First ASML, now TSMC.

If 🔗 ASML Q1 2026 left anyone skeptical, TSMC delivered a quarter that should put to rest any lingering doubts about the sustainability of the AI semiconductor cycle. Revenue hit $35.9 billion (+41% YoY) while gross margins surged to 66.2% from 62.3% QoQ, blowing past the high end of guidance by 120 basis points.

These results are additional evidence of a structural shift in how compute demand is evolving.

Financial Results : Mind blowing 🤯

Note: Charts below as shared from 🔗 SemiAnalysis Source

Revenue has now grown double digits YoY for eight consecutive quarters, with management guiding for that streak to continue into Q2 2026.

Advanced nodes still represent the backbone of the story, accounting for 74% of total revenue (slightly down from 77% in Q4). Within that mix, N5 drove 52% of incremental QoQ wafer revenue, reaching an all-time high contribution, while N3 declined 5% QoQ as the node transitions through its typical ramp dynamics.

Taken together: growth remains anchored at the leading edge where AI demand is concentrated.

The Agentic AI Inflection

Management’s commentary on the transition from “query mode” to “command and action mode” AI is the one of the most telling takeaways from this quarter.

"The shift from generative AI and the query mode to agentic AI and the command and the action mode is leading to another step up in the amount of tokens being consumed. This is driving the need for more and more computation, which supports the robust demand for leading-edge silicon." — Dr. C.C. Wei, Chairman & CEO

Dr. C.C. Wei explained that Agentic AI requires exponentially more token consumption and background compute than the generative AI applications we’ve seen so far. This is a step-function change that’s driving High-Performance Computing (HPC) to 61% of total revenue.

HPC growing 20% sequentially while entirely offsetting an 11% seasonal decline in Smartphone revenues tells you everything you need to know about TSMC’s transformation.

The company has structurally decoupled from consumer electronics cyclicality, and the market hasn’t fully priced this in yet. (In my opinion).

The Margin Story

A 66% gross margin and 58% operating margin at all-time-high in a capital-intensive semiconductor business is exceptional to say the least.

But understanding why this is happening is key.

TSMC is printing these margins because it holds a monopolistic position in sub-5nm AI silicon. When customers are begging for capacity and there’s no viable alternative, pricing power follows.

The interesting thing about N3 (3nm) is that it contributed 25% of wafer revenue this quarter while still being a drag on margins. Management confirmed that N3 margins will cross above the corporate average in the second half of 2026 as the node matures and depreciates. This means the current 66%+ gross margin, already exceptional, represents something close to a trough for the N3 era. Once that node fully matures and becomes a margin tailwind rather than a headwind, we could see even more operating leverage.

That said, let’s not get carried away. The N2 (2nm) ramp in the second half will temporarily dilute margins by 2-3%, and overseas fabs (Arizona, Japan, Germany) will add another 2-3% headwind initially. So the 66%+ we’re seeing now is likely the high-water mark for 2026, and investors should expect some normalization in the back half of the year. This doesn’t change the thesis though, it’s just the natural consequence of ramping cutting-edge nodes and geographic expansion.

CapEx and the Growth Pipeline

TSMC raised its 2026 CapEx guidance to trend toward the high end of the $52-56 billion range. Some investors might view this as concerning, but the reality is quite different.

And you may ask, but why ?

“A very simple answer is, the demand are very robust, especially from the HPC and AI applications. We try very hard to speed it up and pull in all the equipment as we can. Still, our supply is very tight... That’s why we are toward our high end of CapEx forecast.” — Dr. C.C. Wei, Chairman & CEO

Management is aggressively expanding N3 capacity globally (outside Taiwan), not because they want to (they wouldn’t do it a few years ago) but because they have to. The undersupply situation is big and customers with multi-year roadmaps need certainty that TSMC can deliver the silicon.

The beauty of this CapEx increase is what it means, sepcially coming from one of the more cautious players in the industry :

• Management has visibility into 2027 and 2028 demand that gives them confidence to invest at these levels. You don’t drop an extra few billion dollars into fab capacity on a hunch.

• The hyperscalers and AI accelerator clients are committing to volume that requires this build-out, and that commitment is worth more than any guidance number.

Capital intensity staying around 30% despite the absolute CapEx increase also matters. Revenue growth is outpacing CapEx growth, which means this isn’t a margin-destructive investment cycle. It’s the necessary infrastructure spending to capture what management believes is a multi-year AI hardware supercycle.

Btw, WFE companies love this, specially ASML and all their peers playing in the leading-edge field (KLAC, LRCX, BESI, ASM, AMAT, etc.).

Some data: Wafer shipments are at ATH. Advanced nodes are at 100%+ utilization.

The Moat

Another telling part of the conference call was management’s response to questions about competitor foundries and sovereign fab initiatives.

Here is CEO Dr. Wei take on competition and foundry “shortcuts”:

“There are no shortcuts. The fundamental rule of the foundry game never change. Our customers need the technology leadership, manufacturing excellence, and trust... It takes 2-3 years to build a new fab. No shortcuts. It takes another 1-2 years to ramp it up.” — Dr. C.C. Wei, Chairman & CEO

Technology, manufacturing excellence, and customer trust are non-negotiable requirements. Go ask Intel. Go ask Elon, too.

Intel’s foundry ambitions and various “Terafab” (I keep laughing) projects make for good headlines, but TSMC doesn’t view them as near-term threats. And frankly, the numbers support this confidence. When you’re printing 66% gross margins and growing HPC revenue 20% quarter-over-quarter while competitors struggle with yields and customer wins, the competitive gap is widening.

The fact that TSMC won’t gouge customers despite undersupply also matters for the long-term thesis. Management reiterated their partnership philosophy: they’d rather “grow together” with customers and capture their fair share of value over time than extract short-term rents that damage relationships. This is exactly the behaviour you want from a company with a sustainable competitive advantage.

"We always view our customer as our partners... we don’t change our pricing dramatically or something like that... fundamentally, is number one, our customer has to be successful. That’s our consideration, number one. We grow together." — Dr. C.C. Wei, Chairman & CEO

If you read my ASML earnings digest yesterday, management made a very similar point about how they run the business, how they work with customers, and how they approach pricing.

And honestly, I think this is one of the core reasons behind the long-term success of the advanced semiconductor ecosystem.

Collaboration-for-scale is what makes the leading edge so difficult to replicate, and why the winners tend to compound together over decades.

Upgrading Guidance

TSMC raised full-year 2026 revenue growth expectations to “above 30%” in USD terms, with Q2 guidance of $39-40 billion implying 32% YoY growth.

Q2 gross margin guidance of 65.5-67.5% suggests TSMC expects the current operating leverage to persist, even as they begin absorbing early-stage dilution from overseas fabs. The midpoint implies another 30 basis points of expansion, driven by high utilization and productivity gains.

The N2 and Advanced Packaging Roadmap

N2 entered high-volume manufacturing in Q4 2025 with good yields and will be a “long-lasting node” according to management. More importantly, A14 (1.4nm) development remains on track for 2028 volume production with a 25-30% power improvement. (Thanks ASML!)

The roadmap is bulletproof, and nobody else is even close. Period.

Advanced packaging (CoWoS) capacity remains the bottleneck for large AI accelerators, but TSMC is already hands-on, building a pilot line for next-generation advanced packaging processes. This looks like a solvable problem that will require time and investment.

Risks

Consumer softness in memory-sensitive PC and smartphone segments is real, but TSMC’s focus on premium tiers partially insulates them from this.

The more relevant risk is that a competitor (Samsung was specifically mentioned) is winning isolated design wins in the AI accelerator space. While TSMC expects to win next-generation business from these customers, it demonstrates that second-sourcing remains a factor even in leading-edge AI silicon.

Looking Forward

The company is executing flawlessly on the most important secular growth opportunity in semiconductors, with a competitive moat that appears wider than ever. The 66% gross margin and 40% ROE are evidence of supernormal returns from monopolistic control over sub-5nm AI silicon.

The CapEx increase shouldn’t be viewed as a drag on free cash flow but as validation of multi-year customer commitments. When such a cautious, conservative management tells you they’re confident enough to deploy $56 billion in capital (probably higher), they’re telling you the AI hardware cycle has legs extending well into 2027 and 2028.

I would even go as far as to say that TSMC’s profitability targets are conservative given what we’re seeing. The structural shift toward HPC dominance and the pricing power embedded in cutting-edge nodes shows, to this day, that TSMC can sustain higher profitability than the market currently underwrites.

The key question is whether investors are properly valuing a company that has become essential infrastructure for the AI era while simultaneously expanding its competitive lead.

I’d argue the answer is no.

But the geopolitical risk (China) is something Taiwan simply can’t shrug off. It’s likely to remain a persistent ceiling on the multiple for the foreseeable future, regardless of how strong the underlying fundamentals become.

Thanks for following along,

—Nikotes

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Portfolio Corner - 🗓️ Monthly Updates (Last Update: 29-Mar-2026)

Subscribe now

"We do not want EUV to be the bottleneck. So, I think I'd like to say that very strongly... If you compare a bit what we did in 2025 versus what we could in basically '27 with the at least 80 system, we more than double basically the total EUV capacity we can ship to our customer." — Christophe Fouquet, CEO

Keep them shipping Mr. Fouquet 👀

ASML just delivered one of those quarters that forces you to re-examine your priors. Not because the numbers were simply "good" but because they definitively put to rest the AI bear thesis that was hanging over the stock the past year.

Let’s walk through these Q1 2026 results.

Financial Results

Note: Charts below as shared from 🔗 SemiAnalysis Source

What China Cliff ?

For months, throughout the last year, the narrative surrounding ASML was relentlessly negative. One of the may concerns was related to China representing 36% of system sales in Q4 2025 and what was going to happen when export controls tighten and their tech WFE self-sufficiency and that revenue evaporates?

Well, we got our answer.

China’s share of system sales dropped from 36% to 19% in a single quarter, a big decline by any measure. And what did ASML do? They raised full-year guidance from €34-39B to €36-40B.

The thing is this shouldn’t have surprised anyone paying attention to the fundamentals and the real secular trends. The China business was always lower-margin, lagging-edge DUV demand. What’s replacing it is demand for high-margin EUV systems and advanced DUV immersion tools from Taiwan, South Korea, and the US, all driven by AI capacity expansions. South Korea’s share surged from 22% to 45%, Taiwan rebounded to 23%, and suddenly the geographic mix looks far healthier than it did a quarter ago.

Take a look:

And here, management commentaries on the ex-China DUV Rebound and how it scales with EUV demand:

“When it comes to immersion... we believe that for non-Chinese customer, the demand on immersion will scale with the demand for EUV because... there is a pretty clear relationship between the two.” — Roger Dassen, CFO

This is the perfect example of why focusing on quarterly noise rather than structural trends will get you in trouble. The market obsessed over China exposure when the real story was the unprecedented buildout of AI infrastructure globally.

Let’s move on.

The Memory Perfect Storm

Here’s where things get interesting from a long-term perspective. ASML has always been viewed primarily as a Logic play, the company enabling Moore’s Law through EUV adoption at the leading edge. But something fundamental shifted this quarter.

Memory customers. Specifically, DRAM manufacturers supporting High-Bandwidth Memory (HBM) production are now driving EUV adoption at a pace nobody anticipated last year.

“DRAM has been a bit the perfect storm for ASML because of course we have this capacity buildup, but... we have seen a major adoption of EUV in DRAM in 2025... If you are going to use more EUV layers, you are going to need less multi-patterning, and multi-patterning takes a lot of space also in the fab.” — Christophe Fouquet, CEO

Management explicitly stated that Memory customers are “sold out for 2026.” These companies are scrambling to add EUV capacity not just to increase output, but to save precious fab space by eliminating multi-patterning steps.

The math here is quite compelling, looking ahead, towards the next quarters : A single High-NA EUV exposure can replace 3-4 Low-NA exposures, saving a 10x factor in process steps for critical layers. When you’re trying to pack HBM production into existing and new fab footprints, this isn’t just a nice-to-have.

What’s remarkable is that Memory revenue accelerated 38% quarter-over-quarter and now represents 51% of system sales (essentially matching Logic at 49%).

Time will tell but this does not look like a temporary spike. Based on current industry dynamics and management commentaries, it seems long lasting. AI is contributing to the structural expansion of ASML’s addressable market. The company that enables the leading edge for Logic is now becoming indispensable for Memory as well.

Monetizing the Installed Base

One aspect of ASML’s business model that doesn’t get enough attention is their Installed Base Management (IBM). In Q1, the company recognized €2.5 billion from this recurring revenue segment, primarily software upgrades that boost throughput on existing tools.

The new NXE:3800E is not just an upgrade from the 3600D system but a new product which increases wafer-per-hour output from 220 to 230. This might sound incremental, but for customers desperate for capacity right now, paying for a software upgrade that delivers immediate productivity gains is a no-brainer. It’s also extraordinarily high-margin revenue for ASML, contributing to Q1’s impressive 53% gross margin.

As an investor, this is precisely the type of business model characteristic you want to see: razor-blade, the ability to extract incremental value from an installed base while simultaneously selling new systems. It creates a recurring revenue dynamic in what would otherwise be a purely cyclical capital equipment business.

See the blue revenue bars across the quarters in the chart below ? This is only going to go up as a percentage of total sales and it’s the driver behind the increasingly recurring nature of ASML’s revenue mix.

The Pricing Power Philosophy Question

During the call, management addressed something I found particularly interesting: whether ASML should be extracting higher prices given the evident desperation of customers to secure capacity. ASML’s CFO pushed back hard on this, emphasizing their “fair model” philosophy.

"Our model of pricing is not based on the squeeze that our customers find themselves in. That's not the way we do business. The way we do business is that we look at the value that we provide to our customers, generation on generation, tool on tool, and we take our fair share in that." — Roger Dassen, CFO

It’s a win-win healthy long-term relationship : ASML 🤝 TSM and co.

ASML prices based on the generational value and throughput of each tool, not based on supply-demand dynamics. They’re explicitly choosing not to squeeze customers when they have maximum leverage.

While some might view this as leaving money on the table, I think it’s actually brilliant long-term thinking. Semiconductors are a cyclical industry, and these same customers will have purchasing decisions to make during the inevitable down-cycles. Building trust through consistent, value-based pricing creates durable competitive advantages that can’t easily be replicated.

This said, don’t mistake fair pricing for leaving value uncaptured. If you don’t believe me, you just need to dig into their roadmap and pay attention to management comments :

When ASML introduces the new NXE:3800F model in 2027 (with 260 WpH throughput) and the next-gen NXE:4200G (with 300 WpH) in 2029, they’ve confirmed it will carry a meaningful ASP increase. The value capture happens through innovation and productivity improvements, not opportunistic gouging.

And, one last chart for those doubting pricing power over the years:

😶 🤫

The Q2 Margin Guide-Down: Noise ?

Q2 guidance calls for 51-52% gross margins, down from Q1’s 53%. The bears will seize on this as evidence that margins are peaking. I think that’s the wrong interpretation.

“The primary ones being the very high margin elements in the upgrade business that we had in Q1, we’re not counting on such a strong pattern in the upgrade business in Q2... And then... the increase in headcounts that we’re going through to prepare for upticks in the move rate.” — Roger Dassen, CFO

Management was explicit about what’s driving this: hiring and training costs associated with ramping production capacity, plus normalization from Q1’s unusually high software upgrade mix. These are the costs of scaling to meet surging demand, exactly what you want to see a capital equipment company doing when facing a multi-year AI upcycle.

The reality is that ASML is investing to build at least 80 Low-NA EUV systems and 600 DUV systems by 2027. Scaling production by this magnitude requires workforce expansion and operational investment. If you’re going to criticize these margin pressures, you’re essentially arguing they should forgo the capacity expansion, which would be a big mistake given the demand environment.

Looking Forward: The 2030 Framework Remains Intact

ASML reiterated their 2024 Investor Day targets: €44-60B in revenue by 2030 with 56-60% gross margins. With 2026 now guided to €36-40B (up from previous €34-39B), they’re tracking beautifully toward the midpoint or upper end of this range.

What I find most compelling is the visibility :

• Memory customers are sold out for 2026. The company has communicated minimum 2027 production targets.

• The transition from ‘E’ to ‘F’ system tools going into 2027 (with 15%+ higher throughput) provides a clear structural margin catalyst.

• High-NA adoption is progressing ahead of skeptics’ expectations.

This isn’t a company where you need to squint and hope the numbers work out. The earnings power is becoming increasingly visible with each passing quarter.

