That is essentially the story of Duolingo. What started in 2011 as an ambitious attempt to make education universally accessible has evolved into a global consumer technology business with more than 58 million daily active users, 12.7 million paid subscribers and nearly $300 million in quarterly revenue.

But the most interesting part isn’t where Duolingo is today. It’s where the company could go next.

Because Duolingo is slowly becoming something much bigger than a language-learning app.

A Simple Idea With a Massive Problem Behind It

When Luis von Ahn and Severin Hacker founded Duolingo, they weren’t simply trying to compete with Rosetta Stone or traditional language schools. They were trying to solve a much larger problem: high-quality education was expensive and inaccessible to billions of people around the world.

Their solution was radically simple. Make the core product free, put it on smartphones and make learning entertaining enough that people actually want to return every day.

Instead of asking users to sit through an hour-long lesson, Duolingo broke learning into small exercises and added game mechanics such as streaks, rewards, leaderboards and characters.

The green owl wasn’t just branding. It became part of the learning experience.

And this created something extremely valuable: habit.

The Free Product Was Actually the Growth Engine

At first glance, giving away the core product seems like a terrible business model. Every free user costs money to serve, and many users may never pay anything.

Duolingo looked at the problem differently.

The free product allowed the company to build an enormous global audience without relying heavily on traditional advertising or expensive customer acquisition.

Once millions of people were using the product every day, Duolingo could convert a percentage of them into paying subscribers.

But there was another advantage that was even more important.

Every lesson completed, every mistake made and every interaction with the product generated information that could help Duolingo improve its learning experience.

That created a powerful flywheel: more users created more data, more data improved the product, a better product increased engagement, and higher engagement created more users and subscribers.

This is one of the reasons Duolingo’s business became increasingly difficult for smaller competitors to replicate.

Then Duolingo Started Thinking Much Bigger

For years, Duolingo was primarily associated with language learning.

Then management started expanding the platform.

The company launched the Duolingo English Test, followed by Math, Music and Chess. This is a very important strategic shift. Duolingo isn’t necessarily betting that everyone wants to learn Spanish or French.

It’s betting that people enjoy learning when the experience is designed correctly.

If the same product formula can work for languages, mathematics, music and chess, then the company’s addressable market becomes dramatically larger.

This is where the Duolingo story starts becoming much more interesting for investors. The company isn’t simply trying to build the world’s biggest language-learning app.

It is trying to build a consumer education platform.

The Moment the Business Model Changed

Having millions of users is impressive, but investors ultimately need to answer one question: Can those users generate meaningful revenue?

Duolingo has increasingly answered that question with subscriptions.

In 2025, the company generated approximately $1.04 billion in revenue, while subscription revenue reached around $873 million.

Then came Q2 2026.

Revenue reached $298.5 million, an 18% increase from the same quarter a year earlier. Subscription revenue alone reached $258 million.

That means subscriptions accounted for roughly 86% of quarterly revenue. This is an important transformation. Duolingo is no longer primarily an advertising-driven education app. It has become a high-margin, recurring-revenue subscription business.

And that changes how investors can think about the company.

But Growth Is Getting More Expensive

Here’s where the story becomes less straightforward.

Duolingo could simply focus on maximizing profits today. Instead, management has chosen to invest aggressively in user growth, product development, marketing and artificial intelligence.

In Q2 2026, revenue grew 18%, but adjusted EBITDA declined slightly and the adjusted EBITDA margin dropped from 31.2% to 25.9%. For a company that investors have rewarded for strong growth and profitability, that decline matters.

Management is effectively making a bet. They are saying that sacrificing some profitability today can create a much larger and more valuable company tomorrow.

If user growth eventually translates into more subscribers and stronger revenue, the strategy could look brilliant in hindsight.

But if monetization doesn’t keep up, investors may start asking whether Duolingo is spending too much to maintain growth.

Subscribe now

The Real Moat Might Be Experimentation

Most people think Duolingo’s competitive advantage is its gamification. That’s only part of the story. One of its most powerful advantages is its ability to constantly experiment with its product.

Duolingo can test different lesson structures, notifications, pricing strategies, onboarding experiences, rewards and features across millions of users. That creates an enormous feedback loop.

A small improvement in retention may seem insignificant when looking at one user. But when applied to tens of millions of people, even a tiny improvement can create substantial economic value.

This is where Duolingo starts looking less like an education company and more like a sophisticated consumer technology platform.

Its product isn’t static. It is constantly learning from its users.

AI Could Be Duolingo’s Biggest Opportunity

Artificial intelligence could completely change the economics of education. Historically, personalized learning required teachers, tutors and expensive human resources.

AI can potentially provide personalized explanations, conversations and practice at a fraction of the cost. Duolingo is already moving in this direction with features such as Explain My Answer, Roleplay and Video Call.

Imagine learning Spanish and having an AI conversation partner available whenever you want. You don’t need to schedule a teacher. You don’t need to travel. You don’t need another person to practice with.

You simply open the app and start speaking. If Duolingo can make that experience genuinely useful, AI could dramatically increase the value of its subscription.

But there is a catch. AI also makes competition easier. An AI startup can potentially build a language tutor much faster than a traditional education company could build a new product.

So the question isn’t whether Duolingo can use AI. The real question is whether Duolingo’s brand, user base, data and distribution give it an advantage when everyone has access to similar AI models.

The Green Owl Solved One of Education’s Biggest Problems

Education has a motivation problem. Most people don’t quit learning because they don’t understand the importance of education. They quit because they lose motivation. Duolingo attacked this problem through gamification.

A 300-day streak creates psychological pressure to keep going. Leaderboards create competition. Rewards create immediate feedback. Suddenly, learning isn’t just about the long-term benefit. It’s about completing today’s lesson.

That habit is extremely valuable. In Q2 2026, Duolingo reported an 84% current-user retention rate, its highest ever. The company also launched a Streak Revival campaign that allowed users to restore previous streaks. Around 15.4 million learners revived their streaks.

That tells you something important about the product. People aren’t simply using Duolingo. They are emotionally invested in continuing to use it.

But Investors Should Watch Bookings Carefully

The headline numbers look excellent. DAUs grew 23%. Paid subscribers grew 17%. Revenue grew 18%. But bookings grew only 8%. That difference deserves attention.

Bookings are an important indicator of future revenue because they capture sales and subscriptions that may be recognized as revenue over time.

If user growth remains strong but bookings growth continues slowing, investors may eventually question whether Duolingo can convert its growing audience into revenue at the same pace.

This is one of the biggest questions surrounding the company today.

Can Duolingo keep growing users without sacrificing monetization?

The Biggest Competitor Might Not Be Another Language App

Duolingo’s traditional competitors include companies such as Babbel, Busuu and Memrise. But the competitive landscape is changing. The bigger threat may actually be artificial intelligence.

Why pay for a language-learning subscription if an AI assistant can speak with you in Spanish for free? Why follow a fixed curriculum if AI can create a personalized learning plan?

Why wait for an explanation when an AI tutor can explain your mistake instantly? These are real questions. Duolingo’s defense is its ecosystem: brand, curriculum, gamification, habit formation, data, distribution and millions of daily users.

An AI competitor can copy a feature. Copying the entire ecosystem is much harder.

Duolingo Wants to Become the Netflix of Learning

The biggest opportunity may not even be languages.

Imagine opening Duolingo and learning Spanish in the morning, practicing mathematics in the afternoon, playing chess in the evening and improving your music skills at night. That sounds ambitious today.

But Duolingo has already taken the first steps toward this vision. If the company successfully proves that its gamified learning model works across multiple subjects, its addressable market could expand dramatically.

That’s why Math, Music and Chess matter. They aren’t just additional features.

They are experiments testing whether the Duolingo learning engine can work beyond languages.

The Numbers Are Strong, But Expectations Are Higher

Duolingo expects approximately $1.207 billion in FY2026 revenue and around $320 million in adjusted EBITDA. Those are impressive numbers.

But the stock market doesn’t reward companies simply for producing good numbers. It rewards companies when the future turns out to be better than investors expected.

That’s the challenge. Duolingo has already proven that it can grow users, build subscriptions, generate high gross margins and create a powerful consumer brand.

Now investors want to know whether the company can keep doing all of that at a much larger scale.

The Biggest Risk Isn’t That Duolingo Stops Growing

The biggest risk isn’t necessarily a collapse in the business. It could be something much more subtle. Growth slows while expectations remain extremely high.

