was the higher oil prices.

Ben Carlson always has interesting charts. This week, I saw this one on the relationship between higher oil prices and stock market returns.

As you see, the S&P actually goes up more with higher prices.

Of course, again, there are many nuances here. Maybe the speed of the climb plays a role, the impact on inflation if it persists and so on. Still, an interesting chart.

The futures, though, look bad for tomorrow, though.

So, tomorrow could be an awful day. We’ll see. But things can always change fast.

3. Nu Holdings

This week, I found this table, which compares Nu Holdings (NU) after 5 years in Brazil and 5 years in Mexico.

This is another data point that the company keeps crushing it.

The company also announced that it would first target Brazilians in its US expansion, which is coming soon.

4. This Is Getting Crazy

With some news, I have to take a step back and think: what on earth did I just read?

Cortical Labs, a small Australian biotech, just demonstrated something that sounds like it belongs in a sci-fi movie. They grew real human brain cells on a silicon chip and got those 200,000 cells to play Doom. It played the 1993 first-person shooter where you have to navigate a 3D environment, spot enemies, and shoot them in real time.

Even crazier: a developer with no neuroscience background did it in about a week. He used Python through a cloud API. The neurons figured out basic gameplay faster than traditional machine learning algorithms. It took just minutes, not hours.

Cool demo, but so what, you might think, but the implications over the medium-long term might be big. Your entire brain runs very efficiently, much more efficiently, like 50 times or more efficiently, than the most efficient chips right now. So, this could be a huge efficiency leap once this becomes applicable at an industrial scale, if that ever happens, of course. It’s still very early.

The CIA’s venture arm, In-Q-Tel, is an investor in Cortical Labs. The first 115 units shipped in 2025. And if you don’t want to buy the hardware, you can rent access to living neurons through their Cortical Cloud platform. It’s a subscription service where you deploy code to actual brain cells from your browser. The cells are grown from adult skin and blood samples and reprogrammed into neurons.

Read that last paragraph again if you need to.

Biological computing could eventually leave traditional silicon in the dust when it comes to energy efficiency and adaptability. Neurons learn from tiny amounts of data, adapt continuously, and consume almost no power compared to GPUs.

Also, this was just 200,000 neurons. The human brain has about 86 billion. So, you see the potential.

We’re at the absolute beginning of this. And honestly, I find it both fascinating and a bit terrifying.

Do you want a guide during these scary times?

Take a subscription to Potential Multibaggers.

Go to this page.

Subscribe to the annual plan.

Read more

This is such a period in which the punches don’t seem to stop when you are a growth investor. But that’s also often when the best opportunities are found.

Just look at this “unknown” company.

It just delivered record revenue of $23 billion, up 36% YoY, grew net income by 260% to $1.6 billion, and generated $3.4 billion in adjusted EBITDA. Would you think the stock would drop 25%?

Of course, it’s the old story: the stock price determines the stories that are being told. If you think that the fundamentals determine the stories, very often that’s not the case.

Just look at the stock price yesterday, up 8.23%. Fundamentals that suddenly improved in a day? Yeah, right.

I’ve seen this before with Sea (SE). I picked it at $54 back in 2020. I held and bought more when it crashed from $350+ to under $40. And I held when the market declared Sea a “Covid stock.” Every single time, the company has proven that great execution overcomes market sentiment over time.

The question is whether this time is different or just another chapter in the same book.

That’s what we will explore in this article.

The Numbers

Let’s start with the full-year numbers. In 2025, Sea generated $22.9 billion in total revenue, up 36%.

Adjusted EBITDA came in at $3.4 billion, a jump of 75.2%.

As you can see, Shopee did especially well, with more than a fivefold increase in adjusted EBITDA. This was mainly because of economies of scale. As you may know, Shopee raised its take rate, which positively impacted profitability.

I know that if I say take rate, many comment that it’s unsustainable, but the biggest part of that was driven by ads, which are growing fast. Therefore, this is sustainable and probably there’s a lot of growth left.