The Investment Case

ASML just demonstrated that it can absorb a 17-percentage-point decline in China exposure in a single quarter while simultaneously raising full-year guidance. They’re successfully transitioning from a Logic-dominated business model to one where Memory provides equal contribution. The installed base is generating substantial recurring revenue. Production capacity is scaling to meet unprecedented demand.

If you needed this quarter’s results to confirm ASML’s competitive position and market dynamics, your conviction was probably too low to begin with. The company remains the single most irreplaceable asset in the global semiconductor supply chain, and this quarter simply reinforced what was already evident to anyone paying attention.

Thanks for following along,

—Nikotes

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Portfolio Corner - 🗓️ Monthly Updates (Last Update: 29-Mar-2026)

Subscribe now

Revenue grew 16% in Q4, organic growth barely registered at 1% for the year, and M&A deployment was virtually nonexistent at just $23 million for all of 2025.

At first glance, Lumine had what might look like an unspectacular quarter. But here’s the thing: if you understand the Constellation playbook and by extension, the Lumine playbook, this quarter was a masterclass in capital discipline and operational execution.

Lumine spent most of 2025 doing what great capital allocators do when valuations don’t make sense: they wait. They optimize. They print cash. And then, when the right opportunity shows up, they strike. The $306.7 million Synchronoss acquisition announced in February 2026 is precisely that strike, and it was funded by a balance sheet that management rebuilt throughout 2025.

Let’s take a closer look at the ‘not-so-ugly-duck’ of the Constellation family.

Financials

The Cash Flow Story

Let’s cut through the noise and focus on what really matters: Free Cash Flow Available to Shareholders (FCFA2S).

This is the magical metric used across the Constellation Software universe to determine whether management is actually creating value or simply buying revenue.

Lumine Group delivered these numbers: FCFA2S rose 48% in Q4 to $67.1M and jumped 153% for the full year to $217M. On $765M in revenue, that implies a 28.3% free cash flow margin. Let that sink in and then ask most US-listed SaaS companies what they think about that level of cash generation.

Interestingly, this wasn’t driven primarily by revenue growth. Organic growth was essentially flat at 1% FX-adjusted, with total revenue growth in the mid-teens. The real driver was operational efficiency.

Expenses increased just 8% in Q4, while revenue grew 16%. Staff costs (the largest line item for VMS businesses) rose only 11% in Q4 and 6% for the full year.

Huge credit goes to CEO David Nyland and the management team. Lumine is extracting incredible leverage from its existing portfolio by doing exactly what Constellation taught them: optimize costs, shift to lower-cost geographies, strip out redundancies, and let the cash compound.

Why 2025 Was Actually Perfect

Some investors might look at $23 million in M&A deployment and wonder if Lumine has lost its edge. I would argue the opposite. The willingness to sit on your hands when deals don’t make sense is the mark of a disciplined allocator, not a desperate one. Management could have forced acquisitions to show “growth,” but instead they chose to deleverage, accumulate cash, and wait for the right opportunity that meet their expected IRR and hurdle rates.

That opportunity came in the form of Synchronoss Technologies, a $306.7 million acquisition that closed in early 2026. To fund this, Lumine used accumulated cash and drew down $160 million on its expanded credit facility. The company now has a $360 million revolving facility, showing that it has the scale and credibility to swallow public targets in the hundreds of millions. This is a step-function increase in Lumine’s ability to deploy capital.

The reality is that 2025 wasn’t a “slow” year, it was more like a setup year. And given the nature of Lumine Group’s acquisition targets, we’ll likely see more of these periods.

But what matters for shareholders is how that $217M in FCFA2S was used. It wasn’t paid out in dividends or deployed into forced mediocre deals or buybacks. It was stockpiled as dry powder for precisely the kind of larger, higher-IRR carve-out acquisitions that can move the needle.

A Cleaner P&L Finally Showing Up

For those of you who’ve been following Lumine since the spin-out, you’ll know that 2024’s financials were an absolute mess due to a $317 million non-cash expense related to the mandatory conversion of preferred and special securities. This legacy capital structure artifact made the bottom line virtually unreadable, showing a net loss of $258.9 million despite the business printing cash.

2025 wipes most of that away. Net income for the year came in at $118.8 millio, cleanly positive for the first time. Operating income grew 31% for the year. With the preferred share noise gone, the underlying profitability of this business is now crystal clear. And for investors trying to model forward earnings, this is a much cleaner starting point.

The Organic Growth Question

Let’s address the elephant in the room: 1% organic growth is objectively anemic.

If you’re a growth investor, this may look like a red flag. But here’s the thing with Lumine: they aren’t exactly playing the organic growth game like Topicus (to some extent). But, why?

Well, it turns out the Communications & Media vertical where they play is mature. Telecom procurement cycles are long. Customers are budget-constrained. But, in my opinion, none of this is should change the thesis if you understand what’s at play here.

The Constellation model doesn’t require high organic growth to work, it requires high cash conversion and attractive returns on deployed capital. Lumine is delivering both.

So when you ask yourself what game Lumine Group is playing, the answer is simple: the compounding game. And what makes it even more compelling is that they’re doing it from a much smaller market cap base than Constellation Software while operating at the same level of excellence.

Once again, credit to CEO David Nyland and the management team.

Now, on the recurring revenue side: their maintenance and recurring revenue figure represents roughly 70% of total revenue ($537.7 million), providing a great stable base. And when you couple that recurring base with 28% free cash flow margins, you have a business that can fund meaningful M&A without diluting shareholders or over-leveraging the balance sheet.

I would even go as far as to say that if Lumine were growing organically at 10% while generating 15% free cash flow margins, I’d be less interested…. (What are you smoking Nikotes?)

The current configuration, flat organic growth with exceptional cash generation, is actually ideal for a pure single-vertical serial acquirer like Lumine because it means the organic business is essentially self-funding while all the M&A dollars flow to incremental growth. Or, put differently: the base business generates cash to fund carve-outs, so M&A drives incremental growth without diluting returns.

What Synchronoss Acquisition Means for 2026

The Synchronoss acquisition is the real story going forward this year, even though it closed just after year-end. At around $306.7 million in enterprise value, this is the largest deals Lumine has executed since IPO. For context, total M&A deployment in 2025 was $23 million. This is a step-function increase in scale.

Investors should expect a few things in the coming quarters:

First, integration costs and elevated amortization will hit the P&L in H1 2026. This is par for the course with any acquisition of this size, but it will temporarily obscure the underlying profitability improvements.

Second, management will apply the classic Lumine margin expansion playbook to Synchronoss. This means cost rationalization, elimination of public company overhead, and operational efficiency gains. If history is any guide, we should see meaningful margin expansion from Synchronoss by late 2026 or early 2027.

Third, the debt load will increase. Lumine drew down $160 million on its credit facility to help fund the deal. This means interest expenses will rise, and management will need to prioritize debt paydown over the next 12-18 months before they can reload for the next major acquisition.

The Risks: We Need To Talk

Let’s be realistic about the headwinds here particularly as Lumine keeps scaling up and becomes bigger.

Organic growth is barely above zero, and if it turns chronically negative, Lumine will have to rely entirely on M&A to offset the decay of its legacy assets. The telecom sector is consolidating, and customer concentration is a real risk. When telcos merge, Lumine could lose customers or face pricing pressure.

The increased debt load from big acquisitions like Synchronoss also introduces financial risk. While Lumine ended 2025 with an under-leveraged balance sheet, the $160 million draw-down changes that equation once again. If the business hits a rough patch before one of these big acquisitions starts contributing meaningful cash flow, the debt service could become a constraint. But given how good is management at operating this ship, I would not be too concerned about it.

Finally, there’s the execution risk inherent in any large acquisition. Synchronoss was a public company with its own baggage. Integrating a $300 million asset is fundamentally different from bolting on $5-20 million tuck-ins like Topicus or Constellation are used to do. Management will need to prove they can execute at this bigger scale. So far, they are doing so beautifully.

How to Think About Lumine As an Investor

Lumine is a spin-out from Constellation Software focused exclusively on Communications & Media vertical market software. It operates exactly like its parent and sister company Topicus, but with a narrower vertical focus. The business model is simple: acquire sticky, mission-critical software businesses in telecom at reasonable multiples, optimize the hell out of them, and let the cash compound.

The investment thesis is not about organic growth, but about disciplined capital allocation, exceptional cash generation, operational efficiency and the compounding effects of reinvesting that cash at attractive returns. The fact that management sat on their hands for most of 2025 and then deployed $307 million in early 2026 is exactly the behavior you want to see.

For shareholders, the question is whether Lumine can maintain high returns on deployed capital as it scales. The Synchronoss acquisition is another big test of that hypothesis. If they can apply the margin expansion playbook successfully once again and maintain their M&A discipline at this scale, the compounding story remains intact.

Thanks for following along,

—Nikotes 🫡

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Portfolio Corner - 🗓️ Monthly Updates (Last Update: 29-Mar-2026)

Subscribe now

Topicus delivered a truly good Q4 that based on the takes I’ve read on social media (X, Substack, etc.) some users have missed it entirely. Why? Because anyone screening on GAAP Net Income will see a 53% decline and move on. This drop is nothing more than accounting noise, a non-cash artifact of transitioning the Asseco Poland investment to the equity method.

Meanwhile, Free Cash Flow Available to Shareholders (FCFA2S) jumped 40% in Q4 and 23% for the full year. This is one of the key numbers that shows the health of this business, and it’s accelerating precisely when the market narrative suggests VMS should be deteriorating.

Financial Results

Understanding the Accounting Noise

Investors analyzing screeners will see the 53% drop in Net Income and assume something is broken. The MD&A clarifies this is entirely artificial. Initially, Topicus recorded the Asseco stake at Fair Value (FVOCI), riding the stock price up. Once regulatory approval was achieved in September 2025 to increase the stake, accounting rules forced a switch to the Equity Method. This required reverting the asset back to its initial cost, triggering a €221.7 million non-cash reversal through the income statement. (Read more on: 🔗 Fair Value vs Equity Method).

This is pure accounting mechanics with zero cash impact. I’ve said it before with CSU and will say it again here, investors must ignore the P/E ratio and GAAP Net Income entirely. The true valuation anchor is FCFA2S, which, unlike Constellation Software’s cash flow metrics, is a cleaner number because it doesn’t include the IRGA liability deductions that warp CSU’s cash flows as I also explained on my 🔗 CSU Q4 2025 earnings digest.

Note that, EBITDA and the EV / EBITDA multiple could also be useful metrics, in CSU family’s case, although I still prefer the FCFA2S number as a long term trend indicator.

Here’s a great visual shared by the great C.J. on X.

What a beautiful trend, don’t you think?

The Organic Growth Floor

Topicus achieved 6% organic growth in recurring maintenance revenues for both Q4 and the full year. That might not sound particularly exciting at first glance, but it’s actually one of the most telling numbers in the entire release.

This steady organic growth has provided a stable floor for the business while it continues executing its M&A roll-up. As I’ll discuss in more detail later in the AI section for Topicus, it’s also an early signal that (at least for now) AI disruption and the “vibe coding” narrative aren’t driving customers away from these mission-critical systems.

Recurring revenue hit €298 million in Q4, representing 68% of total revenue. The 6% organic growth in this segment outpaced professional services (0%) and licenses (-17%), supporting the transition toward higher-quality, predictable revenue streams following the mothership steps. This ensures management can keep deploying capital aggressively without worrying about cash flows lumpiness.

Capital Allocation Strategy Is Evolving: The Asseco Pivot

Some investors worry that the Asseco and Sabre deals signal mission creep, a departure from the classic CSU playbook. I don’t see it that way. Just like the mothership, Topicus is simply adapting the model to current market realities and capital deployment opportunities, exactly as Mark Leonard long suggested would be necessary.

The company can’t acquire a €2 billion target outright, but it can deploy hundreds of millions to gain influence in a publicly traded peer. This, in my opinion, expands the Total Addressable Market for capital deployment while maintaining the core discipline around return hurdles. The traditional VMS engine continues running in parallel, absorbing €390M in acquisitions during 2025 alone.

To me, this is what successful capital allocation looks like at scale: multiple parallel strategies, each optimized for different market segments and capital requirements.

An Excellent Balance Sheet Favoring More Capital Deployment

Despite 2025 record numbers in capital deployment, Topicus ended the year with just €366M in net debt, lower than their full year Operating Cash Flow. By any measure, leverage remains conservative and the maturity of their debt is clean. Some important debt maturities don’t arrive until 2028, with the largest bullet (€95.4M) not due until 2032. (You can verify this in the MD&A on the Topicus IR website, page 16.)

All in all, this positions Topicus well for capital deployment going into 2026, with dry powder for what I believe is an exceptionally favorable M&A environment.

With SaaS-mageddon creating havoc across public markets, private equity holding periods stretching, and fewer competing buyers, Topicus is stepping into a landscape of compressed SaaS multiples, more willing sellers, and limited capital constraints on its side.

The AI Disruption Narrative Is Empirically Wrong

The market has discounted Horizontal SaaS and VMS stocks indiscriminately on the assumption that AI will erode terminal values. While Topicus formally acknowledged AI risks in their MD&A (twice now), the actual operating data tells a very different story. Let me walk through three leading indicators that should put this bear case to rest, at least for now.

First, organic growth in recurring maintenance revenues remains unbroken. If AI agents were actively displacing Topicus software, we would see churn spike and maintenance revenues drop. Instead, Maintenance & Other Recurring Organic Growth held steady at 6% for both Q4 and full-year 2025. This is consistent with the 5-7% range maintained over the last 3 years. No acceleration, but no deceleration either.

Second, the staff expense figures could be starting to show productivity gains:

Overall costs grew 18%, but R&D expenses surged 32% in Q4 and 22% for the full year. This shows the company is becoming more efficient overall while funneling resources into product development. If anything, this looks more like AI-augmented productivity than AI destroying the business model. Of course, it’s still too early to draw conclusions, we’ll need more quarters to see how this plays out. But it’s reassuring to read between the lines and see that management is being proactive rather than reactive.

Third, I’d like to highlight an excellent analysis done by another Substack writer, Best Anchors, in their 🔗 Topicus Q4 2025 earnings digest. The focus is on contingent considerations and impairments, two indicators that show a lot about the health of the business and the quality of its M&A execution.

Impairments declined slightly (€0.5M vs. €0.6M), while contingent considerations (earn-outs) rose to a €2.3M expense. In simple terms, this means acquired companies are hitting or exceeding their revenue and profit targets, triggering earn-out payments.

If AI were destroying terminal values or derailing near-term performance, you would expect the opposite: impairments rising and earn-outs disappearing. Instead, we’re seeing the exact reverse.

Valuation Context and Forward Outlook

Topicus paid €0 in common dividends in 2025, compared to the occasional special dividends in prior years. This also tells you a lot about management’s view of their internal rate of return environment. They’re retaining every dollar because the opportunity set is exceptional.

The stock been cut off in half from all-time highs, initially driven by Mark Leonard departure narrative (go figure …), subsequently driven by AI panic. Today, investors can buy an accelerating capital deployment machine which has been tripling its year-over-year deployment, at a normalized FCF multiple of around 20x NTM on my numbers, and operating in a highly favorable M&A environment.

What To Watch Going Forward

The key question for Topicus over the next few years is whether they can maintain deployment velocity without sacrificing return discipline. The Asseco deal just like CSU’s Sabre deal represents a new strategic direction, and we won’t know the returns on this capital for several years.

I’ll be watching three things closely:

1. Maintenance organic growth trends - Any consistent deceleration below 4-5% would be concerning

2. Staff expense ratios - Continued improvement or stability here would validate the AI-augmentation thesis vs AI disruption bear case.

3. Capital deployment cadence - Can they sustain €500+ million annual deployment?

Early signs in Q1 2026 are encouraging. Subsequent to year-end, Topicus already completed or committed to acquiring additional businesses for €20.3 million across automotive, government, chemicals, healthcare, and legal verticals. The M&A engine shows no signs of slowing.

I’ll also be sharing my Lumine Group (LMN.V) thoughts later today (time permitting), otherwise early next week.

Thanks for following along,

— Nikotes 🫡

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Portfolio Corner - 🗓️ Monthly Updates (Last Update: 29-Mar-2026)

Subscribe now

After Sir Leonard stepped down last year, the bear case was ready:

• AI disruption

• New CEO doubts: “No one can replace Leonard”

CSU management bringing back conference calls this quarter was a smart move.

The key investor question isn’t just about AI, at least not yet. It’s about capital allocation at scale: How do you deploy billions annually while maintaining 20%+ IRRs?

CSU’s answer: PEMS — Permanent Engaged Minority Shareholder strategy.

A structure to scale capital deployment without diluting hurdle rates. And the 🧙‍♂️ is still there, guiding M. Miller & co. from the back seat.

