These are reasonable questions. But they are the second question, not the first.

The first question is simpler and harder: is this actually a good business?

It sounds obvious. Of course you should own good businesses. Every investor says they want to own quality companies. The word “quality” shows up in fund names, investor letters, and earnings call transcripts constantly. And yet, when you push on what quality actually means in precise economic terms, not “strong brand” or “great management” or “leader in its category,” but the specific mechanics that make a business genuinely worth owning for a decade, most investors go quiet.

I have been on both sides of that silence. Early in my investing, I thought quality was something you could feel, a recognizable brand, a loyal customer base, a product people clearly loved. I was not entirely wrong. Those things matter. But they are symptoms, not causes. A business can have all of them and still fail to compound value over time if the underlying economics are wrong. I learned that the hard way, more than once.

What follows is what I have arrived at after years of looking at this problem: a working definition of a genuinely good business, in economic terms, not narrative terms. Five specific characteristics. Not a checklist that any business can pass with generous interpretation, but a genuine filter that most businesses fail.

That gap is where most mistakes happen.

I have analyzed a lot of businesses over the years, and the pattern I keep returning to is this: the companies that look good from the outside, growing revenue, recognizable names, high-profile products, are not always the companies that are genuinely good. And the ones that are genuinely good are often not obvious at first glance. They tend to be quiet about it. Their advantages are structural rather than visible. Their numbers tell the story, but only if you know which numbers to read.

What follows is my working definition of a good business. Not a great stock, not a great story, not a great brand. A good business, the kind that can compound value over a decade and survive the things that will inevitably go wrong in the meantime.

There are five characteristics. Not all good businesses have every one of them. But the ones I trust most have most of them.

1. It Earns More Than It Costs to Run

This sounds so basic that it barely needs saying. And yet the number of businesses that fail this test, quietly and over a long period, is larger than most investors realize.

The metric that captures it most precisely is return on invested capital, or ROIC. It measures how much profit a business generates relative to the total capital deployed in it, equity plus debt combined. A business with a 20% ROIC is generating 20 cents of profit for every dollar invested in it. A business with a 7% ROIC is generating 7 cents.

Why does this matter? Because capital is not free. Whether it comes from shareholders or lenders, it has a cost, typically somewhere between 7% and 10% for most businesses when you account for both debt and equity. If a business consistently earns returns below that threshold, it is destroying value even when it appears to be profitable. Revenue is growing, net income is positive, the management team is announcing records, and the business is still worth less than the capital put into it.

This is not a theoretical problem. It is very common.

Think about a consulting firm. It employs talented people, charges high day rates, and has a long list of blue-chip clients. On the surface it looks like a successful operation. But consulting businesses are capital-light only until you account for the real cost: the people. Hiring, training, and retaining skilled consultants requires constant investment. If utilization rates drop or day rates compress, margins fall fast. ROIC in professional services often looks impressive in good years and uninspiring across a cycle. The business is not bad. But it is rarely exceptional.

Now contrast that with a software business that sells a product once and charges a recurring license fee. The second sale costs almost nothing. The marginal customer does not require a new employee. That is a fundamentally different economic structure. If the business has pricing power and retention, ROIC compounds upward as the customer base grows, because the numerator (profit) grows faster than the denominator (capital). That is what a genuinely good return profile looks like.

What I look for specifically: ROIC consistently above cost of capital, not in one year, but across a full cycle. Stability matters as much as the level. A business earning 18% ROIC in three out of five years, with two bad years pulling the average down to 11%, is a more honest picture than a single strong year that looks impressive in isolation. The trend matters too. Falling ROIC, even from a high base, is the earliest reliable signal that a competitive advantage is beginning to erode, usually before it shows up in earnings, and well before it shows up in the stock price.

2. It Can Grow Without Constantly Needing More Capital

This is the characteristic that separates a genuinely great business from a merely good one, and it is the most underappreciated quality in investing.

All businesses need some capital to grow. The question is how much. A grocery retailer needs to build stores to add revenue. A manufacturer needs equipment. A bank needs to hold capital against its loan book. This is not a flaw, it is just the cost of the model. The question is whether the business can generate returns on that capital that justify the investment.