Garena’s adjusted EBITDA was up 38%, also very strong for a division that mostly runs on a game from 2017 (Freefire, of course). Monee’s (formerly Sea Money) adjusted EBITDA was up 43% YoY.

If something’s adjusted, it’s always interesting to know what. If you start with net income here, what’s added back (because it’s EBITDA) is the I (interest income or expense), the T (taxes) and the DA (depreciation and amortization). There’s also stock-based compensation and loan provisions.

Net income came in at $1.6 billion, versus $448 million in 2024. That’s a 260% increase. Very strong.

For Q4, revenue was $6.9 billion, up 38% YoY, beating consensus by about 6%. For revenue, that’s a lot.

Gross profit rose 36% to $3.0 billion. Operating income grew to $565 million, nearly doubling YoY.

The EPS came in slightly below analyst estimates. And the 2026 Shopee guidance mentioned ‘at least the same adjusted EBITDA’ despite 25% GMV growth guidance. That alone was enough for a 25% selloff.

So, what is the market actually worried about here? It’s not the growth because that was great. It’s the uncertainty about why continued growth in guidance doesn’t translate into greater profitability.

The market is always very impatient about an investment phase, even if it thinks it’s a good investment. The reason is simple: the market is not dominated by investors, but by traders.

Of course, that means we can focus on the crucial question: is this a temporary and intentional investment or a fundamental weakness in the business?

Let’s first dive into the different divisions.

Shopee

Shopee had a phenomenal 2025. I am sensitive to words and, to be honest, I myself wouldn’t expect the word “phenomenal” in an analysis of a stock that just dropped 25%. That also means that I use it very intentionally. Just look at why I use it.

For the full year, GMV came in at $127 billion, up 27%. As a reminder, GMV or gross merchandise volume, is the dollar amount of everything sold on the platform.

There were 400 million active buyers and 20 million sellers. For Q4, GMV was $36.7 billion, up 29% YoY from 4 billion orders, up 30%. Revenue came in at $5 billion for the quarter, up 36%, with core marketplace revenue (so, the countries in Southeast Asia) growing 50% YoY to $3.6 billion. That’s really impressive.

I have already mentioned this earlier in the article, but adjusted EBITDA for Shopee went from $156 million in 2024 to $881 million in 2025. Remember, this was a deeply loss-making business just a few years ago. The fact that Shopee generated almost $900 million in adjusted EBITDA while growing this fast seems to be completely ignored by the market.

This difference between GMV growth and revenue was mostly driven by ad revenue, which grew over 70% in Q4, with ad take rate up more than 80 basis points YoY. More than 20% more sellers are now paying for ads, and their average ad spend increased by more than 45%.

The overall take rate reached 13.5%, up from about 11.3% just two years ago. That’s a big improvement. And the great thing about ads is that it’s a self-fortifying machine. If more merchants run ads, those that don’t yet see them work also start using them.

But the market only looked at one thing. CEO Forrest Li stated that Sea aims to grow Shopee’s GMV by around 25% YoY, with adjusted EBITDA “no lower than 2025 in absolute dollar terms.” In other words, margins will compress. The market didn’t like that, to put it mildly.

But there’s a fundamental difference between a company that can’t generate margins and one that chooses not to because it sees larger opportunities ahead. If you are a shareholder of Mercado Libre, you have seen this play out there too: invest heavily, market panics, then the investments pay off nicely. Rinse and repeat. Sea’s investments have clear timelines and identifiable returns.

CFO Tony Hou was very explicit about this on the conference call (for context: VIP program is Shopee’s equivalent to Amazon’s Prime):

If you look at the fulfillment network, there will be a period of time we build it up, but there’s a clear investment cycle come with it rather than that is ongoing perpetual investment...

If you look at the VIP program, there’s a pure time that we will kind of educate the market, but as we get everything in place, I think the cost structure will be a lot better.