Music to my ears. It shows CSU understands the laws of large numbers better than anyone and has a plan to evolve its playbook for the next phase.

Financial Results

The Numbers (And Why They Don’t Tell the Full Story)

Revenue grew 18% to hit just over $3 billion for the quarter, reaccelerating from +16%, with 9% maintenance and recurring organic growth in Q4 (6% constant currency). Not too shabby.

Another thing that stood out for some in the Fintwit / X community is that GAAP net income cratered 61% to $110 million and FCFA2S decreased by 16%.

The reality lies, once again, in the details: a $204 million non-cash revaluation charge related to CSU’s investment structures (specifically the IRGA/TSS membership liability tied to Asseco and Sygnity holdings).

This is pure accounting noise. Exclude the IRGA (as investors should) and the picture is clear:

• FCFA2S +15.8% and +28.3% for the full year

• Cash from Operations (CFO) +16% to $788M and full-year CFO at +24%

Focus on the true cash generation, not the GAAP earnings distorted by non-cash revaluations of CSU’s public holdings as they appreciate in value.

PEMS: The Strategic Response to Scale

The most significant development this quarter wasn’t in the financials at all. It was management’s formal articulation of the PEMS strategy, which has been quietly unfolding over the past few years with investments in companies like Asseco with Topicus leading the way, and now, CSI with Sabre.

On the new PEMS Strategy (Permanent Engaged Minority Shareholder) with Mark Leonard helping to craft it from the back seat:

We’ve also developed a new approach to deploying larger amounts of capital, what we’re calling a Permanent Engaged Minority Shareholder strategy or PEMS. Our investment in Sabre is the first meaningful expression of it. And I have to thank Mark Leonard who is helping us with this strategy and helped us explicitly on this particular investment. But the logic is straightforward. Permanent, meaning we’re long-term share -- long-term holders, not traders. We’ll work to ensure these companies endure as institutions. We’re engaged, which means we care about governance, management incentives and capital allocation, and we’ll actively work to have an influence where we think it creates value. Minority, we’re not acquiring these businesses outright. We want to partner with other shareholders, and we hope many of them become engaged long-term holders alongside us. — Mark Miller, President & COO

Here’s the problem CSU has been facing: at nearly CAD 64 billion in market cap, deploying $2-3 billion annually into small VMS acquisitions at 25%+ IRRs is becoming mathematically harder. The traditional playbook, buying $5-20 million revenue software companies from retiring founders at 3-5x EBITDA still works, but it can’t absorb all the cash CSU generates.

PEMS is management’s answer. Rather than trying to force-feed capital into overpriced small deals, they’re taking minority stakes in larger public companies, demanding board seats, and attempting to influence capital allocation and operational decisions. They’re essentially trying to export the CSU playbook to undervalued public companies without paying takeover premiums.

This is very clever, but it’s also an implicit admission that the classic model faces constraints at current scale. The thing is that being an engaged minority shareholder is fundamentally different from owning 100% of a business. You can advise, you can influence, but you can’t dictate. If management teams at portfolio companies like Sabre resist CSU’s governance suggestions, capital could get trapped in suboptimal investments.

Does that mean I’m less excited? Quite the opposite.

If executed well, this could become one of the most powerful levers available to them. With SaaS-mageddon leaving plenty of SaaS “cadavers” across public markets, it might create exactly the kind of hunting ground where an exceptional capital allocator like CSU can thrive.

The AI Question (And Why Management Gets It Right)

Let’s start this section with this quote on AI as a Moat vs. Table Stakes during the CC:

“I want to be direct about something, building products and features faster will not be what differentiates us long term. That capability will become widely available. It’s going to be table stakes. What will matter is what our businesses have spent many years developing, deep vertical knowledge, a genuine understanding of customer workflows... and the trusted relationships they’ve built.” — Mark Miller, President & COO

Management addressed the AI disruption narrative with a pragmatism that’s rare in today’s market. Mark Miller was explicit: AI-assisted coding will become table stakes, not a competitive advantage. Every software company will have developers using AI tools to write code faster.

What won’t be commoditized? Deep vertical knowledge, proprietary customer data, and entrenched workflows.

As I’ve been pointing at on different pieces released recently at Expanse Stocks, this is exactly where CSU’s moat actually lives. Miller’s vision of “knowledge networks”, using AI to mine CSU’s massive trove of customer-specific data to build better solutions, is the right way to think about this.

On Refusing to Centralize AI:

“As far as a large centralizing development or centralizing AI tools, we’re not really a big fan of that because... it means you’re taking away the ability for that business leader who could be in some country, somewhere... to move quickly to give them what they need... we’re very hesitant to centrally control anything here.” — Mark Miller, President & COO

The decentralized approach to AI adoption is also classic CSU. There’s no top-down mandate. Instead, hundreds of business units are experimenting, competing, and sharing best practices. The best ideas win and propagate organically.

While I find this reassuring, there’s still a tail risk worth acknowledging. If agentic AI eventually lowers the barrier to entry for bespoke vertical software over a 5-10 year horizon, it could pressure pricing power or increase churn. Management is confident this won’t happen, and their reasoning is sound.

Capital Allocation & The Buyback That Wasn’t

On Share Buybacks (NCIB):

“We have created a subcommittee within the Board to look at this... But at this point in time, we believe there are ample opportunities to deploy capital as opposed to buying back our shares.” — Jamal Baksh, CFO

CFO Jamal Baksh confirmed that the board evaluated implementing a share buyback program (!!) and ultimately rejected it.

The conclusion? There are still too many high-return opportunities available to justify buying CSU stock at current valuations.

In my view, that’s very good news. It suggests management continues to see attractive capital deployment opportunities and that buybacks simply don’t meet CSU’s self-imposed hurdle rates, targeting 20%+ IRRs.

In Q4 alone, the company deployed nearly $900M across traditional M&A and minority investments. Since year-end, in just a couple of months, they’ve already committed another $802M.

The capital deployment engine is clearly running hot, validating management’s decision to skip buybacks.

What This Means for Returns Going Forward

I’ve written before about the inevitable gravity of scale, and this quarter’s developments reinforce that thesis. CSU is no longer the CAD 10 billion market cap compounder it was a decade ago. At CAD 64 billion, the math changes.

The PEMS strategy is an intelligent adaptation, but it inherently carries different return characteristics than the classic VMS roll-up model. Minority stakes in public companies, even with board influence, are unlikely to generate the same 25-30% IRRs that buying private software companies at 3-5x EBITDA delivered historically.

Management confirmed that PEMS investments are evaluated using the same IRR hurdles as traditional acquisitions, which is reassuring. But the reality is that deploying billions annually while maintaining 20%+ returns is exceptionally challenging, regardless of strategy.

My base case remains that CSU will gradually see returns compress from the mid/high 20s toward the low twenties to high-teens over the next 5-10 years. This is just a graceful moderation driven by scale. Even at 17-22% IRRs, CSU would remain one of the best capital allocators in public markets.

What to Watch: The Numbers That Actually Matter

Here’s what I will be monitoring throughout the year:

• Organic growth: Holding steady. No degradation despite AI fears. Management indicated this is a “good number for future organic growth”.

• M&A volume: $571 million deployed in Q4 for traditional acquisitions, with the pipeline remaining full. Private market valuations haven’t contracted meaningfully, but CSU’s funnel of prospects is intact.

• Free cash flow (exc. IRGA): Growing faster than revenue, which is what you want to see in a compounder. The $802 million in deals already committed for early 2026 shows the deployment pace isn’t slowing.

Thanks for following along,

—Nikotes

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Portfolio Corner - 🗓️ Monthly Updates (Last Update: 29-Mar-2026)

Subscribe now

Mercado Libre reported Q4 2025 earnings that perfectly encapsulate, once again, why this company remains one of the most compelling growth stories in public markets. The headline numbers: 45% revenue growth to $8.8 billion, but operating margins compressed 340 basis points to 10.1%.

The market's initial reaction to the margin compression is, once again, predictable … *sigh* ... but misguided. Let me explain why.

Financial Results

Big Numbers At Bigger Scale

A Margin Compression “Problem”

Operating income grew 22% YoY, but margins fell from 13.5% to 10.1%. Before anyone panics, CFO Martin de Los Santos was explicit about what’s happening here:

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

“We are not managing the business for short-term margin... 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 believe these investments are creating a foundation for future growth.”

— Martin de Los Santos, CFO

“We are not managing the business for short-term margin.” This quote sound tired, at least to the market, but management estimates that strategic investments in shipping subsidies, credit cards, first-party inventory, and cross-border trade are dragging margins by 5-6 percentage points.

In my view, this isn’t margin compression in the way investors should worry about. MELI isn’t seeing margins collapse because its core economics are deteriorating. It’s choosing to deploy capital aggressively across logistics, fintech, and inventory to deepen its moat while the opportunity set is still wide open.

The company is making a bet that owning the user now will generate significantly higher returns over the next decade than optimizing for today’s margin profile.

The unit economics tell you everything you need to know about whether this strategy is working. Despite flooding Brazil with lower-ticket items (which should theoretically pressure unit economics), shipping costs per unit fell 11% YoY. That’s the entire story right there: MELI is subsidizing shipping to win volume, but their logistics network is efficient enough that the actual cost of moving each additional item is declining. This is what operating leverage looks like before it shows up in the P&L.

The R$19 Effect: Brazil Commerce Acceleration

The decision to lower Brazil’s free shipping threshold from R$79 to R$19 has been a key driver of growth. Items sold in Brazil accelerated from 26% growth in Q2 to 45% in Q4.

"New buyers that have come to Mercado Libre 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." — Ariel Szarfsztejn, CEO

New buyer cohorts are showing higher retention and broader category engagement, which shows MELI is successfully training a new generation of Latin American consumers to shop online first.

With 45% volume growth in Brazil, MELI is almost certainly taking significant share from both Asian cross-border players (Shein, Shopee) and local incumbents (Magazine Luiza, Casas Bahia). The heavy logistics Capex of the past few years is now validating itself through competitive positioning that rivals simply cannot match.

Fintech: From Payments Processor to Primary Bank

Mercado Pago continues to be the most underappreciated part of the MELI story.

Assets Under Management grew 78% YoY, driven by interest-bearing accounts. This matters because MELI is starting to fund its massive credit expansion (credit portfolio up 90% to $12.5 billion) with customer deposits rather than wholesale funding. This is how you build a bank, not just a payments rail.

The credit card portfolio is the primary growth driver here, and the unit economics are proving out faster than expected.

Management noted that credit card cohorts in Brazil reach NIMAL breakeven (profitability after losses) in 12-18 months, and 75% of the Brazil card portfolio has already hit this milestone. For context, NIMAL expanded to 23.3% from 21% despite the rapid portfolio growth, which tells you the company is pricing risk correctly.

"More important than NPLs are Net Interest Margin After Losses and those improved... Therefore, what we did was we increased the number of people and the riskier number of people we give credit to, but we price that risk accordingly. And therefore, we ended up having a significantly larger spread than we did on the prior quarter. So I think this was a calculated risk and it worked out well."

"Cohorts that are older than 2 years are already profitable at a NIMAL level... half of the portfolio is already NIMAL [positive] in Brazil."

— Osvaldo Giménez, Fintech President

There will be noise around early-stage delinquencies (15-90 days) ticking up.

But, as Fintech President Giménez explained, they’re deliberately expanding down the credit curve and pricing that risk accordingly. The wider spread (NIMAL) is the reward for taking that risk. This is textbook subprime lending done correctly: price for the risk, accept higher delinquencies, and make sure your net return is attractive.

The Flywheel in Full Display

This quarter perfectly demonstrated (once again) why MELI’s ecosystem is so powerful.

The commerce investments (shipping subsidies) drove traffic. That traffic fueled the advertising business, which grew 67% YoY. The data from that traffic enabled massive credit issuance, which grew 90%. And the credit cards drive “principality” in Fintech, with users making Mercado Pago their primary financial account which closes the loop by driving more commerce volume. The flywheel in full display.

Each piece reinforces the others. This is why MELI can sustain 40%+ growth at $8.8 billion in quarterly revenue, a scale where most companies are happy with mid-teens growth.

AI: Tailwind, Not Headwind

New CEO Ariel Szarfsztejn addressed the “Agentic Commerce” question head-on, the fear that AI shopping agents might commoditize marketplaces by making product discovery platform-agnostic. His response was simple: MELI will build the best agent because it has the best first-party data on Latin American consumer behavior.

"We are betting and putting our efforts on what we can control, which is building the best assistant possible... relying on our first-party data to create the best search, recommendation, and discovery engine... I think that MercadoLibre is well-positioned to capture this technology transformation, which... will accelerate the migration from offline retail into online retail." — Ariel Szarfsztejn, CEO

Far from being disrupted, MELI is using AI as an accelerant. The 67% growth in advertising revenue is partially AI-driven (better targeting, higher ad loads), and credit underwriting is becoming more precise as AI models ingest more data. MELI controls the data layer in Latin America’s two most important digital businesses (commerce and fintech), which makes it exceptionally well-positioned in an AI-driven future.

What to Watch: A Path Back to Margin Expansion

Management made clear that 2026 will see continued aggressive investment. The market needs to accept that ~10% operating margins are the new baseline for the next 12-18 months. The critical question is whether the 5-6 percentage point investment drag eventually converts into operating leverage as volume scales and logistics efficiency improves.

The early signs are encouraging. Unit shipping costs fell 11% despite lower average order values, and credit card cohorts are breaking even faster than expected. If these trends continue, we should start seeing margin expansion in late 2026 or 2027 as the investment cycle normalizes and operating leverage kicks in.

Here’s what I’ll be monitoring:

• Direct Contribution Margin in Brazil: Watch whether logistics efficiencies begin offsetting shipping subsidies in the next few quarters

• NPLs >90 days: Early delinquencies are expected given the credit expansion, but late-stage delinquencies are where real credit losses show up

• NIMAL trajectory: If this continues expanding while the portfolio grows, it confirms the risk-adjusted pricing is working

The Investment Case

MELI is executing a playbook that few companies have the courage to run: sacrifice near-term margins to dominate emerging markets before they fully mature. The difference between MELI and companies that blow up doing this is that MELI’s unit economics are actually improving while margins compress. That’s the entire difference between intelligent capital allocation and reckless growth spending.

The reality is that MELI is the only company in Latin America successfully building both a dominant e-commerce platform and a scaled fintech business. That combination is rare globally and essentially impossible to replicate at this point in LatAm. The moat isn’t just the logistics network or the credit models, it’s the data flywheel that connects them.

Investors need to decide whether they’re willing to accept 10% operating margins today in exchange for owning the digital infrastructure of Latin America’s consumer economy. The items sold acceleration (+45%) and the credit portfolio growth (+90%) tell you MELI is winning the user. The falling unit costs tell you the economics work. The margin compression is the price of admission to capture this opportunity before the market fully consolidates.

If you’re uncomfortable with MELI’s margin profile today, you probably don’t understand what management is building. This isn’t margin compression from competitive pressure (at least not in any material way) or structural deterioration, this is margin investment to cement a two-decade competitive advantage. There’s a meaningful difference between the two, and mistaking one for the other is how you miss great companies at inflection points.

Thanks for following along,

—Nikotes

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Portfolio Corner - 🗓️ Monthly Updates (Last Update: 29-Mar-2026)

Subscribe now

Long time no see, Cloudflare… The very first company I ever covered in my Deep Dive Briefs (🔗 Cloudflare) at Expanse Stocks is back high on my watchlist.

Despite trading at sky-high multiples, Q4 2025 results (presented Feb 10) were spectacular. The initial +15% pre-market spike didn’t last long, the stock quickly drifted back to around $180 as the SaaS-mageddon keeps ravaging across the SaaS world.

The truth is Cloudflare’s Q4 earnings were nothing short of excellent. Revenue growth re-accelerated to 34% YoY while new ACV bookings surged nearly 50%. The numbers speak for themselves, but what’s more interesting is why this is happening now and whether it’s sustainable.

Let’s unpack what’s actually driving this performance.

Q4 2025 Financial Results

The GTM Retooling is Finally Paying Off

For the past two years, Cloudflare has been rebuilding its go-to-market strategy. The concerns around sales execution that plagued 2023 and early 2024? Those are effectively behind us.

"I’m proud to report our go-to-market engine is firing on all cylinders. We blew away our previous record for new ACV in the quarter... In Q4, new ACV booked grew nearly 50% year over year, making it not only a record quarter in absolute ACV dollars but also the fastest growth rate we've delivered since 2021." — Matthew Prince, CEO

Some big words from Matthew, as usual, but the evidence is also reflected in the deal flow. Cloudflare closed its largest annual contract ever at $42.5M/year and saw a record number of new $1M+ customers. Sales productivity surpassed the all-time highs set back in 2021. The thing is, this isn’t just about hiring more reps or improving conversion rates, it’s about fundamental changes to how they contract with customers.