At 58.7 million daily active users, maintaining 20%+ growth becomes increasingly difficult. At the same time, AI competition could increase, marketing expenses could rise and users could become harder to monetize.

The company also depends heavily on Apple and Google for distribution and payments. And while Duolingo has a strong balance sheet, investors still need to consider valuation. A wonderful company can still be a poor investment if you pay too much for future growth.

If you enjoyed this deep dive, subscribe to the newsletter for more simple, data-driven breakdowns of companies, business models and the stocks behind them. I’ll keep digging into the businesses most investors are overlooking—and the stories behind their numbers. See you in the next one.

What if it was a completely new technology that could make your entire category less important?

That is the situation monday.com finds itself in today.

For more than a decade, monday.com built its business around helping companies organize and manage work. But AI is changing what people expect from software. Instead of simply telling employees what needs to be done, AI agents can increasingly do part of the work themselves.

And monday.com isn’t ignoring that shift. It’s betting the company can reinvent itself before the market forces it to.

The interesting part is that this isn’t the first time monday.com has reinvented itself.

It started with a very simple problem

Back in 2012, Roy Mann and Eran Zinman were working at Wix, a rapidly growing company. As the organization expanded, keeping track of projects, responsibilities, and communication became increasingly difficult.

Teams were relying on spreadsheets, emails, and different tools to figure out what was happening. The problem wasn’t a lack of software. The problem was that work was scattered everywhere, and nobody had a clear picture of what was actually going on.

So the founders built an internal tool to solve the problem. That tool eventually became daPulse, and in 2012 they turned it into a company.

The early days were anything but glamorous. The company started from a vacant apartment in Tel Aviv, and Wix became its first customer. From there, the numbers started moving quickly. By 2014, monday.com had more than 3,181 customers, compared with just a handful when it first launched.

But the market was already crowded. Asana, Trello, and Microsoft were all fighting for the same customers. Simply building another project-management tool wasn’t going to be enough.

That forced monday.com to make its first major strategic decision: instead of becoming another task-management application, it wanted to become a broader Work OS.

The first big bet paid off

This decision became especially important in 2020.

COVID suddenly forced companies around the world to move their teams online. Offices closed, employees started working remotely, and companies needed better ways to coordinate work.

For monday.com, this became an unexpected growth opportunity.

The company expanded its platform, launched monday 2.0 and opened its Apps Marketplace. Instead of building a product that forced every company to work in exactly the same way, monday.com gave customers the ability to customize workflows for their own businesses.

That flexibility became one of its biggest advantages.

The company was no longer just selling project management. It was selling a platform that companies could adapt to different teams and different types of work.

And the growth eventually became too large to ignore.

In 2021, monday.com went public on Nasdaq, raising approximately $574 million at a valuation of around $6.8 billion.

But the IPO created another problem.

Once you become a public company, investors don’t care only about your past growth. They want to know where the next $1 billion of revenue is going to come from.

monday.com needed another growth engine.

So it started expanding beyond work management

In 2022, the company launched products across areas including CRM, software development and service management.

The idea was straightforward: if a company already uses monday.com to manage projects, why shouldn’t it use the same platform to manage sales, development and customer service?

This strategy started working.

monday CRM eventually crossed $100 million in ARR, while the overall company continued scaling rapidly.

Revenue went from approximately $308 million in 2021 to $519 million in 2022, then $730 million in 2023, $972 million in 2024, and approximately $1.23 billion in 2025.

That’s a remarkable increase in just four years.

But there was something else hidden inside those numbers.

Growth was slowing.

Revenue growth went from 91% in 2021 to 68% in 2022, 41% in 2023, 33% in 2024, and 27% in 2025.

That’s normal for a company getting bigger, but it creates a challenge.

A $300 million company can grow 50% much more easily than a $1.2 billion company can.

monday.com needed another opportunity.

And then AI arrived.

AI changes the question

The biggest question facing monday.com isn’t whether AI will matter.

That’s already obvious.

The real question is what AI does to the value of traditional work-management software.

Imagine a normal workflow inside a company.

Someone creates a task. An employee researches the problem, writes an email, prepares a report, updates the CRM and eventually marks the task complete.

Software has traditionally helped organize this process.

But AI agents can potentially perform many of those actions themselves.

That creates a problem for companies whose products were designed primarily to help humans manage work.

Why pay for software that helps you manage 100 tasks if an AI agent can complete 50 of them automatically?

In 2025, the company started building its AI strategy more aggressively. Then in 2026, it made the transformation much clearer by repositioning itself from a Work Management Platform to an AI Work Platform.

The difference sounds like marketing.

It isn’t.

The first model says: “Here is software to help you manage your work.”

The second says: “Here is software that can actually do some of the work.”

That is a much bigger ambition.

But every big transformation comes with a cost

And this is where the monday.com story gets much more interesting for investors.

In July 2026, the company announced plans to cut roughly 620–630 jobs, representing around 20% of its workforce.

At almost the same time, monday.com changed how it charges customers.

The traditional SaaS model is relatively simple: companies pay based largely on the number of seats.

But AI doesn’t work that way.

An AI agent doesn’t need a human seat in the traditional sense. It can perform thousands of actions, and the value comes from the amount of work it actually completes.

So monday.com introduced a hybrid seats + AI credits model.

That’s a major economic shift.

Instead of asking only, “How many employees use monday.com?” the company can increasingly ask, “How much work is monday.com doing for this customer?”

If the strategy works, that could eventually create a much larger revenue opportunity.

But there is also a risk.

Customers may not like unpredictable consumption costs. AI infrastructure can be expensive. Net dollar retention could come under pressure. And competitors aren’t sitting still.

Microsoft has Copilot. Notion has AI. Asana has AI. AI-native startups are appearing everywhere.

So the real question isn’t whether monday.com can add AI features.

The question is whether AI becomes a bigger business for monday.com than traditional work management ever was.

Subscribe now

Follow the numbers

This is why I think the numbers are more interesting than the headlines.

monday.com went from $308 million of revenue in 2021 to $1.23 billion in 2025.

That’s roughly a 4x increase in revenue in four years.

The company is also guiding for approximately $1.47 billion of revenue in 2026.

But investors shouldn’t stop at revenue.

The more important question may be what happens to the economics of each customer.

If AI agents become deeply embedded in customer workflows, monday.com could potentially move from a software subscription relationship to something closer to a business infrastructure relationship.

That’s powerful.

A customer might not care much about paying for another software seat.

But if an AI agent is generating leads, responding to customers, preparing reports, updating workflows and completing repetitive tasks every day, the customer may be willing to pay for that output.

That’s the bet behind the credit-based model.

monday.com has done this before

There is another reason I find the company interesting.

monday.com has repeatedly moved into markets that were larger than the one it started in.

It started by solving communication and transparency problems.

Then it became a work-management platform.

Then it expanded into CRM, development and service.

Now it wants to become an AI Work Platform.

The pattern is important.

The company isn’t trying to protect its original product forever.

It’s trying to move one layer higher whenever technology changes.

And that might be exactly what a software company needs to do in the AI era.

Because the companies that simply bolt an AI chatbot onto an existing product may eventually struggle.

The winners could be the companies that redesign their products around AI from the ground up.

monday.com is attempting exactly that.

But the next chapter is far from guaranteed

There are still plenty of things that could go wrong.

Growth is slowing. The company is changing its pricing model. It has reduced its workforce by roughly 20%. AI requires significant investment, and customers may not immediately adopt AI agents at the scale management expects.

There is also the competitive problem.

Microsoft can bundle software with its enormous enterprise ecosystem. Asana and Notion are already building AI capabilities. New AI-native companies are entering the market without the burden of an older business model.

So monday.com has to execute while simultaneously changing the engine of its business.

That’s difficult.

But the company does have something many AI startups don’t have.

250,000+ customers.

Those customers already use monday.com to run real workflows and store operational information on the platform.

That gives monday.com something extremely valuable: distribution.

The company doesn’t have to convince 250,000 businesses to start using AI from scratch.

It can try to introduce AI into workflows they already depend on.

The real bet

And that’s ultimately what makes monday.com interesting.

The investment case isn’t simply that “AI is growing.”

That’s too easy.

The real bet is that monday.com can take an existing customer base of 250,000+ businesses, turn its software from a system that manages work into a system that performs work, and then monetize that additional activity through AI consumption.