But some businesses have a different and more valuable property: they can grow earnings without deploying proportional new capital. Once the infrastructure is in place, additional revenue flows through at high margins with minimal incremental cost. I think of this as the reinvestment runway, the combination of how much room there is left to grow and how capital-efficient that growth is.

A software business with high switching costs, for instance, might need significant capital in its early years to build the product and acquire the first customers. But once embedded in an organization workflow, each renewal and each expansion costs almost nothing to serve. The existing infrastructure handles it. Revenue grows, capital employed barely moves, and returns on invested capital expand automatically. That is reinvestment runway working as intended.

The grocery retailer example is different but still instructive. Each new store requires real capital, land, construction, fit-out, inventory. There is no escaping it. A grocery chain is not a capital-light business. But if the returns on each new store are consistently high, and if there is genuine whitespace left in the market to expand into, then deploying that capital is the right thing to do. The reinvestment runway justifies the capital requirement. The question is always the same: what does the business earn on the money it puts to work?

Where investors go wrong is in confusing revenue growth with reinvestment quality. A business can grow revenue at 15% per year by consistently investing more capital than the growth justifies, through acquisitions, aggressive expansion, or simply maintaining an expensive operation, and destroy value the entire time. The income statement looks fine. The cash flow tells a different story.

The simplest test: track free cash flow as a percentage of net income over five years. If a business consistently converts earnings to cash at high rates, say, 80% or above, the business model is not devouring capital to sustain itself. If earnings are rising but free cash flow lags persistently, capital is being consumed faster than the headline numbers suggest. That gap deserves scrutiny before it deserves investment.

3. It Can Raise Prices Without Losing Customers

Pricing power is probably the single most important quality a business can have. It is also the quality most frequently claimed and least frequently tested.

Every investor and every management team says their business has pricing power. What they usually mean is that their prices have gone up. That is not the same thing.

Real pricing power means a business can raise its prices, meaningfully, repeatedly, above the rate of inflation, without losing material volume. Customers absorb the increase and stay. The revenue line grows not just because more units are sold, but because each unit is worth more. Margins expand. Intrinsic value compounds.

The mechanism behind genuine pricing power is almost always one of two things: either customers cannot easily switch to an alternative, or the product is important enough to the customer’s outcome that price sensitivity is low.

A specialized industrial software business that is embedded in a client’s manufacturing process falls into the first category. The client is not going to switch because of a 10% price increase, because the switching cost, retraining staff, rebuilding workflows, integration risk, far exceeds 10% of the contract value. The price increase lands and the relationship continues unchanged. That is pricing power through inertia, and it is extremely durable.

A business selling a premium product in a category where alternatives are plentiful has a different and harder path. If a consumer can switch to a competitor at no cost and no inconvenience, pricing power depends entirely on brand conviction, on whether customers genuinely believe the product is worth the premium. That belief can be real and durable, but it requires constant maintenance and is vulnerable to a competitor who simply makes a better product. Pricing power through brand alone is softer than most investors model it to be.

The test I use is blunt: go back five or ten years and look at the average selling price trend. Has it gone up faster than inflation? Have gross margins held or expanded alongside those price increases? If yes, the pricing power is real. If gross margins have compressed even as prices rose, it means volume concessions, mix effects, or input cost inflation are eating the gains, and the pricing power may be weaker than it appears.

There is also a version of this test I find even more revealing: what happens when the business raises prices in a bad environment? Any company can hold pricing when the economy is growing and customers are confident. The real test is a recession year, or a year when a major competitor cuts price aggressively to take share. Does the business absorb that pressure and hold its line? Or does it quietly offer discounts, extend payment terms, and add value-in-kind to retain customers? The second pattern is almost never disclosed clearly in annual reports. You have to read for it, in volume trends, in gross margin movements, in the language management uses to explain what drove revenue.

A business that genuinely has pricing power does not need to justify its pricing. It just takes it.

4. The Business Model Gets Stronger Over Time, Not Weaker

Most competitive advantages are temporary. A new product leads the market for a few years, then competitors catch up. A cost advantage gets eroded by a smarter operator. A regulatory barrier gets dismantled. That is normal, and it is why valuation always requires a view on duration, not just whether a business is good today, but how long it will remain good.