Their “Pool of Funds” approach, where customers get flexibility across the entire platform rather than being locked into specific products, accounted for 20% of ACV in Q4. This matters because it removes friction from the sales process and allows customers to make larger, more strategic commitments upfront. It’s a smart adaptation to how enterprises actually want to buy infrastructure.

The result and consequences of this GTM shift? Well… the numbers below speak for themselves:

AI Agents: The Infrastructure Multiplier

Matthew positioned Cloudflare as the “control plane” for what he calls the “Agentic Internet.”

While this sounds like typical CEO hyperbole, the underlying mechanics are proving themselves: AI agents are generating exponentially more traffic than human users ( - Prince noted that weekly requests from AI agents doubled in January alone.

“If you look at the last 30-plus years of the Internet and software ecosystem, they were built for human consumption, people in seats and clicks. Now the agentic Internet is emerging, and we can already see its trends. If humans looked at 5 sites when they were making a decision, agents might look at 5,000.” — Matthew Prince, CEO

The key insight here is simple: if a human looks at five websites, an AI agent might look at 5,000. This creates massive infrastructure demand, and Cloudflare sits right in the middle of it. They’re monetizing this shift through both security (blocking unwanted scrapers) and compute (the Workers platform for AI applications).

Unlike many SaaS companies where AI is still just a narrative, and for which the market has been punishing harshly, Cloudflare is seeing tangible revenue from it. They cited multiple examples of customers choosing Workers over hyperscalers for deploying AI applications, including the $45M deal that displaced a hyperscaler for global application management.

The Hyperscaler Displacement Story

What’s particularly interesting about this quarter is how aggressively management positioned against AWS, Azure, and GCP. Prince didn’t mince words, arguing that hyperscalers have “no incentive” to make AI workloads efficient because they rent servers by the hour. Cloudflare’s serverless model, by contrast, aligns economic incentives with efficiency.

“The hyperscalers are in the business of buying machines and then renting them back... Cloudflare is in the business of getting work done.” — Matthew Prince, CEO

The company is winning not just on price but on performance and deployment speed. They highlighted several instances where customers chose Workers over hyperscaler offerings because they could ship code faster and run it cheaper. This is the classic innovator’s dilemma playing out: the hyperscalers are optimized for their existing business (renting compute by the hour), which makes them vulnerable to a model optimized for the new world (serverless, edge-native applications).

“Vendors are once again being sorted between nice-to-haves and must-haves. The quantitative metrics and qualitative wins this quarter again demonstrate that Cloudflare is squarely in the must-have bucket.”

“We didn’t just replace three incumbents; we replaced a legacy mindset, proving once again when you give developers the best tools, the entire enterprise follows.”

— Matthew Prince, CEO

Of course, whether this displacement thesis plays out at scale remains to be seen. The hyperscalers aren’t going to sit idle, and they have significant advantages in bundling, sales relationships, and wallet share. But Cloudflare seems to be winning more of these head-to-head battles than they were two or three years ago.

The Numbers We (Investors) Should Care About

Revenue growth of 34% at this scale (~$2.5B run rate) is impressive, but let’s talk about what’s driving it and what it means for the model going forward.

First, they’re achieving this while maintaining relatively modest CapEx at 12-15% of revenue compared to the much higher percentages hyperscalers are spending on GPU clusters.

Second, gross margins came in at 74.9%, slightly below the long-term 75-77% target. Management attributed this to mix shift - more paid traffic relative to free traffic and the contribution of the Workers platform. While they emphasized that unit economics remain consistent, this is something to monitor. If gross margins continue compressing, it would suggest either pricing pressure or an adverse shift in the business mix.

Third, new ACV bookings growing at 50% while revenue grows at 34% tells you that the sales pipeline is accelerating faster than the P&L reflects. This is a leading indicator of sustained momentum, assuming these deals convert into revenue as expected.

The company is operating at a Rule of 40 score of around 50 (34% growth + 16% FCF margin). This is best-in-class territory and justifies a premium multiple.

The real question isn’t whether it deserves a premium, but how much of one relative to its durability, growth visibility, and capital efficiency.

Take a closer look at the numbers below and draw your own conclusions.

What Could Go Wrong

The gross margin compression, even if explained by mix, is worth watching. Cloudflare has historically commanded premium pricing because of its performance advantages, but as they move up-market and compete more directly with hyperscalers, pricing power could erode.

The “Pool of Funds” contracting approach, while smart for closing larger deals, introduces potential revenue recognition volatility. As these deals grow to 20%+ of ACV, quarterly revenue could become lumpier depending on how customers draw down their commitments. This isn’t necessarily a problem for the long-term economics, but it could create forecasting noise that the market punishes.

Management also acknowledged a “volatile start to 2026” for the tech sector. While they feel relatively insulated, broad IT spending cuts would eventually impact even the highest-performing infrastructure companies. You get me - no one is immune to a genuine spending freeze.

Finally, there’s the question of hyperscaler retaliation. As Cloudflare increasingly displaces AWS/Azure/GCP for compute and storage (their R2 product), competitive intensity will likely increase. The hyperscalers could respond with aggressive bundling or pricing, which would make Cloudflare’s displacement thesis harder to execute.

At current multiples, perfection is priced in. If any of the above plays out, the downside could be fast and furious.

The Moat Question

20% of the web now sits behind Cloudflare. This scale creates what they call a “data gravity” advantage for training AI models and controlling AI crawlers.

It’s a flywheel: more sites use Cloudflare → more visibility into internet traffic → better data for new products → more sites use Cloudflare.

This network effect is real and difficult to replicate. You can’t build a product like their AI crawl control marketplace (connecting content creators with AI model builders) without having internet-wide visibility. AWS can’t replicate this just by spending money.

The question is whether this moat is sufficient to sustain premium returns as the company scales. At a $2.5B revenue run rate, they’re no longer a scrappy upstart - they’re an established player going head-to-head with some of the best-capitalized companies in tech. The moat needs to be strong enough not just to defend existing customers but to continue winning new ones at attractive economics.

Looking Forward: FY 2026 Outlook & The Act 4

Cloudflare’s FY 2026 guidance calls for 28-29% revenue growth with operating margins around 14%. Management is well know for its conservatism, these numbers probably implies the lowest range of its continued high growth at scale.

The thing is, we need to think about what this business looks like in 5-10 years, not just 2026. Can they sustain 25%-30%+ growth while maintaining or expanding margins? The answer depends on whether the AI agent thesis plays out as dramatically as management suggests and whether they can continue displacing hyperscalers for compute workloads.

My base case is that Cloudflare sustains low-twenties growth for the next several years while gradually expanding operating margins into the high-teens / low-twenties. This would put them in a relatively elite category of capital-efficient growth companies.

The upside case is that AI agents drive even more dramatic infrastructure demand and Cloudflare becomes the default platform for the next generation of applications. The downside is that hyperscaler competition intensifies and gross margins compress faster than operating leverage can offset it.

There’s another layer the market is underappreciating.

Matthew and the management team at Cloudflare have long framed their roadmap around the “Three Acts” strategy. I covered this extensive in my 🔗 Deep Dive Brief: Cloudflare.

But with AI agents moving from concept to deployment, they’ve been subtly hinting at a potential fourth leg of growth for a few quarters now:

Act 4: Monetizing Content for AI.

“In order for AI to be even more incredible, the Internet needs to be even more robust and full of great original content. And in order for that to happen, there has to be a way for those content creators to get paid by AI companies.”

“Because Cloudflare is that trusted neutral third party... we can be that honest broker between them [AI companies and content creators].”

— Matthew Prince, CEO

If the internet is increasingly accessed by agents rather than humans, the gatekeeper to traffic, data access, and usage economics becomes exponentially more valuable.

In other words, what started as a network + security + developer platform could evolve into an AI toll road and that optionality is barely reflected in current narratives.

Investment Implications

The combination of re-accelerating growth, improving profitability, and direct AI exposure without massive CapEx risk is rare. The GTM issues from 2023-2024 appear genuinely resolved rather than just papered over with discounting.

That said, I would caution against extrapolating this quarter’s 50% ACV growth indefinitely. Some of this could represent catch-up from the sales reorganization and pent-up demand. A more realistic steady-state assumption is probably high-20s to low-30s revenue growth with gradual margin expansion.

The key metrics to monitor going forward:

(1) whether gross margins stabilize back above 75%,

(2) continued uptake of the Workers platform as a leading indicator of AI application market share,

(3) the mix of revenue coming from “Pool of Funds” deals and what that means for quarterly volatility.

(4) longer term, the realization of the Act 4, monetizing Content for AI

The company deserves a premium valuation, but that premium should reflect realistic expectations for the next growth phase.

Just like Axon, a phenomenal company with a deep, expanding moat but trading at nosebleed multiples, it’s ultimately up to you as an investor to decide what Cloudflare is truly worth.

Thanks for following along,

—Nikotes

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Portfolio Corner - 🗓️ Monthly Updates (Last Update: 30-Jan-2026)

Subscribe now

Intuitive Surgical reported Q4 2025 results that should reinforce what long-term shareholders already know: this is one of the most durable growth machines in the market.

The company delivered another quarter of solid procedure growth, continued its da Vinci 5 upgrade cycle, and expanded into new clinical frontiers. However, the guidance and management commentary also signalled several headwinds and conservatism that investors need to monitor going into 2026.

Let’s unpack what stood out in this report and conference call.

Financial Results

The Core Engine

The thing that matters most for Intuitive’s long-term thesis is procedure volume growth, and on this front the company continues to deliver. U.S. da Vinci procedures grew 15% in Q4, with strength in after-hours acute care procedures like cholecystectomy and appendectomy, which surged 35%. Outside the U.S., procedures grew an even more +22% YoY.

Global system utilization increased 4%, driven by the efficiency advantages of the new da Vinci 5 platform. This is a key data point that doesn’t get enough attention. But, why?

Higher utilization means the installed base is becoming more productive, which translates directly into recurring revenue growth (instruments, accessories, services). With 81% of Q4 revenue coming from these recurring sources, Intuitive has built a business model that most health-tech companies would kill for.

2026 procedure growth guidance of 13–15% on a base of more than 3.1 million annual procedures highlights the durability of the business. That said, it may have come in a bit light for the Street, especially given tough year-over-year comps and management’s typically conservative posture, familiar to anyone who has followed the name long enough.

Overall, I don’t see a company struggling to sustain growth. Rather, a management team managing expectations while expanding its portfolio of clinical applications.

Da Vinci 5: The Upgrade Cycle Continues

The da Vinci platform continues to be a success. In the US, the company processed 146 trade-in transactions in Q4 versus 62 a year ago. Once again, existing customers view the new system as a meaningful upgrade worth paying for. Surgeon feedback has been positive, and the higher utilization rates confirm that the new da Vinci represents a technological leap forward.

Europe showed a good pace of system placements (118 systems versus 89 last year) and is serving as the primary market for early outside-U.S. da Vinci 5 adoption. The upgrade cycle appears healthy and should continue to drive both system revenue and higher recurring revenue per installed system.

The ASC Strategy: The Positive Surprise for Investors

One of the most interesting developments this quarter was the detail management provided on its multi-year effort to expand into Ambulatory Surgery Centers. The strategy leverages the lower-cost, refurbished da Vinci XIR system and targets higher-volume ASCs, particularly those affiliated with existing hospital customers where surgeons are already trained.

This shows smart capital allocation by management and market segmentation. Intuitive is using a lower-price-point product to address a cost-sensitive setting without cannibalising its premium da Vinci 5 placements.

The ASC market represents a significant untapped opportunity, and the XIR provides a pathway to penetrate it without sacrificing margins on the high end.

Expanding the Clinical Frontier

Intuitive received FDA clearance for several cardiac procedures on da Vinci 5 this quarter. Management emphasised a “measured rollout,” but this officially opens a new, high-acuity market with a TAM estimated at around 160,000 procedures annually.

The company also gained 510(k) clearances for its Single Port system in high-volume procedures including nipple-sparing mastectomy, hernia repair, and cholecystectomy. These clearances should support deeper penetration in 2026, we’ll see.

Management noted that their “line of sight” for addressable procedures has increased from around 7 million in 2024 to approximately 9 million today, driven by strengthening clinical validation in benign procedures and new clearances.

CEO David Rosa framed the company as being in the “early stages of this journey” despite the meaningful progress made. Another indication of the long runway for innovation beyond just procedural penetration.

Headwinds to Monitor

While the core business remains strong, management was transparent about several headwinds facing the company in 2026 in the conference call.

China Competition: Management explicitly called out intensified competition from local players, provincial preferences for domestic suppliers, and pricing pressure that impacted Q4 tender win ratios. This is a key market to monitor. The competitive dynamics in China are deteriorating, and while the company still has advantages, the trajectory here bears watching closely.

Japan Budget Challenges: Persistent government budget constraints continue to pressure capital placements in Japan. This has been an ongoing issue and doesn’t appear to be resolving quickly.

Margin Pressure: The impact of tariffs on gross margin is expected to nearly double in 2026 to 120 basis points of revenue (up from around 70 basis points in 2025). Management’s 2026 gross margin guidance of 67-68% is effectively flat YoY, as cost reduction efforts are being offset by tariff headwinds and product mix.

U.S. Healthcare Policy Uncertainty: Management noted potential impacts from changes to U.S. healthcare subsidies (ACA/Medicaid) as a risk factor. This is a newly articulated concern that could impact hospital finances and, consequently, patient procedure volumes.

Europe Macro Concerns: Broader macroeconomic pressures and capital budget constraints in Europe (particularly the UK) could continue to weigh on system placements.

What This Means for Valuation and Long-Term Returns

In my view, Intuitive remains one of the highest-quality compounding machines in the Healthcare technology industry. The business has a dominant market position, a widening technological moat, expanding clinical applications, and a highly predictable recurring revenue model. These are not things that change overnight.

That said, investors need to be realistic about what they’re underwriting. The 13–15% procedure growth guidance for 2026 is solid and steady, but it does represent a modest deceleration from historical levels, all while the stock trades at roughly 43x forward EV/EBITDA.

Operating expense growth of 11–15%, driven by higher R&D spending and distributor acquisitions, combined with largely flat gross margins, suggests operating leverage will be more muted than in prior periods.

As always with ISRG, it remains to be seen whether this outlook reflects normalization or the management team’s familiar conservatism in setting expectations.

The thing is that even under these conservative assumptions, Intuitive’s competitive position continues to strengthen. The da Vinci 5 extends its technological lead. The XIR system enables better market segmentation to address more cost-sensitive settings. Expansion into cardiac surgery and new Single Port indications reinforces the platform’s clinical utility.

With recurring revenue now accounting for ~81% of total sales, the business offers incredible quality and predictability. This is high-quality compounding at its finest.

What to Watch in 2026

For those of us monitoring Intuitive closely, here are the key questions to focus on:

1. Cardiac and Single Port rollout pace: How quickly do these new indications contribute to procedure growth? Early traction here would be a positive signal.

2. XIR system adoption in ASCs: What are the placement volumes and utilization rates? This could be a meaningful growth driver if executed well.

3. China competitive dynamics: Are tender win-rates stabilizing or continuing to deteriorate outside the US? This will tell us whether the competitive pressure is transitory or structural.

4. MyIntuitive Plus monetization: Beginning mid-year, the digital subscription package becomes a paid offering for renewing customers. The renewal rate and pricing will be important indicators of value delivery and pricing power.

5. Margin trajectory: Can the company offset tariff headwinds through operational efficiency, or will margins face continued pressure?

Thanks for following along,

—Nikotes

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Portfolio Corner - 🗓️ Monthly Updates (Last Update: 29-Mar-2026)

Subscribe now

Mercado Libre reported its third-quarter 2025 results this week, and the quarter perfectly illustrates the company’s aggressive, long-term mindset and why exceptional businesses can justify short-term margin compression.

Once again, short-term-focused investors fixated on the Direct Contribution margin decline as the company continues to sacrifice near-term margins in its largest market, Brazil.

The result? Re-acceleration of user growth and a clear signal that Mercado Libre is pressing its advantage to solidify its dominance for the decade ahead.

Financial Results

The Headline: Brazil

🔹 Top Line Growth Re-acceleration in Brazil (largest market) in Exchange for Short-term Margin Compression.

This quarter’s defining move was simple: dropping the free shipping threshold from R$79 to R$19 in Brazil. This, in my opinion, is a calculated bet, made by a long-term oriented management team, that their logistics network could absorb volume at scale better than anyone else. And they were right.