If it works, the economics could look very different from traditional SaaS.

If it doesn’t, monday.com could end up spending heavily on AI while simultaneously weakening the predictable subscription model that built the company.

That’s the tension investors should watch.

Because monday.com isn’t just adding AI to its product.

It’s trying to change what its product is.

And perhaps the most interesting part of the entire story is this:

A company that started in a Tel Aviv apartment by helping people see their work more clearly is now betting that the future belongs to software that actually does the work.

The next few years will tell us whether that was monday.com’s smartest decision yet...

or its most expensive experiment.

Share Compounding Wealth

Most companies believe growth comes from adding more.

More products.

More customers.

More acquisitions.

More revenue.

ConocoPhillips chose a completely different path.

It deliberately became smaller.

It gave away an entire half of its business.

Investors questioned the move. Analysts debated whether management had made a costly mistake.

More than a decade later, that single decision transformed ConocoPhillips into one of the world’s largest independent exploration and production companies, worth roughly $140 billion.

This isn’t just a story about oil.

It’s a story about focus, discipline, and why saying “no” can create more value than saying “yes.”

Every Oil Company Wanted to Be Bigger

For decades, the energy industry followed one simple playbook.

If you wanted to become stronger, you expanded.

You explored for more oil.

You built refineries.

You owned pipelines.

You marketed fuel.

You diversified across every part of the value chain.

The larger your empire became, the safer your business supposedly was.

Conoco and Phillips Petroleum followed this exact strategy for decades.

One company traced its roots back to 1875.

The other was founded in 1917.

Both grew into respected American oil companies before eventually merging in 2002 to create ConocoPhillips.

At the time, the merger created one of the world’s largest integrated oil companies.

On paper, everything looked perfect.

But management eventually realized something important.

Being good at many businesses wasn’t the same as being great at one.

The Decision That Changed Everything

Ten years after the merger, ConocoPhillips made one of the boldest strategic decisions in the energy sector.

Instead of expanding further...

It split itself in two.

Its refining, chemicals, and marketing operations became Phillips 66, while ConocoPhillips kept only its exploration and production business.

Many investors couldn’t understand why management would willingly shrink such a large company.

After all, integrated oil companies had traditionally been viewed as more resilient.

But management wasn’t optimizing for size.

They were optimizing for returns.

Their thesis was simple:

If every dollar of capital is invested only in finding and producing profitable barrels of oil, shareholders will earn better long-term returns than if capital is spread across multiple businesses.

It was a risky bet.

But history proved them right.

Sometimes Surviving Is Winning

Then came 2020.

Oil demand collapsed almost overnight.

The pandemic brought travel to a standstill.

Factories shut down.

Planes stopped flying.

For the first time in history, U.S. oil futures briefly traded below zero.

Companies across the industry scrambled to survive.

Many borrowed heavily.

Others slashed dividends.

Some filed for bankruptcy.

ConocoPhillips responded differently.

Instead of protecting production growth, management protected something far more valuable:

The balance sheet.

They aggressively reduced capital spending.

They cut operating costs.

They delayed projects that no longer generated attractive returns.

Most importantly, they refused to chase growth simply because competitors were doing it.

That discipline would define the company’s next chapter.

A Different Definition of Success

Most investors assume oil companies compete by producing the most barrels.

ConocoPhillips doesn’t.

Its goal is to produce the most profitable barrels.

That changes every decision the company makes.

Instead of asking,

“How much oil can we produce?”

Management asks,

“Which barrels generate the highest return on capital?”

That philosophy explains why the company continues investing in assets like:

• Permian Basin

• Eagle Ford

• Alaska

• Norway

• High-quality LNG projects

These aren’t necessarily the biggest assets.

They’re the ones capable of generating strong cash flow even when oil prices fall.

In commodity businesses, low-cost producers usually survive the longest.

ConocoPhillips built its strategy around that reality.

Subscribe now

Growth Isn’t the Goal. Cash Flow Is.

This is where ConocoPhillips separates itself from many competitors.

Many companies celebrate record production.

ConocoPhillips celebrates free cash flow.

During the latest quarter, the company reported:

Financial Highlights

• Revenue: $19.5 billion

• Net earnings: $3.9 billion

• Operating cash flow: $7.4 billion

• Free cash flow margin: 22.6%

• Market capitalization: Approximately $140 billion

Those numbers are impressive.

But the more interesting question is

What did management do with all that cash?

Instead of chasing expensive acquisitions or ambitious expansion plans, the company returned roughly $3 billion to shareholders in a single quarter through dividends and share repurchases.

That’s not an accident.

It’s part of a deliberate capital allocation strategy.

Management has repeatedly emphasized that generating cash means little unless it ultimately reaches shareholders.

The Billion-Dollar Efficiency Plan

Even successful companies eventually face pressure.

During late 2025 and into 2026, weaker crude prices squeezed profits.

Many companies would have hoped oil prices recovered.

ConocoPhillips did something else.

It announced a $1 billion cost-reduction initiative.

That decision reveals an important management philosophy.

Great operators focus on what they can control.

They cannot control global oil prices.

They cannot control geopolitical tensions.

They cannot control OPEC production decisions.

But they can control operating costs.

And over long periods, disciplined cost control often becomes a durable competitive advantage.

What Makes ConocoPhillips Different?

Looking at the business today, several themes stand out.

1. Focus Beats Complexity

Rather than operating across every part of the energy value chain, ConocoPhillips concentrates on upstream exploration and production.

Specialization has improved operational efficiency and capital allocation.

2. Discipline Beats Aggression

The company isn’t trying to produce the most oil.

It’s trying to earn the highest returns from every dollar invested.

That often means walking away from opportunities competitors eagerly pursue.

3. Strong Balance Sheets Create Optionality

The discipline shown during the 2020 downturn allowed ConocoPhillips to continue investing when weaker competitors couldn’t.

Financial strength isn’t exciting during boom years.

It’s invaluable during crises.

4. Shareholders Come First

Rather than building an empire, management consistently returns excess cash through dividends and buybacks.

This creates long-term value even when production growth slows.

Lessons Beyond Oil

The biggest lesson from ConocoPhillips has very little to do with energy.

It has everything to do with business strategy.

Many executives believe success requires doing more.

Launching more products.

Entering more markets.

Buying more companies.

Building larger organizations.

ConocoPhillips demonstrates another path.

Sometimes the highest-return decision is eliminating distractions.

Sometimes becoming smaller makes a company stronger.

Sometimes focus creates more value than expansion.

That’s true whether you’re running an oil company, a software startup, or your own investment portfolio.

Final Thoughts

ConocoPhillips didn’t become a $140 billion company because it discovered the most oil.

It became one because management consistently made disciplined decisions when others chased growth.

They separated from businesses that no longer fit their strategy.

They protected the balance sheet during industry downturns.

They invested only in high-return assets.

And when cash started pouring in, they shared it with shareholders instead of empire-building.

In a world obsessed with growth at all costs, ConocoPhillips offers a refreshing reminder:

The best companies don’t try to do everything.

They become exceptional at the few things that matter most.

At first glance, the numbers look fine.

But look closer, and there’s a much more interesting story:

McDonald’s global machine is still working. The U.S. Machine needs fixing.

And that distinction matters for investors.

• $7.1B revenue, up 4%

• $3.32 diluted EPS, up 6%

• $3.38 adjusted EPS

• $37B global Systemwide sales, up 5%

• 1.3% global comparable sales growth

• $2.36B net income, up 5%

On the surface, this looks like another solid quarter.

But the geographic breakdown tells a different story.

The U.S. Is the problem?

U.S. Comparable sales grew just 0.8%.

That’s down sharply from 2.5% a year ago.

Internationally:

International Operated Markets: +1.5%

International Developmental Licensed Markets: +1.9%

So while McDonald’s is still growing globally, its largest market is clearly losing momentum.

And there’s an important detail hiding underneath that 0.8%.

Check growth was positive. Guest counts were negative.

In other words:

McDonald’s is getting more money from customers, but fewer customers are walking through the door.

That’s something investors should watch.

But here’s the part I find fascinating

Despite weaker traffic, McDonald’s is still producing enormous economics.

The company generated $37 billion in global Systemwide sales in one quarter.

That’s roughly:

$400+ million every day.

And McDonald’s doesn’t need to own every restaurant to benefit from that scale.