The businesses I trust most are the ones where the competitive position improves with time rather than decaying. Not just holds, improves.

This dynamic shows up most clearly in businesses where use creates value. An accounting software platform that has processed a client’s books for seven years holds something competitors do not: history, context, embedded integrations, and institutional knowledge. The longer the relationship, the more painful it becomes to leave. The moat deepens automatically, without the business doing anything extra. Retention rates are high not because marketing is working, but because the alternative has become genuinely unattractive.

The same dynamic can exist in physical businesses. A grocery chain that has built stores in rural areas, owned the land outright, and served those communities for years, has a position that becomes harder to attack over time, not easier. A competitor entering the same market finds the best locations are taken, the customers are habituated, and the cost of replication grows every year.

This is the question I ask when I evaluate a competitive position: is this moat wider at year ten than it was at year one? Or does success invite competition that gradually narrows it?

Technology businesses are particularly susceptible to the narrowing problem. The company with the most advanced product today competes in a market where everyone is trying to build a more advanced product tomorrow. The moat is only as durable as the next development cycle. That is not a reason to avoid technology businesses, some of them have structural reinforcements that slow down competitive erosion significantly. But it is a reason to be honest about which part of the competitive position is genuinely durable and which part is just a head start and not easy to master.

The honest framing is this: a business whose competitive position strengthens automatically, simply by doing its job well, is worth a premium over one that has to fight to maintain its position. Constant fighting is expensive. It shows up in marketing spend, pricing concessions, and management distraction. Compounding quietly in the background is a very different thing.

5. Management Treats Capital Like It’s Their Own

The final characteristic is the hardest to quantify and the easiest to underweight.

Capital allocation is the job of management, and most investors do not evaluate it carefully enough. They look at earnings growth, return on equity, maybe the dividend history. But the question underneath all of those metrics is simpler: does management deploy capital in ways that create value for shareholders, or in ways that create activity?

The distinction matters more than most people acknowledge.

A business generating strong free cash flow has options. It can reinvest in organic growth. It can acquire other businesses. It can return cash to shareholders through dividends or buybacks. It can pay down debt. Each of these is rational under certain conditions. None of them is automatically the right answer. The quality of the capital allocation decision depends entirely on the price paid and the returns earned.

The pattern I have seen most often in businesses that disappoint over time is not fraud, not management incompetence, and not competitive failure, it is simply the steady deployment of capital into returns below the cost of capital. Acquisitions made at prices that assumed perfect integration. Expansion into new markets that required more capital than the returns justified. Share buybacks executed when the stock was expensive rather than cheap.

None of these are dramatic failures. Each one seemed defensible at the time. Together, they compound into a business that has consumed its own cash flow without creating commensurate value.

What good capital allocation looks like in practice: management has a demonstrated preference for reinvesting at high returns when those returns exist, and for returning capital to shareholders when they do not. They are honest about the difference. They do not pursue acquisitions to hit growth targets. They do not buy back shares indiscriminately. They understand that sitting on cash is sometimes the right answer, even if it looks passive from the outside.

The track record matters more than the stated philosophy. Most management teams have good stated philosophies. Very few have consistent track records of allocating capital at above-average returns across a full cycle, including through periods when the business was under pressure and the temptation to do something was highest.

What This Looks Like Altogether

A genuinely good business, by this framework, earns consistently high returns on capital, converts those earnings to cash at high rates, can raise prices without losing customers, has a competitive position that compounds over time rather than decaying, and is managed by people who treat capital allocation with the same discipline they bring to the operating business.

Finding all five in one company is rare. When it happens, the resulting investment is usually not cheap, because the market tends to recognize quality over time even when it initially misprices it. The job is not to find the perfect business at a throwaway price. It is to develop enough clarity about what you are looking for that you can recognize it when it appears, assess how much of it is genuinely durable, and decide what that durability is worth.

Most of the mistakes I have made as an investor did not come from overpaying for genuinely good businesses. They came from misidentifying which businesses were genuinely good in the first place. A business with a famous brand I assumed had pricing power it did not. A business growing quickly I assumed was reinvesting efficiently when it was consuming capital. A business with high margins in a good year I treated as a structural outcome when it was a cyclical one.