The results? User growth of 29% YoY in Brazil, the fastest pace since Q1 2021. Items sold accelerated sharply to 42% YoY (from 26% in Q2), and FX-neutral GMV grew 34%.

But here’s what really stood out to me: despite a 28% QoQ surge in shipment volume, unit shipping costs in Brazil decreased 8% QoQ in local currency. Let me repeat that, they added massive volume and their costs went down! This is the power of operating leverage in action, and it’s exactly the kind of competitive advantage that’s nearly impossible to replicate.

The Ecosystem Flywheel Accelerates Across The Board

What makes MELI’s model so compelling is how each segment reinforces the others. The shipping investment drove record traffic to the marketplace, which feeds their high-margin advertising business (Ads revenue grew 63% YoY) and provides a massive, low-cost acquisition funnel for Mercado Pago.

Speaking of fintech… The credit portfolio now exceeds 27 million users and $11 billion in size, growing at an 83% YoY clip. The thing is that management isn’t sacrificing quality for growth; 15–90-day NPLs remain stable at 6.8%. The Brazil credit card business tells the story best: roughly half the portfolio is now profitable, with cohorts older than two years consistently hitting profitability. This is disciplined scaling, not reckless expansion.

The “principality” metric, i.e. the share of users for whom Mercado Pago is their primary financial account, rose 11 percentage points YoY in Brazil. This matters because it shows deepening customer relationships, not just transaction volume.

Beyond Brazil: The Broader Picture

While Brazil grabbed headlines, the other geographies delivered solid results. Mexico posted another strong quarter with items sold up 42% YoY and FX-neutral GMV up 34%. Argentina remained resilient despite election-related turbulence, with items sold up 34% YoY.

The geographic diversification is underappreciated. MELI isn’t dependent on any single market, and they’ve demonstrated an ability to navigate macro volatility across multiple countries simultaneously.

Management’s Long-Term Mindset

Conference call commentary was, once again, honest and straight forward regarding how they approach managing the business:

“We are not managing the business for short-term margin. We are managing for long-term value creation... We are not going to hesitate to invest behind that even if we put some short-term margin pressure.” — CFO Martin de Los Santos.

On the Brazil investment, Commerce President (future CEO) Ariel Szarfsztejn noted they’re seeing “amazing growth, higher frequency from our buyers, record conversion rates, record retention rates.” When asked about competitive rationality, his response was pointed: “Just look at the results... That is a very rational move that significantly strengthens our competitive position.”

To me, this is what intelligent capital allocation looks like. Management identified an opportunity to lock in a new cohort of users and deepen their structural advantages, and they executed without hesitation.

The Margin Question

Let’s address the elephant in the room: Direct Contribution margin in Brazil declined YoY. This will bother some investors, and rightfully so. But context matters.

MELI is generating 39% revenue growth at a $16.5 billion quarterly revenue run rate. They’re adding 7.8 million new buyers to the platform (+26% YoY), users who will generate value across the entire ecosystem for years. The temporary margin compression is the price of securing long-term market dominance in a market where e-commerce penetration is still in the mid-teens.

The key question isn’t whether margins compressed, it’s whether the investment will generate returns above their cost of capital. Given the volume growth, cost efficiencies, and ecosystem effects we’re already seeing, I’d argue the answer is clearly yes.

Risks Worth Monitoring

I’m not blind to the risks here. Argentina’s macro instability remains a persistent wildcard: Q3 showed this clearly with slower consumption and higher funding costs.

The credit portfolio is growing explosively (83% YoY), and while NPLs are stable now, rapid growth in volatile markets always carries risk.

Competition in Brazil is “intensely competitive” (management’s words), which could require further margin-compressing moves. And there’s always execution risk when absorbing this kind of volume growth.

What Matters Going Forward

For investors, the key metrics to watch are:

1. Brazil items sold growth: Can they sustain the 40%+ pace? This would validate the aggressive investment thesis in exchange for short-term margin compression.

2. Unit shipping costs: Continued decline would confirm their logistics moat is widening

3. Credit portfolio NPLs: Any deterioration would be an early warning sign

4. Principality metrics: This determines whether fintech becomes a durable profit engine

The primary risks remain the same: near-term margin pressure from these strategic investments and the persistent macroeconomic volatility in Latin America.

However, this quarter was a textbook example of the powerful flywheel at work. By leveraging its scale to invest in a better customer experience, Mercado Libre is actively expanding its kingdom. This is a company that looks in control of its own destiny, and its long-term compounding story is intact.

Thanks for following along,

—Nikotes

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Portfolio Corner - 🗓️ Monthly Updates (Last Update: 30-Nov-2025)

Subscribe now

Cadence just posted what CEO Anirudh Devgan called their best positioning “in several years.” After going through the numbers, I tend to agree with a dose of caution.

This quarter reinforced Cadence’s status as an indispensable partner for anyone building AI infrastructure. Well… almost anyone, except China despite short-term normalization. That remains the wildcard, and it’s one factor out of Cadence’s hands.

Key Financials & Operational Metrics (Q3 2025)

Quarterly Highlights:

• Revenue: +10% YoY, ahead of +9% midpoint guidance

• Non-GAAP EPS: +18% YoY, beating +16% midpoint guidance

• Backlog: $7B, up 17% YoY and +9.4% QoQ (from $6.4B)
  → ~$150M (≈25%) was a catch-up from delayed China business (Q2)
  → Remaining $450M reflects organic strength across segments

• IP Business on track for FY2025 >20% growth, outpacing EDA business.

• Expect substantial shares buyback before year-end 👀.

  From the call: “We expect to use half of our free cash flow to repurchase Cadence shares”

The AI Design Tsunami

The primary story is straightforward: companies designing AI chips need Cadence’s tools, and they need them badly. And the best news is the “Design for AI” megatrend is already driving results today. The collaboration with OpenAI (using Palladium hardware) and the broad proliferation of Cadence’s tools at leading semiconductor companies for AI designs tell you everything you need to know about where the industry is heading.

The thing is that designing AI chips is fundamentally different from traditional chip design. These are highly complex system-on-chips (SoCs) with multi-die systems that require verification capabilities most tools simply can’t handle. This is why Cadence’s hardware verification business (Palladium and Protium) had a record Q3 and it’s becoming the “de facto choice for AI designs” because nobody else offer comparable capacity and performance.

The IP Business Stealing the Show

What really caught my attention this quarter was the exceptional performance of the IP business. This segment is firing on all cylinders, driven by three converging factors:

• Strategic portfolio focus: Cadence has concentrated its IP offerings on exactly the right areas - high-growth AI/HPC interconnects like HBM4, DDR5, and PCIe. These are key building blocks for AI infrastructure.

• Foundry diversification: The business is benefitting from new advanced-node foundry options beyond TSMC, including Samsung and Intel. This geographic and customer diversification reduces concentration risk.

• Competitive wins vs SNPS (?): They’re taking market share (SNPS warning last Q reinforces this assumption) with what appears to be a superior IP portfolio. The key HBM4 win at a marquee memory company is the type of design win that sets up years of recurring revenue.

The IP segment is evolving from a complementary business into a significant growth engine. I would even go as far as to say this could become Cadence’s most attractive business segment from a competitive positioning standpoint.

Management’s Confidence Is Palpable

CEO Devgan’s tone on the conference call was quite bullish, repeatedly emphasizing that Cadence is “uniquely positioned” to capture what he called a “generational opportunity” presented by AI. Maybe just corporate hyperbole… But what I really find compelling is that this confidence is backed by concrete customer evidence:

• Samsung achieved a 4x productivity improvement using Cadence 🔗 Cerberus AI Studio

• NVIDIA, Samsung, and Qualcomm highlighted 5x to 10x improvement in verification throughput using 🔗 Verisium AI

These do not seem marginal gains, they’re the type of productivity improvements that could justify premium pricing and drive rapid adoption. When your tools can deliver a 10x improvement in verification throughput, price becomes a secondary consideration, don’t you think?

Building Another Growth Pillar

The announced acquisition of Hexagon’s MSC Software (structural analysis) signals Cadence’s strategic push into System Design & Analysis (SDA). This is about positioning for what management calls the “Physical AI” wave: robotics, autonomous driving, drones - where system-level simulation becomes critical. Pretty exciting, right?

I view this as management proactively building a third billion-dollar-plus growth pillar before they need to. The core EDA business is strong but diversifying into adjacent high-growth markets while you’re winning looks like smart capital allocation. The MSC acquisition complements the existing portfolio and expands Cadence’s total addressable market significantly.

The China Wildcard

Following Q2’s export-control disruption, China business is back to normal… But is it?

Management highlighted strong design activity, including catching up on delayed hardware shipments, and expects YoY growth in FY2025. They also see little signs of customers pulling forward demand.

That said, I remain cautious. Geopolitical risk is high, and CDNS themselves noted in the call: “…the FY2025 outlook assumes China export controls remain similar for the rest of the year.” Plus, they are reducing the EPS forecast for Q4 due to China/US trade related tensions.

Remember that Cadance revenues mix includes a substantial part from China:

What I’m Watching

For the long-term thesis to continue playing out, I’ll be monitoring:

1. Backlog trajectory in Q4: Continued growth would set up another strong year in 2026

2. Hexagon’s MSC Software integration: Successful execution here validates the SDA expansion strategy into the “Physical AI” opportunity.

3. IP business margins: As this segment scales, the margin profile becomes increasingly important

4. Geopolitical stability: Any shifts in U.S./China regulatory landscape could impact growth

The Bottom Line

The raised guidance and continued record backlog tell you that this momentum isn’t fleeting. Management’s commentary about exiting FY25 with a record backlog “positions us very well for next year” suggests that they’re confident about maintaining double-digit top-line growth going into 2026.

What matters most is that Cadence is executing at an elite level across all business segments simultaneously. The comprehensive portfolio from silicon-to-systems, combined with AI-infused software that delivers measurable productivity gains, creates a widening competitive moat.

The contrast with more cautious commentary from competitors (SNPS) suggests Cadence is taking market share, particularly in the IP business. When you’re winning in the most strategic areas of AI chip design while your competitors are hedging, that’s the type of competitive positioning that justifies premium valuations.

Thanks for following along,

—Nikotes

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Portfolio Corner - 🗓️ Monthly Updates (Last Update: 30-Nov-2025)

Subscribe now

Intuitive Surgical reported its Q3 2025 results and it turned out to be a “beat and raise” quarter that should put to rest any lingering concerns about a slowdown in the business.

While the headline numbers were impressive, 23% revenue growth and 30% pro forma EPS growth, the real story is about the da Vinci V system proving it’s not only an incremental upgrade but a catalyst for the next multi-year phase of growth.

The procedure growth engine is key, and it is firing on all cylinders. It validates the core thesis that the adoption of robotic-assisted surgery is continuing and accelerating.

Headline Numbers

Strong across the board:

• Total Procedure Growth: Accelerated to 20% year-over-year from 18% in FY24.

• Revenue Growth: 23% increase, beat driven by both systems and procedures.

• Pro Forma EPS Growth: A 30% increase to $2.4 (it includes a $0.12 benefit from tax item).

This performance gave management the confidence to raise full-year 2025 guidance for both procedure growth and gross margin, showing their optimism for the near-term momentum of the business.

The Key Catalyst: The da Vinci V Adoption

The star of the quarter was the successful launch of the new da Vinci V system. V is no longer a “show me” story but the main growth catalyst, and its impact is being felt across the entire business.

1. Multi-Year Upgrade Cycle: In the U.S., 141 hospitals are upgrading from older systems compared to just 38 a year ago. Hospital capital budgets are being prioritized for robotic surgery, especially for technology that enhances efficiency.

2. Increasing Utilization: Management noted that da Vinci V’s utilization is already outpacing the previous-generation Xi system at the same point post-launch. U.S. utilization ticked up 2% this quarter, a positive inflection after the flat trends we saw earlier in the year.

3. International Expansion: First da Vinci V placements in key international markets of Japan and Europe, marking the beginning of its global rollout.

During the CC, Management noted new FDA clearances for its software, which enables remote updates and in-console features like “force gauge” and 3D model viewing. This reinforces the strategy of using the V as a springboard for future digital and AI-driven surgical insights.

Geographic Performance

🔹 US Growth

Intuitive placed 263 da Vinci systems (up from 219 last year), driven primarily by V upgrades.

The +18% total procedure growth was not driven by a single area but broad-based across general surgery categories: cholecystectomy, appendectomy, hernia repair, plus strength in acute care surgery.

🔹 International markets

While procedure growth was exceptional at +25% for da Vinci, system placements remained roughly flat at 164 units. Management pointed to “ongoing government budget challenges in Japan and the UK, plus a competitive environment in China”. However, this was offset by strength in India, Korea, Brazil, and distributor markets.

The reality is that international penetration remains far below U.S. levels, which creates a big long-term runway. OUS utilization grew 8% this quarter, suggesting the installed base is increasingly productive even as new placements face near-term headwinds.

The ION and SP Opportunities

While da Vinci is the key driver, the ION platform continues its growth trajectory with 52% procedure growth. This lung diagnostics system provides an opportunity that remains in early days but represents a potential TAM expansion beyond traditional surgical robotics.

The SP (Single Port) platform saw 91% procedure growth, led by Korea, and is awaiting key U.S. regulatory clearances in general surgery. Once those approvals come through, SP could unlock another wave of growth in procedures where single-port access offers meaningful clinical advantages.

Headwinds

While the quarter was very positive, management was transparent about a few targeted headwinds they are navigating:

• The impact of GLP-1 weight-loss drugs continues to pressure bariatric procedures, but this is now a small and shrinking part of the business, representing less than 3% of total procedures.

• In certain international markets like the UK and Japan, government budget constraints are creating a challenging capital environment, a key area to monitor.

• Gross margin declined 110bps YoY due to the mix of new, lower-margin products and higher service costs. Management raised guidance here, but investors should watch whether scale and cost reductions can offset mix pressure over time.

• While Medtronic’s Hugo and J&Js Ottava haven’t materially impacted Intuitive’s dominance yet, their progress remains a factor to monitor.

Investor Takeaways

This was a high-quality quarter that reaffirmed the investment thesis.

• Growth Story is Accelerating: The jump to 20% procedure growth is key, underlying demand for robotic surgery.

• The da Vinci V is Proving Itself: The new system is successfully driving an upgrade cycle and increasing the value proposition for hospitals, solidifying Intuitive market leadership.

• Portfolio Strategy is Expanding: Management’s plan to offer refurbished Xi systems alongside the premium V. A wise move, as it will allow them to address the entire market, from high-end academic centers to more cost-sensitive community hospitals and ambulatory surgery centers (ASCs), expanding their total addressable market. The ION and SP opportunities remain interesting optionality worth monitoring closely.

• The International Opportunity: Despite near-term budget issues in a couple of countries, the 25% procedure growth outside the U.S. highlights the big runway that remains as the rest of the world adopts this technology.

For investors, the key is recognizing that Intuitive is reshaping how surgery is performed globally. The combination of an embedded installed base, high margin recurring revenue, and multiple vectors for growth creates one of the most attractive business models in healthcare.

Yes, the valuation is rich. But when you’re looking at a business with 20%+ procedure growth, expanding margins, and a decade-plus runway for international adoption, perhaps the right question isn’t “Is it expensive?” but rather “What would you pay for a 20-year growth compounder with a wide moat in healthcare?”

I dropped a new Deep Dive Brief a few days ago on Intuitive Surgical.
If you want to go beyond the headlines and understand the full investment case, it’s still free to read for one more week:
🔗 Deep Dive Brief: Intuitive Surgical

What To Watch Next

1. Da Vinci V adoption in Europe and Japan: International rollout will be critical for sustaining growth as U.S. replacement cycle matures

2. Sustainability of high-teens U.S. procedure growth: Can this momentum carry into Q4 and 2026?

3. Gross margin trajectory: Will scale offset the mix headwinds from newer, lower-margin products?

4. Single-Port (SP) regulatory progress: U.S. clearances in general surgery could unlock meaningful growth

5. Hospital capital spending trends: Current strength is impressive, but healthcare capital cycles can turn quickly

Thanks for following along,

—Nikotes

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Full Portfolio Content - 🗓️ Monthly Updates (Last Update: 30-Nov-2025)

Subscribe now

This was a quarter that separated the signal from the noise. The noise looks like a temporary, cyclical dip in volumes. The signal is the strength of Copart's competitive moat.

Let's unpack the results.

Financial Performance Table

The Headline: Profitability Shines, Volume Softens

A key takeaway from the quarter is that Copart delivered double-digits growth where it matters most: profitability.

• Net Income: Up 22.9% YoY to $396.4 million.