Its franchise-heavy model allows the company to participate in the economics of thousands of restaurants while keeping the capital requirements relatively attractive.

That’s the real McDonald’s advantage.

Not burgers.

The system.

The hidden growth engine: loyalty

This may actually be the most important number in the entire release.

McDonald’s says that across 70 loyalty markets:

90-day active loyalty users reached nearly 220 million.

That’s up 13% year over year.

And trailing-twelve-month Systemwide sales to loyalty members reached:

$40 billion

Up more than 20%.

Think about what that means.

McDonald’s isn’t simply trying to get people to buy another Big Mac.

It’s building a global customer-data and digital engagement engine.

More users → more transactions → more data → better personalization → better promotions → higher frequency.

The restaurant is increasingly becoming the physical layer of a digital customer relationship.

And this is where the U.S. reset gets interesting

McDonald’s just appointed Skye Anderson as President of McDonald’s USA.

The message from management is pretty clear:

The U.S. needs more urgency.

McDonald’s wants to improve execution, strengthen value, and regain momentum in its biggest market.

That’s important because the company doesn’t need some revolutionary new business model.

It needs to make its existing machine work better.

Value is becoming a competitive weapon

Consumers are still feeling pressure from inflation and higher everyday expenses.

McDonald’s knows this.

That’s why the company has been leaning harder into value and its McValue platform.

The strategy is straightforward:

Get customers through the door.

Then use:

• menu innovation

• upselling

• digital offers

• loyalty

• personalized promotions

to increase the value of each customer.

The objective isn’t necessarily to win every transaction on price.

It’s to make McDonald’s feel like the obvious affordable choice.

Here’s what the market may be missing

McDonald’s doesn’t need 10% comparable-sales growth to remain an excellent business.

It needs:

modest sales growth + franchise economics + operating leverage + buybacks + dividends.

That’s a very different investment thesis from a high-growth company.

You’re not betting that McDonald’s suddenly becomes a hyper-growth business.

You’re betting that a global consumer brand with enormous scale can keep compounding.

And Q2 gives us both sides of that argument.

Subscribe now

Bull case

Global Systemwide sales +5%.

Revenue +4%.

EPS +6%.

Loyalty sales +20%+.

Active loyalty users +13%.

Every major geographic segment delivered positive comparable sales.

Bear case

U.S. comparable sales only +0.8%.

Guest counts declined.

Global comparable sales slowed to +1.3%.

China remained a weak spot.

The U.S. is still the company’s most important market.

My takeaway

I wouldn’t look at this quarter and say:

“McDonald’s is broken.”

I’d say:

“McDonald’s is mature—and now execution matters more than ever.”

The company already has the restaurants.

It already has the brand.

It already has the franchise network.

It already has hundreds of millions of loyalty users.

The question is whether management can turn all that infrastructure into faster customer traffic and sustained comparable-sales growth.

Because if it can...

the machine doesn’t need to be reinvented.

It just needs to run faster.

And that’s exactly what makes the next few quarters interesting for $MCD investors.

The 3 numbers I’d watch next

1. U.S. comparable sales

Can management get that 0.8% number back above 2–3%?

2. Guest counts

Check growth is nice.

But sustainable growth eventually needs more customers, not simply bigger checks.

3. Loyalty monetization

220 million active users is impressive.

The bigger question is:

How much more revenue can McDonald’s generate from them?

That’s where the next chapter of the McDonald’s story could be hiding.

In 1977 three engineers bet against the entire computing industry and almost lost everything. Oracle Corporation wasn’t supposed to win. Data was messy, slow, and locked in a rigid system; even IBM didn’t believe the future of databases had a real market. But Larry Ellison saw something others ignored. a radical idea from a research paper organizing data into tables that could be queried instantly. Sounds obvious today, back then? It was heresy. Their first big customer was the CIA, then the Air Force. And then it nearly collapsed.

By 1990 Oracle was on the edge of bankruptcy, with fake growth and misreported revenue. The stock crashed 80%. Most companies don’t come back from that; Oracle did. Why? Because they weren’t just selling software; they were building the backbone of modern business computing. But that wasn’t the end of the story. It was the beginning of who Oracle Corporation would become.

In the late 70s, while giants like IBM were still iterating relational databases as experiments, Oracle shipped the first commercial SQL database. Not perfect, not polished, but first, and in technology, first often means everything. Then came a quiet masterstroke. They rewrote the database in C. It didn’t just improve performance; it made Oracle profitable. While competitors were locked to specific hardware, Oracle could run anywhere, on mainframes or servers. the early PC wave. They didn’t just build software; they built leverage.

After surviving the 1990 collapse, Oracle didn’t play safe; it went on the offensive. Oracle becomes the gold standard enterprise, standardized on it. Once your data lives inside Oracle. Switching isn’t just hard. It’s dangerous. That’s when Oracle stopped being a product and became infrastructure. The acquisition of Sun Microsystems gave them Java, MySQL, and control from software to hardware. But the next wave almost left them behind. Cloud and Amazon are moving faster. Oracle looked late and defensive, and then again, they adapted. autonomous databases in industry-specific SaaS, massive cloud infrastructure bets, and now AI. Partnering with NVIDIA and OpenAI, deploying tens of thousands of GPUs, Oracle is quickly becoming the backbone of AI workloads.

Do I understand the business model?

To understand Oracle in 2026, you have to ignore what it was and focus on how it gets paid today. This is no longer a database company. It is a monetization engine built on two layers: a legacy cash annuity and a high-growth AI infrastructure business sitting on top of it.

The legacy layer is misunderstood. Oracle database and support contracts behave like a tax on enterprise IT. Once a company runs critical workloads on Oracle. Switching becomes expensive and risky. That creates recurring high-margin capital; it is stability capital. The second layer is where the market is undercutting the story. Oracle Cloud Infrastructure is not trying to win by being the biggest cloud. It is positioning itself as the most efficient place to run AI workloads. instead of competing broadly. Oracle is concentrating on IHG performance. compute leasing GPU-heavy infrastructure to companies training large models. This is a very different economic model.

The strategic twist is Oracle’s willingness to embrace multicloud. Rather than forcing a lock, Oracle is embedding itself across competitive platforms. This runs rivals into distribution channels and expands total addressable demand without requiring Oracle to wield the entire cloud stack. Hundreds of billions in contracted future revenue provide visibility that most technology companies simply do not have. This is not speculative growth; it is already committed demand waiting to be recognized over time. The put-together model is asymmetric. a durable cash-generating core, a high-growth AI infrastructure layer, and long-duration contracts locking in future revenue.

Is management capable?

On execution the answer is clear. Management has delivered one of the most dramatic pivots in enterprise software, successfully repositioning Oracle from a legacy database vendor into a core player in AI infrastructure. Revenue is growing at a double-digit pace. Cloud infra is compounding at triple-digit NARR, and the backlog has exploded into the hundreds of billions. These are not incremental improvements; they are signs of decisive, high-conviction leadership.

The question is whether they are too aggressive for the balance sheet they are running.

But capability in execution does not eliminate risk; it amplifies it. Oracle leadership is making a deliberate, concentrated bet that AI infrastructure demand will remain structurally undersupplied for years. The response has been aggressive capital allocation, tens of billions in annual capEx, rapid data center expansion, and large-scale GPU commitments. This is not a cautious transition; it is a front-loaded land grab. At the same time, management is reshaping the cost structure; workforce reduction on an unprecedented scale signals a willingness to sacrifice near-term stability for long-term operational leverage. It is a classic oracle move: optimize ruthlessly, then reinvest where it matters most.

The tension lies in the financials. Rising debt levels and negative free cash flow suggest that this strategy has little room for error. if dmenad matrieze as expected . Oracle emerges as a critical player in the AI stack with significant pricing power; if it does not, the fixed cost base and leverage could compress returns quickly. Leadership transition adds another variable. This departure of Safra Cats introduces uncertainty at a time when strategic consistency matters most. Markets tend to discount execution risk during periods of structural changes, and Oracle is undergoing two simultaneously. a business model shift and a leadership shift.

Does the company have a sustainable competitive advantage?

At its core, Oracle Moat is still the database. This is not just software; it is deeply embedded infrastructure inside the world’s largest enterprises. Replacing it is expensive. risk and often unnecessary. that creates high switching costs, longer customer lifecycles, and predictable revenue. In many ways, Oracle doesn’t need to win new customers to survive; it needs to not lose existing ones. It’s so white that advantage cannot outlast the capital intensity required to sustain it.