He also added something crucial: the 2-3% EBITDA of GMV margin target for e-commerce remains firmly achievable. And if you look at Q4 2025’s EBITDA margin at around 0.55% of GMV, that still improved year-over-year. The sequential decline from H1 to H2 was a conscious ramp-up of investments, not a competitive threat, which the market seems to assume. It’s funny how the market always seems to prefer the negative explanation. Of course, if you are an analyst, it’s better to be negative than positive. It makes you sound smarter, as you seem to warn investors. The funny thing is that many bring this as if bringing negative news is hard. It’s exactly the opposite. If you are positive, you get a lot more mud thrown at you in the market. Trust me, I know. :-)

So, where is Sea investing?

(If you don’t want to miss these articles, don’t forget to subscribe, for free or paid, if you want more).

Subscribe now

Logistics

SPX Express, the company’s fulfillment arm, now processes over 30 million parcels every single day, making it one of the largest e-commerce logistics providers in its markets. In cities like Bangkok and Jakarta, instant and same-day delivery already made up more than 10% of order volume by the end of 2025.

That’s not just a charity service, of course. Aside from appreciating the service and not churning as quickly, such buyers also spend around 15% more on average than before they started using the program. That’s a great flywheel: better delivery leads to more spending. This is a part of the Amazon playbook, of course.

The great thing is that buyers have the choice. In Indonesia, orders using the cheaper shipping option more than doubled in Q4. Fulfillment services, from warehouses closer to buyers, already handle more than 10% of orders in some markets. The goal is to double that by the end of 2026.

Shopee also expanded instant delivery into groceries, partnering with local supermarkets in Thailand to deliver fresh groceries in as little as 1 hour. This expands the addressable market.

Then, Brazil. CFO Tony Hou explained the Brazil logistics angle nicely:

We have been operating with a much more efficient logistics network compared to what’s available to our — to the other players in the market with much lower cost. And we are able to run the businesses profitably with sort of a much lower basket size.

Brazil was again the fastest-growing market for Shopee in 2025. Delivery times improved by about 1.5 days YoY in Q4, which is quite impressive.

Shopee Mall GMV more than doubled YoY, and more than 300 new brands were onboarded. That matters because Shopee Mall is the higher-end segment. It has branded sellers and, therefore, higher basket sizes. It shows Shopee is moving upmarket in Brazil, not just sticking to cheap goods, as it’s often portrayed wrongly. That’s just the foot between the door.

The next step in Brazil is rolling out fulfillment capability at scale in 2026, another investment.

The VIP Membership Program

This is Sea’s equivalent of Amazon Prime. Subscribers get better free shipping deals, daily vouchers, and exclusive discounts.

Total VIP members surpassed 7 million by year-end. That’s impressive because it’s more than double what it was just a quarter ago. In every market where it launched, members increased their spending by double digits after joining. In Indonesia, VIP members spend 30-40% more than before joining. In some markets, VIP members already contributed more than 15% of total GMV in Q4.

Tony Hou shared a very interesting detail. Historically, the biggest challenge for subscription programs in Southeast Asia was the low payment success rates. Most people don’t have credit cards for auto-renewal. Retention rates were only around 40%. But by working closely with Shopee and Monee to enable smooth payment processing, Sea increased subscription retention from 40% to 70% in Indonesia over just a few quarters. And probably, there’s more room for improvement. This may seem like a detail, but it’s fundamental.

It also shows that the SEA ecosystem works. It’s a structural advantage that competitors without a fintech arm will have a very hard time replicating. Such seemingly “small” details add up.

The Content Ecosystem

This was an investment that already showed clear positive results. Orders driven by YouTube content more than tripled in Q4 YoY. The Meta partnership, launched in October, already has more than 3 million affiliates who linked their Shopee and Facebook accounts. Content’s share of business is already more than 20% and growing.

AI across Shopee

This one doesn’t get enough attention. I mentioned it in the Overview Of The Week a few weeks ago.

Sea is deploying AI across its e-commerce platform in ways that directly impact revenue and cost. On the search and recommendation side, they rolled out multimodal search: users can now search with a picture plus a description, similar to how you’d use Gemini or ChatGPT. CFO Tony Hou said the higher ad take rate is partly a result of the AI improvements: the platform now understands products and what users are looking for much better.