• Gross Margin: Expanded by 280 basis points to 45.3%.

• Operating Margin: Grew by 310 basis points to 36.5%.

This performance was driven by two key factors. First, the company demonstrated excellent operational leverage. Second, and more importantly, they achieved significant growth in Average Selling Prices (ASPs), which rose 5.7% in the U.S. This key metric more than offset a surprising softness in Q4 U.S. insurance unit volumes, which declined by 2.1%.

Deconstructing the Volume Headwind

The decline in U.S. insurance volumes was the primary point of discussion on the earnings call. Management was transparent about the drivers, attributing the weakness to cyclical macro factors rather than a structural issue with the business.

1. The Underinsurance Trend: A 4.3% decline in "earned car years" (a proxy for insured vehicles) versus only 1.3% growth in the vehicle population. Simply put, consumers are dropping collision coverage due to higher insurance premiums which have been ramping up. When people don't have comprehensive coverage, there are no total loss claims to process.

2. Insurer Activity: Management also noted the normal "ebbs and flows of business activity among individual auto insurance carriers" as a contributing factor. So, business fluctuations among individual carriers contributed to the softness. This is the kind of quarterly noise that's inherent to any business with concentrated customers.

This still looks like a headwind, but I view the underinsurance trend as cyclical. As insurer profitability improves and premium inflation cools, this pressure should eventually abate.

Management seemed refreshingly transparent about these headwinds. Copart’s CEO did not spin this as anything other than what it looks: a cyclical downturn that doesn't change the fundamental thesis.

The Core Thesis is Intact: All About Auction Liquidity

In my view, the most important part of the earnings release was management’s confident articulation of Copart's core competitive advantage. CEO Jeff Liaw spent some time in the CC highlighting what separates them from the competition: superior "auction liquidity".

He stated, "the critical value we provide sellers... is that our auction platform will find the highest and best use of every vehicle anywhere in the world."

This is the heart of their moat along with their physical network of salvage yards. Copart's global buyer base of 300,000+ members create intense competition for every vehicle, which drives up selling prices. This, in turn, generates higher net returns for their insurance clients.

Management also delivered a “knockout punch” with a direct competitive statement. They noted their ASPs "grew at a rate more than fivefold that of service providers similar to ours."

This is a clear signal to the market and their clients that Copart is a premium partner that delivers superior financial outcomes, creating value for insurance clients. This is how you win and retain business over the long term.

About the Margin Expansion

Another net positive about these results: gross margins expanded 280 basis points to 45.3% and operating margins hit 36.5% up 310 basis points despite lower volumes. This demonstrates the high operating leverage inherent in Copart's business model. Fixed costs get spread over fewer units, yet margins still expanded because of pricing power (higher ASPs) and operational efficiency.

International Growth Offsetting Domestic Weakness

While U.S. volumes declined, international units’ volume grew 3.3% in the quarter and 8.1% for the full year. The shift from purchase contracts to consignment models in markets like Germany is driving higher-margin revenue growth. This geographic diversification provides a partial hedge against cyclical downturns in any single market.

Capital Allocation Flexibility… Maybe Too Much?

With $4.8 billion in cash and securities plus total liquidity of $6 billion, Copart has enormous financial flexibility, maybe too much?...

My guess: management still favors buybacks as the main capital return lever, but they won’t pull the trigger while multiples stay elevated. M&A? Still a wildcard.

What to Watch Going Forward

The key question is whether Q4's volume weakness represents a cyclical trough or the beginning of a more persistent trend. I'm watching three things:

1. Q1 2026 U.S. insurance volumes: Is this a one-quarter anomaly or something more persistent?

2. Insurance industry dynamics: With combined ratios improving, there may be more competition for policy growth, which could eventually benefit claims volumes.

3. ASP sustainability: Can Copart continue outpacing used vehicle indices, or was this quarter's 5.7% growth a one-time thing?

The Bottom Line

This quarter reinforces why Copart is a quality company. The combination of pricing power, margin expansion, and structural industry tailwinds creates a business model that can generate attractive returns even during periods of volume softness.

The reality is that Copart's competitive moat, with their global auction platform and superior liquidity, becomes more valuable during challenging periods, not less. Insurance companies need maximum recovery rates when they're under pressure, which plays directly into Copart's value proposition.

For long-term investors, quarters like this should increase conviction rather than diminish it. Any business can grow when volumes are tailwinds. Only exceptional businesses can expand margins and grow profits when volumes are headwinds.

Thanks for following along,

—Nikotes

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Full Portfolio Content - 🗓️ Monthly Updates (Last Update: 11-Sep-2025)

Subscribe now

Heico delivered another quarter that showcases why this aerospace compounder deserves its premium valuation. Net sales grew 16% to $1.15 billion, and the group showed even better operating leverage with operating income up 22% and net income up 30% to $177.3 million, or $1.26 per share as the company's razor-and-blade model keeps firing on all cylinders.

Performance Drivers

1. Flight Support Group (FSG): The Growth Engine

The star of the show remains the FSG, which posted its twentieth consecutive quarter of sequential sales growth. Think about that for a moment: five straight years of growth in the commercial aerospace aftermarket, weathering everything from supply chain disruptions to post-pandemic recovery volatility.

The numbers:

• Revenue grew 18% to $802.7 million. Contribution from 13% organic growth. This organic growth figure is particularly positive because it strips out the acquisition noise and shows the true demand strength in their core markets.

• But what really caught my attention was their operating income growing 29% to $198.3 million. That's serious operating leverage, with margins expanding from 22.5% to 24.7%. When you're selling aftermarket parts to airlines with a growing number of old fleets that need their planes flying, pricing power shows up in exactly this way.

2. Electronic Technologies Group (ETG): Solid 10% Growth with Margin Pressure

The ETG delivered but its profitability story was more subdued.

• Revenue increased 10% to a record $364.5 million, with 3% contributions from acquisitions and a 7% increase in organic sales.

• Operating Margin compressed to 22.8% from 23.5%. Management attributed this decline to an increase in performance-based compensation expenses, probably higher bonus accruals.

Profitability Paired with Lower Cash Flow Conversion

Unlike companies where GAAP earnings can be misleading, Heico's reported profit closely reflects its operational performance. The 30% increase in net income was driven not only by higher sales volumes but, more importantly, by substantial margin expansion within the larger FSG segment.

Cash generation was okay, with Cash Flow from Operations up 8% to $231.2 million. The slower growth versus net income was largely due to a sizable increase in Net Working Capital. While cash flow conversion bears watching, I’m not overly concerned given management’s strong track record in capital allocation.

The Capital Allocation Story

What stands out most about Heico is its ability to absorb the large Wencor acquisition while still deleveraging. Over the past nine months, Net Debt to EBITDA improved from 2.06x to 1.90x, giving them firepower for their next wave of acquisitions.

This is textbook disciplined capital allocation, funding growth without overextending leverage, while generating enough cash to simultaneously reduce debt.

The Aerospace Aftermarket Thesis Intact

These results reinforce why I remain bullish on the aerospace aftermarket thesis. Once airlines commit to specific aircraft platforms, they’re locked into the parts ecosystem for decades. Heico’s FSG capitalizes on this dynamic through its Parts Manufacturer Approval (PMA) business, producing FAA-approved alternatives to OEM parts at superior margins. I broke this down in detail in my recent Heico investment thesis: 🔗 Deep Dive Brief: Heico

FSG’s twenty consecutive quarters of growth shows the resilience of aftermarket demand. Airlines may delay new aircraft purchases during downturns, but they can’t defer maintenance on their existing fleets.

What to Watch Going Forward

Key metrics I’m tracking are straightforward:

• FSG organic growth sustainability: Can they maintain double-digit organic growth as comps get tougher?

• Acquisition pipeline execution: With leverage improved, they have room to reaccelerate M&A following the Wencor acquisition in 2023.

• ETG margin trajectory: Is the compensation-driven margin pressure temporary, or does it show something more structural? Ideally, this segment sustains low double-digit growth.

Aerospace recovery cycles tend to be long and durable once underway, and we may still be in the early innings of post-pandemic normalization.

Valuation Context

At current levels, Heico trades at premium multiples that clearly reflect its quality and growth (47x FCF NTM and 31x EBITDA). The 30% net income and EPS growth this quarter shows why investors are willing to pay up, even Buffett picked up shares in Q2 2025. When the model works, the operating leverage is substantial.

In my view, businesses with genuine moats in expanding end markets deserve premium valuations. Heico’s combination of aftermarket defensibility, disciplined M&A, and best-in-class capital allocation places it in that camp.

You won’t get it cheap. But if the stock dips, I’d better be ready. Quality compounders rarely stay discounted for long.

Thanks for following along,

—Nikotes

Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Full Portfolio Content - 🗓️ Monthly Updates (Last Update: 30-Nov-2025)

Subscribe now

Constellation Software just delivered what I'd call a "textbook CSU quarter", one that shows really well why focusing on GAAP earnings for serial acquirers is a fool's errand. While reported EPS decreased 68% to $2.66, the underlying cash generation shows a different story.

The Numbers That Matter

Let me cut straight to what matters: Cash Flow from Operations surged 63% to $433 million, and Free Cash Flow Available to Shareholders grew 20% to $220 million. These are the metrics that determine CSU's ability to compound capital over the long term, and they're well on track with past performance but, at a higher scale.

Revenue growth of 15% was driven by the usual suspects, about 10% from acquisitions and 5% organic growth. The organic number is particularly encouraging, with maintenance and recurring revenue (now 75% of total revenue) growing 7%.

This is the sticky, predictable cash flow that makes CSU's acquisition model so powerful.

The GAAP Earnings Noise

The 68% decline in net income is almost entirely noise. Here's what happened:

• Foreign exchange losses: $118 million vs. $4 million last year. This is primarily non-cash revaluation of foreign-denominated liabilities, i.e. FX headwinds.

• IRGA/TSS liability revaluation: $126 million charge vs. $8 million last year, tied to Topicus performance. A revaluation that is non-cash but flows through P&L statement. This is a big one-off (or at least period-specific) expense.

• Higher amortization expenses: $29 million increase from recent acquisitions.

These are exactly the types of non-cash accounting adjustments that make GAAP earnings misleading for understanding CSU's true economic performance. The thing is that none of these items affect the company's ability to generate cash or reinvest it profitably.

Why This Quarter Reinforces the Investment Thesis

What I find most compelling about these results is how they demonstrate CSU's operational resilience. Q2 is typically their weakest quarter for cash flow seasonality, yet they delivered exceptional cash generation relative to other quarters. This shows CSU’s portfolio businesses are performing well despite all the accounting noise.

The company's balance sheet remains fortress-like with $2.58 billion in cash and over $1 billion available in credit facilities. They've got more than enough dry powder to continue their acquisition spree, having already deployed $474 million in H1 and committed another $320 million post-quarter.

Borrowing a great chart from X user @wlep, which shows the rolling 4-quarter cash deployment into acquisitions for consolidated CSU, Core Constellation Software (CSI), and the spin-offs (TOI & LMN).

As you can see, despite a slow start to 2025, core CSI is beginning to pick up steam again while getting a solid boost from its spin-offs, particularly Topicus in recent quarters.

The Scale Question Remains

While these results are encouraging, they don't change the fundamental question many have been wrestling with: can CSU maintain its historical returns as scale increases?

The company deployed $380M in acquisition capital this quarter, down from $452M a year ago (-16% YoY). This could signal tougher market conditions and the growing challenge of sourcing attractive deals at their scale; or it might simply be a matter of timing, as acquisition opportunities rarely arrive in a straight line. Personally, I lean toward the latter explanation.

In any case, the reality is that as CSU continues to scale, it faces the unavoidable mathematical constraints of large-scale capital deployment. Even with robust cash flow growth, sustaining 20%+ IRRs on an ever-expanding capital base becomes progressively harder.

What to Watch Going Forward

I'm focused on two key metrics:

1. FCFA2S growth trajectory: The 20% growth this quarter is solid, but can they sustain this as the capital base grows?

2. Acquisition pipeline quality: The slight decline in acquisition spending isn't necessarily concerning, but I want to see if they're maintaining pricing discipline or struggling to find attractive deals.

The Bottom Line

This quarter perfectly illustrates why CSU remains one of the highest-quality compounding machines in public markets, but also why investors need to think carefully about valuation. The cash generation is exceptional, the business model is intact, and management continues to execute flawlessly.

However, I'd caution against extrapolating historical returns indefinitely. The so-called ‘smart money’ is modeling gradual return compression from the current 20%+ IRRs toward something more sustainable in the 15-20% range over the next decade. Even at those levels, CSU would remain an exceptional business, just not the perpetual 20%+ compounding machine some investors expect.

One thing that gives me more comfort as a CSU shareholder is Mark Leonard and his team’s extraordinary ability to execute almost flawlessly, regardless of market conditions. As a firm believer in their skills as exceptional capital allocators, I’m confident entrusting them with my capital, knowing they’ll continue working to deliver outstanding shareholder returns, even if the pace inevitably moderates over time.

For long-term investors, quarters like this reinforce why CSU deserves a premium valuation. Just make sure that premium is based on realistic expectations about the future, not a mirror of the past.

Thanks for following along.

—Nikotes

Subscribe now

Let’s start with the headline numbers, then we will dig deeper into what they mean for MELI and long-term investors down the line.

Key Financials (YoY % Growth)

Thoughts:

Margins compressed as MELI leaned into strategic investments, particularly in 1P logistics and free shipping in Brazil. But the trade-off seems smart given the traction across core metrics.

Geography Check

Thoughts:

Argentina stood out thanks to strong macro tailwinds and execution. Brazil saw lower margins due to macro environment (higher interest rates) and MELI’s aggressive free shipping moves, but as we’ll see, this is part of a broader strategy.

eCommerce: Still the Engine

• Active buyers: +25% YoY

• GMV: +21% overall (FXN +37%)

• Supermarket vertical: +65% items sold

• 0–1-day shipments: +28% to 162.9M

• Listings: 536M (+20%)

Commerce continues to lead MELI’s flywheel. Fulfillment surpassed 60% in Brazil, helping reduce order costs. The playbook—improve UX, broaden product selection, lean into logistics—still works.

Management Commentary:
“We are convinced that the best way to serve our customers in Brazil is by offering more free shipping… this creates long-term value.”

They cut free shipping thresholds again (BRL 79 → BRL 19), a move aimed at boosting engagement, frequency, and market share.

Fintech: On ‘Scaling’ Mode

Fintech now starting to rival ecommerce as MELI’s crown jewel. They’re offering 6–7% yields with daily liquidity in the wallet, beating most local banks. And lending is becoming a real profit center.

Management noted:

"Credit card profitability is improving across cohorts, and Brazil is already breakeven at the NIMAL level. Defaults are stable, with longer delinquencies up slightly but within manageable bounds.

Ads: Becoming a Margin Accretive Pillar

• Ad revenue: +38% USD / +59% FXN

• Display & video: Nearly doubled YoY

• New Smart TV App (Mercado Play): Potential future ad inventory engine

MELI is leaning into ads—with UX improvements and smarter seller tools (budget suggestions, AI-based item selection, better attribution). They also launched integration with Google Ad Manager and AdMob, allowing off-platform targeting with MELI’s rich first-party data.

Management Commentary:

“We see ads as a percentage of GMV accelerating… Good news, but the opportunity is huge ahead of us.”

Ecosystem Flywheel in Motion

What ties it all together is scale and integration. Sellers borrow from MELI. Shoppers pay with MELI. Ads are sold on MELI. And logistics is the moat gluing it all together.

Despite lower P&L margins in the short term (especially in Brazil) the decisions made this quarter are deeply strategic. Management is trading margin for future dominance, confident that volume-driven efficiency will eventually pay off.

Final Thoughts

It was a solid quarter, not because every number was perfect, but because the ecosystem is humming.

Revenue grew 34%, EBIT rose 14%, and yet margin compression tells you MELI is still heavily investing. They pulled the shipping lever again in Brazil (cutting the free shipping minimum from BRL 79 to BRL 19), a move that will hit short-term margins; but, in my view, one that might prove brilliant over the long term. This is MELI playing offense, not defense.

Fulfillment reached 60%+ penetration in Brazil, helping cut per-order costs. Meanwhile, Argentina did some heavy lifting this quarter, balancing weaker operating leverage in Brazil and Mexico. FX-neutral GMV growth was especially strong in Argentina (+126%), with items sold up 46%.

In fintech, the story continues to compound. Wallet MAUs hit 64M (+31%), and users are sticking around because of yield-on-deposit features. These keep converting passive users into buyers and credit users.