That alone is a durable foundation. But Oracle is not standing still. It is layering a second moat on top of AI infrastructure. Through its cloud platform, Oracle is positioning itself as a specialized provider for high-performance workloads. instead of competing broadly. It focuses on efficiency at scale. particularly for AI training. This creates a niche advantage where performance and cost matter more than ecosystem size. There is also a strategic shift in distribution. Oracle is no longer forcing customers into a closed system by embedding its database across multiple cloud environments. It is expanding reach while reducing frictions. This turns competitors into channels and increases the surface of its moat. However, this advantage comes with a cost. High leverage, heavy capital expenditure, and relatively constrained free cash flow introduce a different kind of risk. The most is not eroding, but it’s becoming more expensive to maintain, unlike software, whose AI infrastructure models require continuous reinvestment just to stay competitive. This creates tension.

on one side: a deeply entrenched, high switching cost ecosystem
On the other hand, a capital-intensive race where competitors are spending aggressively on innovation.

Is the company active in an attractive end market?

In 2026 Oracle Corporation is no longer operating in a mature, slow-growth category; it is positioned at the center of two of the most capital-intensive and fastest-growing markets in the world: cloud infrastructure and artificial intelligence. This matters more than anything else.

Oracle Cloud Infrastructure business is not just growing; it is compounding at a rate typically reserved for early-stage companies. Demand is being driven by a structural shift. Enterprises are no longer just strong data; they are training models. running inferences. and consuming massive amounts of compute. In that environment, capacity becomes the bottleneck. Oracle is selling that capacity. The AI layer amplifies this dynamic. Large-scale contracts, multi-year commitments, and rapidly expanding backlogs indicate that demand is not cyclical; it is being locked in ahead of time. This creates a retreat setup where growth is not just visible but already contracted. At the same time, Oracle retains exposure to a more stable layer through its application business. While not high growth, it provides continuity and ends up delving deeper into customer workflows. The integration of AI into these applications is less about disruption and more about incremental monetization, improving pricing power without requiring entirely new customer acquisition.

Oracle does not need to dominate the IT cloud market. It only needs to be relevant in the segments that are growing the fastest and require the most capital to serve. That is exactly where it is positioned today. These markets are attractive precisely because they demand continuous investment. Infrastructure must be built ahead of demand, and returns are realized over time. This creates a lag between growth and cash generation. And in markets like these, the winners are not just those who participate but those who can sustain the pace of investment long enough to capture the upside.

What are the main risks for the company?

Oracle is attempting one of the most capital-intensive pivots in enterprise software, building AI-scale infrastructure at speed. This required massive upfront investment funded largely through debt. Total debt has crossed ~$100B+, while fiscal 2026 capital expenditures alone reached ~$55–70B. The result is a balance sheet that leaves little room for missteps. Free cash flow is already under pressure. Heavy capital expenditure has pushed it deep into negative territory (+$20B+), meaning the company is effectively betting today’s balance sheet on tomorrow‘s demand. If AI workload scales as expected, this investment cycle will look prescient. If not, the first cost structure could compress returns quickly.

There is also concentration risk embedded in growth. A meaningful portion of Oracle AI demand is tied to a small number of large customers contributing to a backlog of ~$600B+. While this provides visibility, it also creates dependency. Oracle is committing capital upfront while customers retain flexibility. Competition adds another layer of pressure. Amazon, Microsoft, and others operate at a significantly larger scale with a deeper ecosystem and stronger developer lock-in. Oracle, with ~4% cloud market share, is committed to efficiency, not dominance, which reduces the margin for error. Operationally, the model depends on a supply chain that Oracle does not control. Access to high-end GPUs from NVIDIA is critical. Any disruption, delays, shortages, or pricing shifts directly impact Oracle’s ability to convert its $600B+ backlog into realized revenue.

Finally, there is no longer risk of structural risk. As AI-native architecture evolves, parts of Oracle’s legacy business, still generating ~10B+ per quarter in support and license revenue, could face gradual erosion. The same technological shift driving growth in infrastructure could over time challenge the economy in its highest-margin segments. In other words, Oracle is no longer a defensive compounder; it is a leveraged bet on the durability of AI demand.

Does the company have a healthy balance sheet?

On the surface, Oracle’s scale is undeniable. Total assets have reached ~245B, up ~20%, supported by aggressive infrastructure expansion. But the other side of that growth is leverage. total debt now stands between ~$129B and ~$135B, up sharply from ~$85B Just a year ago, making Oracle one of the most leveraged nonfinancial companies globally. Liquidity exists. but it is not dominant. Cash and short-term investments sit at ~$39B, covering only ~29% of total debt. Shareholder equity is also ~39B, which puts the debt-to-equity ratio in the range of ~3.4x–3.6x, higher than ~97% of software peers.

This is not a conservative balance sheet. Interest coverage remains manageable at ~5.9x, indicating Oracle can service this debt for now. But the cushion is thinner than what investors typically expect from a company undergoing a double-capital-intensive transition. The real pressure shows in cash flow. despite record revenue free cash flow is deeply negative (~-$23.7B), driven by massive capital expenditure of ~55.7B In effect, Oracle is spending nearly $2 for every $1 it generates in operating cash. This is a deliberate choice, not a temporary anomaly.

Management is front-loading investment to capture AI-driven demand, betting that future cash flows will justify today’s balance sheet expansion. There is, however, a counterbalance. Oracle’s performance obligations have surged to ~$638B in contracted future revenue that provides long-term visibility. In addition, financing strategies like customer prepayments (targeting up to ~$25B in FY2027) and flexible infrastructure models help to reduce immediate funding pressure. So the balance sheet is not broken.

Does the company need a lot of capital to operate?

Most people still think Oracle is a capital-light software company. But the reality? It’s turning into a capital monster.

$55.7B capex in FY26 (vs $21.2B last year).
$70B+ expected cash outlay in FY27.
$23.7B negative free cash flow.

Why? AI infrastructure. Oracle is building hyperscale data centers at breakneck speed, funded by $48B raised in 2026 and another $40B planned. Even customers are prepaying $75B just to secure capacity. This isn’t SaaS anymore. It’s heavy industry. Yes, there’s a massive $638B backlog. But the question is, can growth outrun the debt? This is why Oracle isn’t a typical tech story anymore; it’s a high-stakes capital allocation bet in the AI race.

Is the company a great capital allocator?

Oracle used to be a textbook capital allocator. Today it’s a question mark. On the surface, record revenue is a massive $638B backlog. AI demand is exploding underneath; the math is uncomfortable.

ROIC sits around ~10–12%.
Cost of capital? ~13.6%.

The gap matters. It means every dollar deployed isn’t compounding; it’s leaking value for now. Then comes the shift.

Free cash flow: -$23.7B.
Capex: $55B+ and heading toward $90–95B.
Buybacks: basically gone.
Equity issuance: back on the table.

This is a complete identity flip from shareholders’ returns to capital consumption. Yes, dividends are steady, but that’s not the story anymore. Oracle is choosing scale over efficiency. future dominance over present discipline. The bet? Build so much AI infrastructure that returns eventually outrun the cost. I guess that happens; today’s bad allocation becomes visionary. If it doesn’t, it’s expensive overall. That’s the tension. Oracle isnt failing at capital allocation It’s redefining in real time, and markets haven’t decided yet whether that’s brilliance or a misstep.

How profitable is the company?

In FY2026 it delivered record numbers across the board. Net income hit $17B (+36%), operating income reached $20.6B, and EPS climbed to $5.83. Even more striking, operating cash flow surged to $32B (+54%), showing the core business is generating serious money. On a quarterly level momentum stayed strong; Q4 alone brought in $4.2B in profit with a ~44.8% operation margin, which is elite for any tech company.

Margins still remain healthy. Net margins sit around ~21%, and despite some pressure, gross margins are still solid at ~67%; that’s the signature of a high-quality software-driven business. But here’s where it gets interesting. Free cash flow is deeply negative. (-$23.7B). Not because the business is weak but because Oracle is spending aggressively ($55B+ capex) to build AI infrastructure. In simple terms profits are strong, but cash is being reinvested faster than it’s generated. So Oracle is profitable? Absolutely. But its choosing to convert profits into future capacity instead of near-term cash that makes this a different kind of profitability story, one driven by scale, not short-term efficiency. and that distinct matters more than numbers.