On the seller side, Shopee built AI chatbots that sellers can customize for their own stores. This helps sellers reduce employee or service costs. At the same time, it improves upsells to buyers.

Shopee also introduced tools for sellers to create product videos, pictures and descriptions using AI. All of these come with a positive return on investment for the ecosystem.

As mentioned in the Overview Of The Week, Sea also announced a partnership with Google, which is still in development. There are no details yet, but it builds on what they’ve been doing together for years on Google Shopping and Google Ads. On top of that, Google’s AI would also help with game development for Garena.

Forrest Li framed the company’s AI philosophy pragmatically:

We try to focus on the applications and how to connect that fantastic technology to people’s daily life from every corner of the world... We’re not going to try to make some fundamental large language model breakthrough. We’re not going to build data centers.

In other words, Sea will be an AI applier, not an AI builder. That’s a good strategy, I think.

Taiwan and competition

One more highlight: Taiwan. GMV growth accelerated to double digits in 2025. Shopee’s automated locker network grew to over 2,800 locations, making it the only e-commerce player at that scale in Taiwan. These lockers run at 30% lower per-order costs while also functioning as last-mile hubs. This kind of infrastructure moat is an extra defense against competitors.

When you look at the competitive landscape, CFO Tony Hou noted it remains “relatively stable across most of our markets.” No dramatic shifts is good news.

The US recently scrapped the de minimis rule, which allowed packages under $800 to enter the country tax-free. That was the backbone of how companies like Temu and Shein shipped cheap goods directly to American consumers. Now that this loophole is closing, those players need to find growth elsewhere and Sea’s markets in Southeast Asia and Brazil are the obvious targets.

That’s what the market was afraid of and yes, competition got tougher. But SEA doesn’t seem to suffer. Shopee’s take rate keeps rising, order growth accelerated to 30% in Q4, and the active buyer base keeps expanding. Companies that are losing a competitive fight don’t show numbers like that.

One more thing on the guidance. Sea has a long history of guiding conservatively and then beating. At the start of 2025, they guided for around 20% GMV growth. The end result was 27%. For 2026, the 25% GMV target will probably be a floor, not a ceiling.

Monee

Monee, Sea’s crown jewel, as I like to call it, keeps delivering. This segment generated $3.8 billion in revenue for the full year, up 60% YoY. Adjusted EBITDA crossed $1 billion, up 43% YoY. The fintech arm alone generated over a billion in adjusted EBITDA.

The loan book reached $9.2 billion at year-end, up 80% YoY. An 80% loan book expansion sounds risky, right? I heard this as an explanation for the big stock price drop.

But there’s a very important number to look at here, the 90-day NPL ratio. NPL stands for non-performing loans or, in other words, loans that can’t be paid back.

The 90-day NPL ratio held steady at 1.1% at the end of Q4. That’s very strong and even an improvement from a year ago. For context, this is far more conservative than comparable operations like Mercado Pago, where similar delinquency metrics can run much higher. For comparison, the average NPL ratio across Asia was about 2.7% in 2024.

On top of that, remember that Sea is lending to consumers and small businesses in emerging markets, the riskiest segments. Traditional banks in those same markets, with their decades of experience and conservative underwriting, are running NPL ratios 2-3x higher.

But, also very important to acknowledge, a loan book growing so fast will automatically lower the 90-day NPL ratio. Loans need time to go bad and if you grow your loan book fast, this will naturally bring down the 90-day NPL ratio. Nevertheless, the numbers look very conservative.

The growth came from three drivers.

Do you want to see if Sea is a BUY here? Become a paying member of Potential Multibaggers!

(If you already are, just scroll past this)

What do you get as a member?

• ✍️ Multiple high-quality articles per week

• 📚 Full access to our entire library of deep-insight articles

• 🔎 Full investment cases

• 📊 Access to the Private Community

• 🎥 Regular Webinars

• 📈 The Best Buys Now every month

• ✍️ The Overview Of The Week each Sunday

• 🔎 Deep earnings analysis

Many Multis say that the community alone is already worth the price, but you get so much more.

Go to this page.

Subscribe to the annual plan.

Read more