The credit book hit $9.3B (+91% YoY) and metrics suggest healthy stability: short-term delinquencies are down, long-term delinquencies are stable, and profitability is improving. MELI’s ecosystem advantage shows here: lending to users that already shop and sell on their platform.

Ads are picking up real momentum. Revenue grew 50% (59% FXN), and product ads and display doubled YoY. The launch of Mercado Play (a smart TV app) allows MELI to build ad inventory on its own rails. As a user-first company with rich 1P data, they are well positioned to build a full-funnel ad platform, especially after the Google Ad Manager integration.

Capex and credit growth drove a small free cash flow dip this quarter (-$10M), but MELI is clearly still in build mode. With 85% of LATAM retail still offline, the TAM is wide open and MELI is investing because it knows that.

To wrap it up:

There’s a reason MELI continues to outperform in one of the world’s most challenging markets. They’re not playing short-term games. They’re building infrastructure, investing into scale, and creating leverage across verticals. With 85% of LATAM retail still offline, MELI’s opportunity is far from priced in.

Yes, margins took a hit. But when that hit comes with +34% revenue growth, +55% OCF growth, and market share gains across commerce, fintech, and ads, it feels like the right call.

Still early innings. Still best-in-class.

Thanks for following along.

—Nikotes

Subscribe now

📰 What’s New at Expanse Stocks?

📊 Quarterly Update - Portfolio Composition (by Industry and Geography) + Valuation Metrics → 🔗 Behind-The-Scenes [Free access]

📚 Articles

🔎 Deep Dive Briefs
⛅ Cloudflare | 👷‍♂️ Parsons | 𓇲 MPS | 🥼 Medpace | 🔌 Arista | 🛒 Amazon | 🤖 ASML | 🦎 Topicus | 💡 Lumine | ✈ HEICO

💸 General Investing
💎 Hidden Gems Series
✨ Annual Specials – Annual Letters, Investing Philosophy & Top Picks of the Year
📚 Resources for Investors
📢 Latest Stock News
💬 Join My Chat → [📎 Learn more]

👀 Recent Releases!

• 💆‍♂️ A Reflection on my Investing Mistakes

• 💸 Optionality: Hidden Engine of Long-term Compounding

• 💸 Why Equities Win Against Other Assets

👀 Coming Soon

• 💎 Hidden Gem - Fifth Edition

➕ Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

🔗 Full Portfolio Content - 🗓️ Biweekly Updates (Last Update: 11-Sep-2025)

1. 📈 Biweekly Report – Portfolio updates, recent moves & watchlist additions

2. 💼 Full Portfolio Access – Holdings, valuation metrics & performance tracking

3. 📊 Valuation Tools – DCF, reverse DCF, capital efficiency & growth models

4. 🎧 Podcast Picks – Biweekly handpicked episodes on business & investing

5. 🔎 Stock Picking Framework – My methodology & investing philosophy

6. 🎯 Swing Trading – Short-term strategy & latest swing trades

7. 💬 Private Community Chat - 📎 Learn more

8. 🔐 Full access to all my articles

Financial Snapshot: Q2 2025

Key Highlights:

🔹 Net Sales: €7.69B (Est. €7.51B) 🟢 | EPS: €5.90 (Est. €5.22) 🟢

🔹 Gross Margin: 53.7% (Est. 51.9%) 🟢

🔹 Net Bookings: €5.54B (Est. €4.80B) 🟢; -1% YoY and +41% QoQ → For those still focused on QoQ net bookings beats, this Q delivered. But if, like me, you're paying closer attention to total backlog, the real picture is starting to come into focus. I'll dive deeper into that in the Outlook & Guidance section and expand on it in the Final Take below.

🔹 Net Income: €2.29B; UP +45% YoY

CC Breakdown & Management Themes

AI is the Unquestionable Growth Engine

CEO Christophe Fouquet was unequivocal: “Artificial intelligence is currently the main driver for growth for both Logic and Memory.” This theme was the central pillar of the call. Demand from customers adding capacity in the most advanced nodes for AI applications is fueling growth in Logic, while Memory remains robust due to investment in HBM and DDR5. This is directly reflected in the bookings, where Logic accounted for 84% of the €5.5 billion total in Q2.

EUV Business Strength & The Shift from Multi-Patterning to Single Exposure

A core theme was ASML's role in improving its customers' manufacturing economics. As chip features shrink, fabricating a single layer on a wafer can require multiple complex exposure steps (multi-patterning). This process is expensive, time-consuming, and increases the risk of defects.

ASML’s new, more productive tools are key to solving this. CEO Fouquet explained that the higher throughput of the new NXE:3800 Low-NA EUV tool "allows customers to shift more multi-patterning layers to single exposure," which in turn helps them "reduce complexity, reduce yield loss and improve cycle time."

CFO Roger Dassen provided details of the outlook for the Extreme Ultraviolet (EUV) business, projecting "approximately 30% increase of the EUV business" for the full year. This growth is being achieved through higher throughput from the new NXE:3800 tools, allowing customers to meet their “30% extra capacity” needs with a similar number of Low-NA systems as last year.

High-NA Progress & Margin Impact

On the technology front, ASML is making steady progress with its next-generation platform, now working with three High-NA customers. The company shipped its first EXE:5200 High-NA system, the tool intended for high-volume manufacturing.

However, management was clear that these early systems are margin dilutive in the short-term. The recognition of more High-NA tools in the second half is a key reason for the guided-down gross margin, which is expected to be around 50% for H2 2025.

On the other hand, according to CEO Fouquet the new EXE:5200 High-NA system that will be used in production will achieve a “60% increase in efficiency compared to the EXE:5000 High-NA currently used by customers’ R&D teams”. This efficiency gain is key to justify the need for this technology and marks a critical milestone in preparing customers for the next generation of chipmaking.

Geopolitical and Macro Uncertainty Remains an Overhang

While the technological demand picture is clear, both executives highlighted growing uncertainty for 2026, coming from “macroeconomic and geopolitical consideration... and that includes, of course, tariffs.” Management claims they are working to mitigate the direct impacts of potential tariffs on systems and parts shipped to the US but acknowledged that the indirect macroeconomic impact remains "very uncertain."

Business & Regional Highlights

Q2 saw some interesting shifts in the sales mix compared to the prior quarter:

• By End-Use: The strength in AI was evident as Logic system sales grew 16% sequentially, accounting for 69% of net system sales, while Memory sales declined 28% after a very strong Q1.

• By Technology: In a notable reversal from Q1, DUV system sales grew 19% sequentially, while EUV system sales declined by -16.5%. Despite shipping fewer EUV units (11 in Q2 vs. 14 in Q1), the EUV Average Selling Price (ASP) increased 6% q/q to approximately €244 million, reflecting a richer mix of the new, more productive NXE:3800 tools.

• By Region: The geographic mix algo changed substantially. Sales to Taiwan (TSM) more than doubled sequentially, making it the top region at 35% of system sales. In contrast, sales to South Korea (-54%) and the US (-39%) saw sharp declines from a high base in Q1. China remained a stable contributor at 27% of sales.

Outlook & Guidance

ASML reaffirmed outlook for both the upcoming quarter and the full year, as well as its long-term targets.

Key Highlights:

🔹 FY25 Revenue Growth: ~15% YoY (Prior: €30–35B) → unchanged

🔹 FY25 Gross Margin: ~52% (Prior: 51%–53%) → unchanged

🔹 Q3 Revenue: €7.4B–€7.9B (Est. €8.21B) 🟡

🔹 Q3 Gross Margin: 50%–52% (Est. 51.4%) 🟡

Bookings and Forward Visibility: Bookings and the resulting total backlog now provide approximately 80% coverage for 2026 revenue estimates, suggesting 2026 will be another growth year, though management stopped short of an official commitment.

Long-Term (2030) Outlook: The 2030 targets of €44-€60 billion in revenue and a 56-60% gross margin remain intact.

What am I making of it, and where does my conviction stand on ASML’s long-term prospects?

Read more

📎 Full Portfolio Content - 🗓️ Biweekly Updates (Last Update:15-May-2025)

Subscribe now

Key Financial Metrics

Regional Performance

➕ Positives

• Revenue Growth at +27% and RPO Growth at +39%. Very solid and consistent.

• Free Cash Flow growth machine, +50% YoY. Expecting 2/3 of FCF generation for H2 2025.

• Large Customer and Paying customer, mid-twenties % growth. Also, solid and showing consistency.

➖ Negative

• Margins flattish (Operating margin) or declining (gross margin). Will discuss this, later on, in the Risks section.

• Net Dollar Retention flattening at 111%, good but not spectacular as it used to be. Something to monitor, as long as it stays at these levels.

• Stock-Base Compensation (SBC) still at outrageous levels, 20+% of total revenues. Diluting shareholders and reducing GAAP margins.

CFO Thomas Seifert on guidance:

“We’re taking a prudent approach given macro uncertainty, but our strategy of reinvesting in sales and developer traction is working.”

Enterprise Momentum & Record Deals

Closed the largest deal in company history: $130M "pool of funds" contract powered by Workers AI.

Strong traction in Zero Trust and SASE solutions with government and global enterprises.

Multiple multi-year contracts signed, including 7-year commitments.

Sales productivity improved QoQ, with pipeline ahead of forecasts.

CEO Matthew Prince:

“We are winning not just on cost or speed but on platform capability. Our largest deals increasingly represent full-stack adoption of Cloudflare.”

AI & Developer Platform

AI-related traffic exploded 👀

• +4000% YoY increase in Workers AI inference requests

• +1200% increase in AI Gateway requests

🔗 Model Context Protocol—open standard that connects AI systems with external applications— deployments gaining traction: partnerships with Stripe, Atlassian, PayPal.

Cloudflare is positioning as a key AI infrastructure layer, combining security, inference, and low-latency edge compute.

CEO Matthew Prince:

“80% of major AI firms are already on Cloudflare. We're not just adjacent to AI, we’re becoming essential to it.”

Cybersecurity

Record sales in Zero Trust and SASE, with key competitive wins over legacy providers.

Government and enterprise adoption accelerating due to hardware-to-cloud shifts and regulatory mandates.

“Security is moving from boxes to the edge. Tariffs and volatility are just speeding that up.”

Network Infrastructure Spending & Tariffs

Cloudflare continues expanding global network capacity without CapEx overshoot. Network CapEx held steady at 17% of revenue but FY25 expected to decline at 12–13%.

Tariff and geopolitical risks are mitigated via multi-vendor, multi-region infrastructure strategy.

“Legacy hardware vendors can’t match our global flexibility and customers are noticing.”

Main Risks From CC

Management mentioned “slower revenue recognition from “pool of funds” deals may obscure short-term growth visibility”.

Slight compression in gross margin due to increased traffic from high-value customers.

Uncertainty around macro remains high across global markets, hence management prudent guide vs estimates.

Final Take

Another strong Q, with high-quality growth in the areas that matter the most (AI, SASE and Zero Trusts), showing operating leverage, and growing relevance in enterprise.

The company's developer platform and Workers AI are fast becoming differentiators, and its ability to close more and more multi-million-dollar, multi-year contracts reinforces its competitive edge.

Cloudflare’s infrastructure efficiency and full-stack architecture continue to make it a rare combo player in security, edge compute, and AI.

Bottom Line: Long-term tailwinds remain intact. Execution and product depth keeps justifying the premium valuation. Keep an eye on large deal velocity and AI monetization in 2H FY25.

No PT, hard to model long-term given their high-growth, high-multiple nature.

Ready to dive deeper? Here’s what I offer and how I can help you

➕ Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

📎 Full Portfolio Content - 🗓️ Biweekly Updates (Last Update: 15-May-2025)

1. 💼 Full Portfolio Access – Holdings, Valuation metrics & Performance tracking

2. 📈 Live Trade Updates – My latest trades & watchlist additions

3. 📊 Valuation Tools – DCF, Reverse DCF, Capital Efficiency & Growth models

4. 🎧 Podcast Picks of the Week – Handpicked episodes on businesses & investing

5. 🔎 Stock Picking Methodology – Learn my long-term investment strategy

6. 🎯 Swing Trading Strategy – My short-term trading approach

7. 💬 Private Community Chat - 📎 Learn more

8. 🔐 Full access to all my articles

Subscribe now

📎 Full Portfolio Content - 🗓️ Biweekly Updates (Last Update: 11-Sep-2025)

Subscribe now

Mercado Libre (MELI) Q1 2025 Earnings and Conference Call Highlights

Strong growth in Q1 2025, reporting expansion across both its e-commerce, advertising and fintech businesses. MELI is proving, once again, it can scale profitably while maintaining momentum across the broad in an increasingly competitive Latin American market.

📊 Financial & Operating Highlights

• Revenue: $5.94B (+37% YoY, +64% FXN) 🟢

• Gross Profit: $2.8B (margin 46.7%), -0 bps YoY 🟡

• EBIT: $763M (+45%), margin 12.9%, +67 bps 🟢

• Adjusted EBITDA: $935M, margin 16%, +1 bps 🟡

• Free Cash Flow: $749M, margin 13% (down 45% YoY due to higher CapEx, i.e. fulfillment centers and fintech investments) 👀 ST 🔴 —> LT 🟢

• Operating Cash Flow: $1.02B, margin 17%

• Shareholders dilution: 0% Y/Y 👌

📌 CFO De Los Santos:

“We are excited to kick off 2025 with another great quarter at MercadoLibre as we continue to deliver strong growth across both e-commerce and Fintech. In Q1, we maintained the rapid pace of net revenue growth achieved in 2024, while making solid progress on all of our strategic initiatives. This momentum is fueled by continued investments and improvements in the valuable position for us.”

🛒 E-Commerce Segment: Strong Momentum, Especially 1P

All metrics beating market’s expectations:

• Commerce Net Revenue: $3.3B (+32%)

• GMV: $13.3B (+17%)

• Active Buyers: 67M (+26%)

• Items Sold: 492M (+28%)

• Advertising Revenue: +50% FXN

1P (first party) sales grew +102% YoY, outperforming 3P, particularly in the supermarket segment.

📌 On 1P strategy and margins:

“1P has better unit economics, particularly in supermarkets. We’re operating with greater efficiency and scale through warehouse palletization and master case handling.”

📌 On FMCG opportunity:

“E-commerce penetration in Fast-Moving Consumer Goods (FMCG) remains low, making it a major long-term growth opportunity. We’re seeing a halo effect where buyers end up purchasing more across other categories.”

🌎 Regional Performance

• Brazil: Revenue +20% YoY; GMV +10% 🟡

• Mexico: Revenue +26% YoY; GMV +2% 🟠

• Argentina: Revenue +125% YoY; GMV +77% 🟢

• Argentina DC Margin: 47% (+1047 bps YoY), standout profitability contributor

📌 On Argentina’s recovery:

“Argentina had exceptional performance this quarter. Macro stabilization and our leading position helped fuel both top-line and margin growth. Income from operations grew faster than revenue.”

📌 On Argentina’s credit behavior:

“Delinquency rates are the lowest among our markets. Users rely on us as their primary financial platform, which reinforces repayment discipline.”

💳 Fintech: Payments, Credit & Digital Banking Expansion

Highest Growing Segment 🟢

• Fintech Net Revenue: $2.6B (+43%)

• Total Payment Volume (TPV): $58.3B (+43%)

• Credit Portfolio: $7.8B (+75%)

• Fintech MAUs: 64M (+31%)

• Assets Under Management (AUM): $11.3B (+103%)

• NIMAL (Net Interest Margin After Losses): 22.7%, down from low-30% YoY 🟠

  —> moving upmarket + increasing the share of credit cards which have lower NIMAL (i.e. revenue generated from lending (interest and fees) after accounting for losses from bad debts and funding costs.)

Mercado Pago continues to gain share with a growing user base (nod to Nubank 😉), successful credit card strategy, and expanding investment product offerings.

📌 On digital banking positioning:

“Offering attractive deposit rates has been a powerful customer acquisition and retention tool. We’re increasingly seen as the go-to digital bank in Brazil.”

📌 On NIMAL & credit strategy:

“We’re moving upmarket with safer portfolios. NIMAL is trending lower, but risk is also down, which is the right trade-off for sustainable profitability.”

“We are increasing the share of credit cards to the overall portfolio, the share of credit card portfolios. Even though, credit cards have a lower NIMAL than other parts of the portfolio the NIMAL of credit card itself has been improving. We never had such a good quarter as this one. So, the asset quality is super good. What is changing is the share there. And also, as we grow our portfolio and all the cohorts mature the percentage of cohorts that are older and therefore, with a positive NIMAL will increase.”

📌 On Argentina credit card rollout:

“We expect to launch our credit card in Argentina in the second half of 2025, building on our experience in Brazil and Mexico.”