Does the company use a lot of stock-based compensation?

In FY2026 SBC comes in heavy with roughly ~$1.2B in just Q4 and roughly ~$11B+ on a trailing basis. That sounds big until you scale it against $67.4B in revenue. Depending on how you measure it, SBC sits around ~7% to 17% of revenue, which is actually reasonable for a cloud company competing for top talent. For years Oracle neutralized this dilution with aggressive stock buyback issues → stock repurchased, net impact minimal. But the playbook just changed. With $50B+ being poured into AI data centers, buybacks have slowed to a crawl. At the same time, Oracle is raising fresh capital, including equity. That means SBC isn’t being absorbed like before; it’s starting to show up in the share count.

Oracle excludes SBC in its non-GAAP numbers, which is why margins like ~44.8% in Q4 look exceptionally high. So yes, SBC is large in absolute terms, but the real story isn’t size; it’s timing. In the capital-heavy AI phase, SBC shifts from managed cost to potential dilution risk, and that subtle shift is what investors are watching closely.

Did the company grow at attractive rates in the past?

Revenue growth past 5 years (CAGR): 10.7% (revenue growth > 5% ✅)

Revenue growth past 10 years (CAGR): 6.2% (revenue growth > 5% ✅)

EPS growth past 5 years (CAGR): 5.1% (EPS growth > 7% ❌)

EPS growth past 10 years (CAGR): 10.9% (EPS growth > 7% ✅)

The 5-year EPS growth misses the 7% bar largely because of heavy reinvestment and rising costs tied to its AI infrastructure push. However, zoom out to 10 years, and EPS growth looks much stronger and more consistent.

Does the company trade at a fair valuation level?

Market cap sits at ~$529B. Revenue hit $67.36B, growing 20.6% YOY. Net income reached $16.98B, up 25.6%... On the surface, this is a company in hypergrowth mode. But then comes the twist: free cash flow is -$23.68B. Why? Because Oracle is spending aggressively. Operating cash flow is strong at $31.98B, but capital expenditures have exploded to $55.66B, driven by AI data center expansion. In simple terms, the business prints cash and then immediately reinvents even more. that breaks a traditional FCF-based valuation model. So instead we normalize. Using the forward P/E of 22.94x, we get a forward earnings yield of ~4.35%. Think of this as steady-state cash generation once the heavy investment cycle stabilizes.

Now layer in growth. Oracle’s current growth (20-25%) is unlikely to be sustained forever, so a realistic long-term assumption is ~12% to 15%; that’s still strong but grounded. then valuation. The multiple has already compressed from ~52x to ~31. 6x trailing and ~22.8x forward. That suggests limited upside from re-rating. If anything, assume a slight drag of -1% or at best flat. Put it together.

Base case:
4.36% yield + 12% growth – 1% multiple = ~15.3% expected return

Optimistic case:
4.36% yield + 15% growth + 0% multiple = ~19.3% expected return

Despite negative cash flow, Oracle still clears the 10-12% hurdle rate comfortably. But here’s the real insight: most people miss this entire thesis, which depends on one assumption: that today’s massive capes actually translate into durable future returns. If it does, Oracle is early in a new growth cycle. If it doesn’t. You’re looking at capital intensity without payoff. It’s a bet on whether $55B+ annual spending turns into long-term dominance, and that’s where the real risk and opportunity lie.

How did the owner’s earnings of the company evolve in the past?

For years this was a cash machine. From 2018 to 2021, owner earnings stayed strong, climbing a peak of ~$14.5B in 2021. even in 2019 and 2020. It consistently generated $11B–$12B+. Simple, predictable, high-quality cash flow. Then something changed. Starting in 2022, the numbers began to weaken; owner ratings dropped from ~$5.3B in 2022, stayed around $5–7B through 2023–2024, and then collapsed to -$4.4B in 2025.

Negative, so what happened? The real reason isn’t declining profits; net income actually grew from $6.7B (2022) to $12.4B (2025). The hidden shift is capital expenditure. Capes exploded from just ~$2.1B in 2021 to over $21B in 2025; that’s a 10x jump. This is key, kind of. Most people miss Oracle’s DoDAT loss and its earning power; it chooses to reinvest aggressively, like into cloud and AI infrastructure. So the owner’s earnings didn’t fall because the business weakened; they fell because the company is spending ahead of growth. This is why the numbers look worse, but the story might be stronger.

Did the company create a lot of shareholder value in the past?

Over the last decade, it delivered ~405% total shareholder return, quietly turning a boring enterprise software name into a serious compounder. Even in the last 5 years, the stock moved from ~$83 to ~$184 (+131%), backed not by hype but by steady earnings growth and cloud momentum. But the real alpha wasn’t just growth; it was capital allocation. Oracle aggressively returned cash when it matters most. Between 2019 and 2021 alone, it repurchased tens of billions in stock ($36.6B in 2019 and $21.6B in 2021), shrinking share count and mechanically boosting EPS. at the same time, dividends steadily scales from $0.60 in 2016 to $1.80 in 2025, adding consistent type of layer to returns

That combination of buybacks + dividends + earnings growth is what provides outsized shareholder outcomes. Now comes the twist. Market cap surged from ~$200B in 2022 to ~$576B in 2025, but returns on capital are compressing (ROIC down from ~13% to ~7.25%). The company is shifting from returning capital to consuming it, fueling AI and cloud infrastructure.

To put it simply, Wall Street remains overwhelmingly bullish on ORCL, with 83.3% of analysts rating it a Buy or equivalent and a consensus of strong Buy across 30 total ratings; not a single bearish call on the board. The avg price target sits at $268 with a median of $250, though the range is wide from a cautious $158 all the way up to a $400 bull case, reflecting meaningful disagreement on how aggressively Oracle will monetize its AI infrastructure buildout.

We would love to hear what you think.

Leave a comment

Compounding wealth work is provided for informational purposes only and should not be construed as legal, business, investment, or tax advice. You should always do your own research

Find businesses with:

High return on capital

Long reinvestment runways

Great management

That’s exactly why serial acquirers fascinate me.

They don’t just run businesses.

They allocate capital.

Today, let’s explore NVEC. Most people ignore a company this small, but sometimes the most interesting businesses aren’t the loudest ones.

Company name: NVE Corporation
ISIN: US6294452064
Ticker: NVEC
Stock Price: ~$102
Market cap: ~$488.31 million
Average daily volume: ~90K shares
Type: NVE Corporation is a leader in the practical commercialization of spintronics, a nanotechnology that relies on electron spin rather than electron charge to acquire, store, and transmit information.

Do I understand the business model?

Imagine owning a business where your product costs just two dollars, but if a customer tries to replace you, it could cost them millions and break their entire system.

That is the reality for NVE.

They don’t make flashy microchips for iPhones or AI servers. Instead, they harness spintronics, manipulating the literal spin of electrons to build microscopic magnetic sensors and isolation components.

But wait...

What is spintronics?

Spintronics manipulates both electric charge and the magnetic spin of electrons to create digital devices that are faster, smaller, and vastly more energy efficient than traditional electronics.
It powers RAM, an ultra-fast memory that instantly saves your data even when the power is completely cut, outperforming standard flash memory.

instead of fighting tech giants in a race to the bottom. NVE plays a completely different game.

The Component Trap: when an industrial manufacturer designs an NVE tiny sensor into a massive factory machine or medical device. That design gets locked, tested, and legally certified. Changing a 2-dollar NVE chip means rewriting engineering blueprints and restarting a multi-year regulatory approval process. Customers don’t leave. They just keep ordering, creating a subscription-like revenue stream without an actual subscription.

The Paid Sandbox: Governments and deep tech firms pay NVE to build next-generation spintronic tech. while this revenue is lumpy and project-based. It lets NVE invent future products on someone else’s dime.

NVE isn’t a commodity chip company riding the volatile tech cycle. It is a quiet design win fortress. They don’t win by scaling to billions of users; they win by becoming entirely impossible to replace.

Subscribe now

Is management capable?

This is where things get interesting. because NVE is not a typical company that needs great operators. It’s already a high-margin, niche business with strong pricing power.

So the real question is:

Can management allocate capital well?

At first glance, the answer looks like yes. The company consistently generates strong profits. Maintain zero debt and deliver a high return on capital. That tells us management is disciplined and knows how to run a business efficiently.

But when we go deeper, a different picture emerges.