Risk Management & Profitability

• Provisioning remains high to cover growing credit book:

  • 90 days past due: 18%

  • Provision coverage: 101% (>15 days), 147% (>90 days)

• No deterioration in NPLs, even amid aggressive portfolio growth.

• Past-due credit ratios remain within comfort zone, with newer cohorts showing lower defaults.

📌 Management:

“Our recent Brazil cohorts show the lowest first-payment defaults we've ever seen. Asset quality is as strong as it's ever been.”

🚚 Logistics & CapEx

• Fulfillment growth remains on track, but fewer new centers opened in Q1 due to seasonal effects.

📌 Management:

“We continue to expand our logistics footprint as needed to support demand. CapEx levels remain consistent with last year.”

👀 Competition & Market Trends

• On TikTok Shops’ entrance in Mexico and Brazil:

“Early days. They're targeting low-value goods from local sellers. We’re watching closely, but their playbook seems more aligned with other platforms than ours.”

• On U.S. tariff impact:

“So far, we’ve seen no negative effects. We’re importing from the U.S. into Argentina, and we view increased volume as an opportunity.”

• On competition in Mexico:

“Yes, we saw aggressive pricing and financing from competitors. In some cases, we lacked the right selection, but we’re already addressing that with pricing, rebates, and selection expansion.”

Final Take

MercadoLibre continues to show why it's Latin America’s most dominant tech platform. Strong top-line growth, disciplined credit management, and rising fintech adoption, MELI is executing exceptionally well, even amid complex macro and competitive conditions.

Definition of a high-growth disruptor, high-quality company with top management.

Ready to dive deeper? Here’s what I offer and how I can help you

➕ Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

📎 Full Portfolio Content - 🗓️ Biweekly Updates (Last Update: 11-Sep-2025)

1. 💼 Full Portfolio Access – Holdings, Valuation metrics & Performance tracking

2. 📈 Live Trade Updates – My latest trades & watchlist additions

3. 📊 Valuation Tools – DCF, Reverse DCF, Capital Efficiency & Growth models

4. 🎧 Podcast Picks of the Week – Handpicked episodes on businesses & investing

5. 🔎 Stock Picking Methodology – Learn my long-term investment strategy

6. 🎯 Swing Trading Strategy – My short-term trading approach

7. 💬 Private Community Chat - 📎 Learn more

8. 🔐 Full access to all my articles

Subscribe now

📎 Full Portfolio Content - 🗓️ Biweekly Updates (Last Update: 11-Sep-2025)

Subscribe now

Arista Networks Q1 2025 Earnings Call Summary

• Revenue: $2.04 billion (+27.9% YoY), above their own guide $1.95B and above consensus $1.97B

  

• Revenue Breakdown:

  • Services: 17.1% of revenues, +28.7% YoY and Products: 82.9% of revenues, +27.3% YoY

  • North America contribution 80% (very strong), International Contribution 20%

  • Market, Geographic and Product trends breakdown:

    

• Growing Deferred Revenue: Total deferred revenue (current and non-current) rose to $3.1B—up 10.7% QoQ—showing a healthy backlog and visibility into future revenues.

• Gross Margin: 63.7% (GAAP) | Operating Margin: 42.8% (GAAP) vs 42% last year

• Adjusted EPS: $0.65, above consensus $0.59

• Share Buybacks 👀

  • Completed shares repurchase of $787M in Q1 (highest ever)

  • Board authorizes $1.5B stock repurchases in May.

💬 My Commentary:
Classic Arista. The company beat across the board—driven by strong top-line growth, robust profitability, and continued momentum in the U.S. market.

Is this strength a result of customers front-loading orders ahead of potential tariffs? Personally, I don’t think that’s the primary driver.

Company buying back shares like crazy (highest ever) after post-liberation day shenanigans.

Another positive point is their market sector trends; enterprise segment is Arista’s highest revenue growth engine again, leveling up their revenue share vs cloud titans’ (Microsoft and Meta).

Guidance for Q2 2025

• Revenue: $2.1 billion, vs consensus $2.03B

• Non-GAAP Gross Margin: 63% | Operating Margin: 43%

Guidance for FY25

• Top line guide unchanged.

• Gross margin range of 60% to 62% a decrease from the 64% level this Q.

• Cautious tone as H2 2025 “holds significant ambiguity related to the tariff scenarios.”

AI Revenue Outlook

Analyst Question: How meaningful can AI be in 2025 revenue?

CFO Ita Brennan:

“Too early to break out specifics, but we expect AI-related revenue to ramp meaningfully in the second half of 2025.”

💬 My Commentary: Management keeping their usual cautious tone despite strong underlying metrics. While AI-related visibility continues to improve, Arista is wisely tempering expectations due to ongoing tariff uncertainties.

The biggest concern? Margin contraction. Elevated margins in late FY24 and Q1 2025 were largely driven by a highly efficient, tariff-free supply chain. With potential tariffs and some supply chain disruption on the horizon, that advantage may fade in H2 2025. That said, the company remains confident it can sustain healthy margins “through a mix of supply chain optimization, tariff absorption and potential price increases if required.”

As for AI, no precise revenue target given, but AI is clearly becoming more than just a pilot—indicating potential material impact in H2 and beyond.

AI Networking Momentum

AI Infrastructure and Ethernet vs InfinBand:

AI-driven network architecture is gaining traction; Arista reiterates Ethernet has competitive edge and is the cost-effective alternative to NVDIA’s InfiniBand.

Analyst Question: Will Ethernet truly displace InfiniBand in AI clusters?

CEO Jayshree Ullal:

“We’re not claiming Ethernet will replace InfiniBand in every use case… but for scale-out AI, Ethernet’s economics and openness are more compelling.”

“We believe Ethernet is emerging as the fabric of choice for AI networking… Arista is uniquely positioned with our 800G AI platforms.”

Ita Brennan:

“AI orders contributed meaningfully to bookings this quarter… although still early days for revenue recognition.”

💬 My Commentary:
AI is an emerging tailwind. While not yet a major revenue driver, strong bookings growth shows potential ramp through 2025–2026. Arista is quietly framing itself as an early leader in AI-scale Ethernet networking.

Cloud Titans and Enterprise Demand

• Cloud Titans: Showed sequential improvement; spending stabilizing outside AI 👀

• Enterprise Segment: Growth of over 20% YoY, driven by demand from financial services, healthcare, and public sector.

Jayshree Ullal:

“We’re starting to see early signs of recovery among Cloud Titans… while enterprise remains broad-based and resilient.”

💬 My Commentary:
The bifurcation between cloud and enterprise spending appears to be easing. Which is pretty good news for shareholders as it reduces the risks of customer concentration and cloud titans’ dependency. Enterprise segment remains a growth engine, while hyperscale customers (like Meta, Microsoft) may re-accelerate in 2H 2025.

Campus and Routing Progress

• Cognitive Campus: Driven by Wi-Fi 6E and wired refreshes.

• Routing Portfolio: Gains in peering, edge, and AI clusters.

Jayshree Ullal:

“Our campus and routing initiatives continue to deliver, especially in financial and education verticals.”

💬 My Commentary:
Campus builds and routing (historically weak spots) are maturing into complementary revenue contributors. Cross-selling opportunities are growing as Arista broadens its portfolio beyond switching. More revenue streams diversification = more good news if you are a shareholder.

Competitive Landscape

Analyst Question: How do you view competition from Nvidia, Broadcom?

Jayshree Ullal:

“Competition keeps us sharp. We partner and compete with many… what differentiates Arista is our software-first, cloud networking DNA.”

💬 My Commentary:
Arista sees its EOS software and cloud-native architecture as key moats against chip-heavy rivals. The firm’s strength lies in integration, scale, and programmability.

Supply Chain and Lead Times

• Improvement in Lead Times: Material lead times coming down, allowing more predictable delivery.

Ita Brennan:

“We’ve seen significant improvements in component availability… helping us return to normalized lead times.”

Commentary:
Better supply chain dynamics (tariffs permitted) should improve revenue linearity, customer satisfaction and support acceleration in AI deployments and enterprise rollouts.

📌 Takeaways

• Arista is navigating the macro uncertainty with caution but in a position of strength with more diversified demand across enterprise and hyperscale.

• AI represents a key strategic growth area, but management is wisely (IMO) tempering near-term expectations.

• The Ethernet for AI narrative is gaining traction, positioning Arista as a long-term beneficiary.

• Margins and execution remain top tier, enabling strong profitability even as the company invests in growth.

Ready to dive deeper? Here’s what I offer and how I can help you

➕ Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

📎 Full Portfolio Content - 🗓️ Biweekly Updates (Last Update: 11-Sep-2025)

1. 💼 Full Portfolio Access – Holdings, Valuation metrics & Performance tracking

2. 📈 Live Trade Updates – My latest trades & watchlist additions

3. 📊 Valuation Tools – DCF, Reverse DCF, Capital Efficiency & Growth models

4. 🎧 Podcast Picks of the Week – Handpicked episodes on businesses & investing

5. 🔎 Stock Picking Methodology – Learn my long-term investment strategy

6. 🎯 Swing Trading Strategy – My short-term trading approach

7. 💬 Private Community Chat - 📎 Learn more

8. 🔐 Full access to all my articles

Subscribe now

📎 Full Portfolio Content - 🗓️ Biweekly Updates (Last Update: 30-Apr-2025)

Subscribe now

📰 What’s New at Expanse Stocks?

📊 Quarterly Update - Portfolio Composition (by Industry and Geography) + Valuation Metrics → 📎 Behind-The-Scenes [Free access]

📚 Articles

🔎 Deep Dive Briefs
⛅ Cloudflare | 👷‍♂️ Parsons | 𓇲 MPS | 🥼 Medpace | 🔌 Arista | 🛒 Amazon | 🤖 ASML

💸 General Investing
💎 Hidden Gems Series
🎅 Christmas Special - Top 5 Picks for 2025
📚 Resources for Investors
📢 Latest Stock News
💬 Join My Chat → [📎 Learn more]

🔐 Paid Subscriber Exclusives

💼 Full Portfolio Content – Positions, valuation, trades, performance [Details below]
💎 Hidden Gems - Special Editions
🎁 Bonus Picks – From the 2025 Christmas Special
🤫 Exclusive Sections – From select Deep Dive Briefs & Hidden Gems

👀 Coming Soon

• 🔎 A Hidden Gem Deep Dive Brief → Featuring Topicus & a 💎 Hidden Gems – Special Edition III → Featuring Lumine (both, CSU spin-offs)

• 📘 The Serial Acquirers Playbook: A Primer

• ✈ Deep Dive Brief: Heico

• 📊 Q1 2025 Earnings Reviews + Stock News

➕ Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

📎 Full Portfolio Content - 🗓️ Biweekly Updates (Last Update: 30-Apr-2025)

1. 💼 Full Portfolio Access – Holdings, Valuation metrics & Performance tracking

2. 📈 Live Trade Updates – Latest trades & watchlist additions

3. 📊 Valuation Tools – DCF, Reverse DCF, Capital Efficiency & Growth Models

4. 🎧 Podcast Picks of the Week – Handpicked episodes on businesses & investing

5. 🔎 Stock Picking Methodology – Learn my long-term investment strategy

6. 🎯 Swing Trading Strategy – My short-term trading approach

7. 💬 Private Community Chat - 📎 Learn more

8. 🔐 Full access to all my articles

Subscribe now

Q1 2025 Earnings and Conference Call Highlights

Amazon delivered a solid start to the year, with double-digit revenue growth, strong margin expansion, and record profitability across key segments. Retail demand held up well despite macro concerns and tariff-related uncertainty, while AWS posted its highest-ever operating margins, even as topline growth moderated. Ads continued to outperform peers, and Amazon emphasized resilience, efficiency, and AI innovation across its ecosystem.

Financial Highlights

• Total Revenue: +10% YoY (FX neutral), driven by broad-based growth across 1P, 3P, Ads, and AWS.

• Retail: 1P and 3P sales grew at similar rates, a shift from the usual 3P outperformance.

• Advertising: +19% YoY, outpacing both Google and Meta.

• AWS: +17% YoY growth, slightly slower than previous quarters but with a record 40% operating margin. Rule of 40 for AWS reaching 57%, not bad!

• CapEx: 16% of revenue (vs. 10% in Q1'24), reflecting increased infrastructure spend for AI and fulfillment.

• Free Cash Flow: Weakest point but not too bad, considering one-time charges related to inventory timing and returns (driven by tariffs).

📌 Management on efficiency:

"We’re now seeing unit growth consistently outpace logistics costs — a sign of real operating leverage within our fulfillment network."

AWS: Cloud Resilience, Soaring Margins & AI Demand

While AWS growth slowed to 17%, it continues to generate exceptional profits, with operating margins reaching an all-time high (~40%).

• Backlog: $189B (+20% YoY), with an average remaining contract life of 4.1 years.

• AI Revenue: “Multibillion-dollar run rate” growing at triple digits YoY.

• Capacity bottlenecks remain due to surging demand for GPU clusters and components.

📌 CEO Andy Jassy on AWS strength:

"We’re still in the early innings of AI. But to get full value from it, customers must modernize infrastructure and move data to the cloud — AWS is uniquely positioned to help them do that."

📌 On AI supply constraints:

"As fast as we deploy capacity, it's being consumed. Bottlenecks in motherboards and other components will ease over the year."

Retail: Steady Demand & Margin Expansion Despite Tariffs

Amazon saw resilient consumer demand, especially in everyday essentials, which grew twice as fast as the rest of retail.

• North America and International Operating Margins: 7.2% and 3.7% (excluding one-time charges), up from 5.8% and 2.8% YoY.

• Grocery strength: Amazon surpassed $100B in gross grocery sales last year — one in every three units sold in Q1 was an essential.

📌 Management on tariff impact:

"We haven’t seen demand pull back. If anything, some categories saw forward buying. We believe Amazon is more resilient to tariffs than other retailers."

📌 On everyday essentials and pricing power:

"Most of our sales aren’t high ASP items — essentials dominate our mix. That’s why we're confident in our value proposition, even in uncertain economic environments."

Advertising: Outperforming the Market

• Ads revenue grew 19% YoY, gaining share against Google and Meta.

• Amazon Ads' incremental revenue as a percentage of Google Ads’ growth rose from 33% to 40% YoY.

📌 On monetization momentum:

"Amazon’s ad business is now a critical driver of margin expansion — and we’re just scratching the surface in terms of targeting and visibility."

Strategic Commentary & Outlook

• 2Q 2025 Guidance: Revenue and EBIT slightly below consensus, though leadership emphasized long-term strength.

• Ongoing investments in logistics, AI infrastructure, and international expansion remain top priorities.

📌 Jassy on positioning amid macro uncertainty:

"Customers turn to the providers they trust most in uncertain times — and we’re gaining market share during these periods. We’ve done it before, and I’m optimistic we’ll do it again."

Risks & Considerations

• Tariff volatility and macroeconomic uncertainty could weigh on consumer sentiment.

• AWS growth deceleration compared to Azure and GCP, though margin leadership remains unmatched.

• CapEx ramp-up may pressure near-term FCF, though efficiency gains help balance the equation.

Final Take

Amazon entered 2025 with healthy top-line growth, rising profitability, and sustained leadership across retail, cloud, and ads. While AWS growth slowed slightly, record margins and surging AI demand — capped by AI-constrained supply — underscore the unit's long-term strength.

Retail demand remains resilient, with Amazon leveraging its wide selection of products, pricing, and customer trust to navigate tariff-related uncertainty. And with Ads gaining market share (over Google and Meta) and CapEx fueling long-term infrastructure, Amazon looks well-positioned to scale further in the AI and e-commerce eras.

📌 Jassy’s closing note:

"We’re staying focused on what we can control — delivering value, speed, and trust to our customers — while building the infrastructure for what’s next."

Thanks for reading Expanse Stocks! That’s all for the second Big Tech earnings digest of this season.

Ready to dive deeper? Here’s what I offer and how I can help you

➕ Unlock Premium Content – For just $0.39/day ($12/month) or $0.27/day ($100/year)!

📎 Full Portfolio Content - 🗓️ Biweekly Updates (Last Update: 30-Apr-2025)

1. 💼 Full Portfolio Access – Holdings, Valuation metrics & Performance tracking

2. 📈 Live Trade Updates – My latest trades & watchlist additions

3. 📊 Valuation Tools – DCF, Reverse DCF, Capital Efficiency & Growth models

4. 🎧 Podcast Picks of the Week – Handpicked episodes on businesses & investing

5. 🔎 Stock Picking Methodology – Learn my long-term investment strategy

6. 🎯 Swing Trading Strategy – My short-term trading approach

7. 💬 Private Community Chat - 📎 Learn more

8. 🔐 Full access to all my articles