Growth has been stagnant. Revenue peaked and then moved sideways. So what is management doing with all the cash this business generates?

They are returning it. almost all of it. Year after year, NVE pays massive dividends, sometimes even more than its free cash flow. This makes them very shareholder-friendly. But it also reveals their mindset. Instead of reinvesting aggressively to grow their business, they choose to distribute earnings.

no bold acquisition. no major expansion. no clear attempt to compound capital at a higher rate. As a result, cash reserves and equity have slowly declined over time.

So, is management capable?

Yes, but in a very specific way. They are conservative, disciplined, and excellent at preserving value. But they are not aggressive capital allocators. They are not trying to build a compounding machine. They are running NVE like a high-quality cash-generating asset.

Management is good.
But whether they are great depends on what you want as an investor.
steady cash today…
or long-term compounding tomorrow.

Does the company have a sustainable competitive advantage?

At first, NVE doesn’t look like a company with a moat. no massive scale. No global dominance and no billion-dollar hype.

but look closer… and you see something far more powerful.

NVE operates in a world most companies don’t even understand spintronics. where data isn’t just carried by electrons but by their spin.

This isn’t just innovation. its specialization.

With decades of research and over 100 patents worldwide, NVE built a quiet fortress around its technology. Its sensors are smaller, more precise, and more energy efficient, making them critical in high-stakes applications like medical devices and industrial systems.

and the numbers prove it.

~79% gross margin.
~ 58% net margin.
~ 26% returns on capital.

That’s not normal for hardware; that’s pricing power. But every moat has cracks. A larger chunk of revenue depends on a single customer. and growth is almost flat.

So yes, NVE has a real competitive advantage, but it’s not built for scale. a niche moat. deep, durable… but not expanding.

Is the company active in an attractive end market?

At first glance, NVE doesn’t seem diversified.

No multiple segments.
No global expansion story.
no total addressable market hype

Instead, it does something very different.

It goes deep… not wide.

NVE operates in highly specialized markets where failure is not an option: medical devices, industrial automation, and high-reliability electronics.

Think about it.

If a sensor inside a factory machine fails, production stops.
If a component inside a medical implant fails, the consequences are far worse.
In these markets, performance matters more than price.

And that’s where NVE thrives.

Its spintronic sensors are not competing in the mass market like consumer chips. They are designed for precision, reliability, and a long product lifecycle.
This creates sticky demand and long-term relationships.

But here’s the trade-off. These markets are not huge.

Growth is steady but limited. and a large portion of revenue depends on a single customer. So yes, the end market is attractive. but not because it’s large. because it’s critical, niche, and extremely hard to disrupt.

What are the main risks for the company?

At first glance, NVE looks almost too good to be true.

High margins.
strong returns
zero debt

But every great business comes with hidden risk.
The biggest one?

Customer concentration: A large portion of NVE revenue depends on a single customer. If that relationship weakens, the impact on the business could be significant.
Growth is another concern. Despite its strong positioning, revenue has remained largely flat over time. This suggests that while NVE has a strong niche, it may not have a large runway to expand. Then comes capital allocation. Management returns almost all of the earnings as dividends, even exceeding free cash flow at times. While this is attractive for income investors, it raises a key question:

Is the company underinvesting in its own future?
Finally, there is the nature of its market. NVE operates in a highly specialized niche. This creates strong pricing power but also limits scale. so while the moat is deep, it may not be wide enough to support long-term high growth

In short:
NVE isn’t risky because it’s weak. It’s risky because it might never become bigger.

Does the company have a healthy balance sheet?

I look at 3 simple signals to judge NVE’s balance sheet:
debt level: $0 (no interest & no financial risk)
liquidity: cash + investment = $20M (strong cushion for a ~$488M company)
Equity trend: $64M to $58M (slowly decline over time).

This looks perfect—no debt and plenty of cash—but here’s the part most people miss.
The balance sheet isn’t weakening because of losses; it’s shrinking by design.
NVE returns almost all its earnings as dividends, sometimes even more than its free cash flow, so instead of compounding internally, it distributes capital externally.
NVE isn’t using its strength to expand; it’s using it to pay you and that tells you everything about the business

Does the company need a lot of capital to operate?

Most semiconductor companies require massive capital. Billions are spent on factories, equipment, and infrastructure just to stay competitive. But NVE operates very differently, and that difference is easy to miss at first glance.

Instead of pouring money into physical assets, NVE invests more in ideas. In 2026 the company will spend around $2.19M on capex and $3.16M on R&D. That tells you something important. This is not a manufacturing-driven business. It’s an intellectual property-driven one where value comes from innovation rather than scale.

The balance sheet reinforces this. With total assets of $60M and liabilities of just $2.1M, NVE operates with almost no leverage. A debt-to-equity ratio of 3.6 is extremely low, meaning the company doesn’t rely on external capital to fund its operations.

But here’s where it gets interesting. Not needing capital is a strength, but it also creates a constraint. Businesses that require heavy reinvestment often have larger growth runways. NVE, on the other hand, generates high returns without needing much capital, which limits how much it can reinvest internally. NVE is a capital-light, highly efficient business. But its future growth depends less on spending more and more on finding new opportunities where technology can be applied.

Is the company a great capital allocator?

Yes. The company consistently generates a high return on capital with an ROIC of 23-24% and ROE of 24-26% over the last few years. There are strong numbers for any business, especially in hardware. It tells us one thing clearly: management knows how to turn capital into profit efficiently.

The real question is not just how well they generate returns... but what they do with those returns.

And this is where NVE becomes interesting. Instead of aggressively reinvesting into growth management, they chose to return most of the cash to shareholders. The company has paid a steady $4.00 per share dividend, distributing around $19.35M in 2026, which is actually more than its net income of $15.2M. This is rare.

It signals to prioritize cash returns today over uncertain growth tomorrow. At the same time, the business remains extremely efficient with minimal debt, low capital requirements, and strong operations cash flows. $16.6M in net value comfortably funds both operations and dividends without external financing. It isn’t trying to maximize growth; it’s trying to maximize certainty.

How profitable is the company?

exceptionally, but the real question isn’t profitability; it’s what the company chooses to do with it. For a hardware company, you expect moderate margins, heavy costs, and constant reinvestment. NVE breaks those expectations completely.

The company operates with an operating margin of ~60% and a net profit margin of ~58%; that means out of every dollar it earns, more than half turns into pure profit. Not occasionally but consistently over the past 3 years, margins have remained above 57%, showing strong pricing power and a highly efficient business model.

But margin alone doesn’t tell the full story.

Look at how efficiently BVE uses its capital. With a return on assets of ~25% and a return on equity of ~26%, the company generates significant profit without needing a large asset base. In simple terms, it doesn’t need to spend much to earn a lot. And yet there is an interesting contrast. Despite such profitability, revenue has remained relatively flat around $25-30M, with net income steady near $15M.

Does NVE use a lot of stock-based compensation?

In most tech companies, stock-based compensation quietly translates into shareholder return. It’s often hidden in plain sight. But NVE tells a very different story. The company uses almost no stock-based compensation. In 2026, SBC was just ~$93K on $26.3M revenue; that’s roughly 0.36%. Even in previous years it stayed below 0.6%; that’s not just low, it’s negligible.

Low SBC means minimal dilution. Existing shareholders are gradually being watered down through stock issuance. What you own today is very close to what you’ll own tomorrow. It also tells you something deeper about the business.

Many companies surely rely on SVC to boost reported cash flows since it’s a non-cash expense. NVE doesn’t play the game. Its cash flows are real, driven by actual operations, not accounting adjustments. And then there’s the management signal. A company that avoids excessive SBC is usually more conservative and disciplined. It aligns with everything else we’ve seen: zero debt, high margins, and a focus on simplicity. So does NVE use a lot of SBC? Not even close, and in a world where solutions are common, that’s a pretty big advantage.

Did NVE grow at attractive rates in the past?

At first glance the answer looks like a strong yes. There were moments where growth exploded. In 2023 alone revenue jumped ~42% and net income surged ~56%; even before that, 2022 delivered solid double-digit growth. It looks like the start of a powerful compounding story. But then the trends break.

What follows isn’t steady growth; it’s a sharp reversal. Revenue dropped ~22% in 2024, followed by another ~13% decline in 2025. In 2026 growth barely recovered at ~1.7%. This isn’t linear growth; it’s cyclical and volatile. Zoom out further, and the picture will be even clearer. Over nearly a decade, revenues have remained almost flat, moving from around $28M to ~$26M. That’s not expansion; that’s stagnation. And yet there‘s something NVE does remarkably well: it stays profitable. Every single year, so did NVE grow at an attractive rate? In short, yes, but over a long term, growth has been inconsistent and largely flat.

Does the future look bright?

NVE is highly profitable. with ~58% net margin and consistent earnings of around $15M on a relatively small revenue base. Few companies operate this efficiently. And the market isn’t ignoring it either. With a PE of 29, investors are already pricing in a certain level of quality and stability. But the real question is not about the present. It’s about the future.

And this is where things become less clear. NVE operated in niche, high-reliability markets like medical devices and industrial systems. They are stable and difficult to disrupt, but they are not fast-growing. There is no massive expansion story. No obvious scaling engine. No clear path to significantly higher revenue.

At the same time, management continues to prioritize dividends over reinvestment, limiting the company‘s ability to compound internally. So what does the future look like? Stable. Profitable. Predictable. But not explosive. NVE doesn’t look like a company about to take off. It looks like one that continues doing exactly what it has always done: generate high returns, stay disciplined, and reward shareholders along the way. And whether that brings profit or not depends entirely on what kind of investor you are.

Is NVEC a good buy at $100?

A high-quality business has strong margins and consistent profitability, but valuation changes everything. Let’s break it down

At current prices of $102, NVE trades roughly 34X free cash flow, giving a cash flow yield of just 2.9%. That’s your starting point, and it’s low. So the obvious question is, can growth make up for it? Here’s where things get tricky. Over the past few years both revenue and free cash flow have actually declined. That makes aggressive growth assumptions risky. So let’s stay conservationists.

If the business grows at 0%, your expected return is roughly around 0.9% per year. Even if we assume a modest recovery of 3% growth, the return only improves to about 2.1% annually. Why so low? Because of one key factor, most investors ignore multiple contractions. NVEC currently trades above its historical range. If the valuation normalizes from 34X to 23X, it creates a 3.8% annual drag on returns. So even though the business is strong, the you pay today limits your future returns. For this to become attractive, the price likely needs to fall closer to $45–$50, where cash flow yield improves and valuation risk disappears until you’re paying for quality but not getting returns.

How did NVE owner Earing evolve over time?

Everything looks stable. Profits are consistent. Cash flows are strong. But the real story is hidden in how much cash is actually left after investment. Theaters are where owner earnings come in. And this is where things get interesting. It looked like a perfect cash machine, but then the trend shifted.

In 2025–2026 owner earnings dropped to around $13M–$14.5M. Not because the business collapsed but because reinvestment increased. Capital expenditure rose to ~$2.2M in 2026, the highest in years. So what does this tell us? I’ve always generated more cash than I needed.

Even in weaker years, owner earnings remain solidly positive. solidly positive. That’s a sign of a resilient, high-quality business. But it also reveals something deeper. The company doesn’t need much capital to operate. And when it does, reinvesting it slightly reduces the cash available to shareholders but not enough to change the overall story. just consistently strong, and that consistency is exactly what defines NVE.

Did NVE create a lot of shareholder value in the past?

The answer looks like a strong yes. high return on capital. consistent profitability, massive cash distribution. It checks all the boxes. But the way NVE creates value is different.

Instead of reinvesting profits to grow the business, NVE returns most of its earnings directly to shareholders. In 2026 alone, it paid out ~$19.3M in dividends, which is actually more than its net income. That’s not common. It means shareholders are getting paid consistently and generously.

At the same time, the business remains highly efficient. With ROE between 22% and 32% and ROIC above 20%, NVE generates strong returns on every dollar it uses. So where’s the catch? Look at the balance sheet. The book value per share has slowly declined from $13.37 to $12.04 over the past few years. That’s because instead of compounding internally the company is distributing its equity base back to investors.

So did NVE create shareholder value? Yes, but not through growth. It created value by paying it out. And that’s the key distinction. This isn’t a compounding machine. It’s a cash return machine. and where that’s attractive depends on what you value more: immediate income… or long-term reinvestment.

A headline flashes across the screen.
" Talks progressing"
"Tensions easing"

Market breath a quiet sigh of relief.

Equities stabilize. Volatility softens. And Somewhere in the background, a familiar
narrative begins to form the worst is behind us.

we’ve seen this movie before.

but oil isn’t buying it.

While financial markets celebrate the possibility of peace, the oil market remains stubbornly cautions. it doesn’t react to optimism. it reacts to reality.

And right now reality looks very different from the story markets are telling themselves.

Oil doesn’t move on Headlines

Markets runs on expectations. oil runs on constraints.

you can reprice a stock in seconds. you can not rewire a global energy system overnight.

every barrel of oil must be extracted, transported, refined and delivered through a network that spans continents. pipelines don’t speed up because diplomacy improves. tankers don’t move faster because headlines turn positive.

even if tensions disappear tomorrow the system doesn’t reset.

production takes time. shipping takes time. refining takes time.

The market may celebrate in minutes.
Oil responds in months.

And that gap between expectations and reality is where things begin to break.

The Fragile Supply Chain

from a distance the global oil system looks resilient.

up close, its anything but.

a large share of worlds oil flows through narrow chokepoints like the strait of

hormuz thin arteries carrying immense pressure. when uncertainty rises, even slightly the system tightens.

insurance cost climb routes shift. decision slow down.

at the same time inventories sit uncomfortably low. there isn’t much margin for error left. what appears as “spare capacity” often exist more in spreadsheets than in physical barrels ready to move .

then comes refining the final bottleneck. even if crude supply improves turning it into usable fuel is another challenge entirely.

each layer depends on the next and when one weakens the entire structure feels it.

this is not a system that absorbs shocks easily.

Why Peace Doesn’t Fix the Problem

there is quite assumptions embedded in today’s optimism:

if the conflict cools, oil prices will follow.

it sounds reasonable. its also imcomplete.

because peace, while powerful politically, does very little to immediately repair physical systems.

Disruptions leave scars. shipping routes don’t normalize instantly. contracts don’t
renegotiate themselves overnight. confidence once shaken returns slowly.

and beneath all of this lies a deeper issue : years of under investment in energy infrastructure.

even in calm conditions increasing supply is difficult. in a stressed system , it becomes even harder.

peace can remove fear.

but it can not create capacity.

Subscribe now

What the Market is Getting Wrong

the market is doing what it always does looking ahead.

but this time it may be looking too far, too fast.

there is a growing belief that normalization will come quickly. that supply will return prices will ease and the system will rebalance itself.

but this belief rests on a fragile assumptions : that the system is flexible enough to respond.

it isn’t.

what markets are pricing is a story.
what the oil market is reflecting is a constraint.

and when those two diverge reality eventually forces a correction.

not suddenly. but inevitably.

The Ripple Effects Nobody is Talking About

oil doesn’t stay in the energy sector.

it moves through the entire system.

higher oil prices raise transportation costs. they push up manufacturing expenses. they slowly seep into the price of everything from food to services.

inflation once thought to be cooling begins to stir again.

central banks already walking a tightrope find themselves under renewed pressure. interest rates stay higher for longer. growth expectations adjust.

markets that priced in relief may have to reprice risk.

for countries like India, the impact is even more direct. higher crude prices mean larger import bills, currency pressure and tighter economic conditions.

what begins as an oil story rarely ends there.

we often forget what markets are build on.

not headlines. not narratives. not hope.

but reality.

you can negotiate peace.
but you can not negotiate physics.

oil moves when pipeline flow, when tanker sail, when refineries operate. it doesn’t respond to sentiment it responds to structure.

and structures take time to rebuild.

the system doesn’t flip from tight to loose because the news cycle improves.

it moves slowly. and when it does move. it carries everything else with it.

Reality vs Expectations

Markets may continue to rally on better headlines. that’s what they do.

but beneath that optimism the oil market is telling a quieter, more uncomfortable story.

A story of tight supply.
of limited flexibility.
of a system stretched closer to its limits than most realize.

peace if it comes will help. it will calm sentiment. it will reduce uncertainty.

but it will not instantly refill inventories.
it will not immediately expand capacity.
and it will not undo years of structural strain.

the world may get better headlines.

but until barrels begin to move freely again. the oil market will remain defined by scarcity.

and when scarcity persists, its consequences rarely stay contained.

Subscribe now