The common thread is that I was looking at the surface rather than the structure. Revenue growth without asking what it cost. High margins without asking if they would hold. A strong market position without asking whether it would be stronger or weaker in five years.

Good businesses are patient. They do not announce their quality in headlines. They reveal it in the economics, steadily, over time, through the numbers that most people do not bother to track carefully.

Those are the ones worth looking for.

Disclosure

This newsletter is published for educational and informational purposes only. Nothing written here constitutes financial advice, investment advice, or a recommendation to buy or sell any security.

I am not a licensed financial advisor, investment advisor, broker, or dealer. All analysis reflects my personal research, opinions, and framework as an individual investor. It may contain errors, omissions, or outdated information, and should not be relied upon as the basis for any investment decision.

Investing involves risk, including the possible loss of principal. Past performance, of any security, strategy, or analytical approach discussed here, is not indicative of future results. Markets are unpredictable, and even well-researched theses can and do go wrong.

I may personally hold positions in securities mentioned in this newsletter, either long or short, at the time of publication or at any point thereafter, without obligation to disclose changes. My interests may not align with yours. Always conduct your own independent research and consider your own financial situation, objectives, and risk tolerance before making any investment decisions. Consult a qualified financial professional if you need personalized advice.

This newsletter is not affiliated with, endorsed by, or associated with any company or security mentioned herein.

It was 280 pages. The table of contents had 22 items. I had no idea what I was looking for, so I was looking at everything equally, which is the same as looking at nothing.

That was the wrong approach, and it took me longer than I’d like to admit to figure out why. A 10-K is not a book you read front to back. It’s a document you guide. Large portions of it are legally required disclosures that tell you almost nothing about the quality of the business. Other sections are written by lawyers whose job is to ensure completeness, not clarity. A meaningful amount of it is designed, not deceptively, but structurally, to bury the interesting material in procedural language.

The goal of reading a 10-K isn’t to get through it. It’s to “separate the wheat from the chaff” as quickly as possible, and then spend your real time on the parts that actually matter.

Here is how I do that now.

What I skip entirely (and why)

Starting with what I skip sounds backwards. But the most valuable thing I learned about 10-Ks is that roughly half the document can be deprioritized without losing much analytical value.

The cover and Part I boilerplate. The opening sections of most 10-Ks are a formal recitation of what the company does, structured for regulatory completeness rather than investor insight. If you already passed the initial filter, you understand how the business makes money in ten minutes, you don’t need to read three pages of legal description of the same thing. Skim for facts you don’t already know, skip the rest.

Risk factors. This is the most reliably over-read section in any 10-K. It is also, in most filings, the least useful per page.

Risk factor sections are written by legal teams whose objective is comprehensive coverage, not prioritization. The result is a list of twenty to forty risks, presented with roughly equal weight, that ranges from “we could be adversely affected by a pandemic” to “our CEO could leave.” Genuine risks are buried next to boilerplate risks that apply to every company in every industry.

I do not read risk factors systematically. I search them. If I already have a thesis, I look for whether my specific concern appears and, crucially, how management characterizes it. If the risk I’m most worried about is mentioned in one sentence among forty other risks and assigned no more language than the generic cybersecurity disclosure, that tells me something. It means management either does not see it as primary, or is managing its prominence carefully. Both are informative.

What I never do is build my risk assessment from the risk factor section. By the time something is in that list, the market has usually priced it.

Properties and legal proceedings. Unless you are analyzing a real estate-intensive business (in which case properties matter) or a company currently facing material litigation (in which case legal proceedings matter), these sections are background noise. I scan them in under two minutes and move on.

Selected financial data tables. These are duplicated elsewhere in the document in more useful form. Skip.

Where the signal actually lives

The 10-K has four sections where I spend the majority of my time, and I go to them in a specific order.

1. The MD&A, but not the way most people read it

Management’s Discussion and Analysis is the most important section in the document. It is also the section most people read the wrong way.

Most investors read the MD&A looking for the numbers: revenue change, margin change, segment performance. Those numbers matter, but they are available faster and more cleanly in the earnings release. What the MD&A contains that you cannot get anywhere else is management’s explanation of those numbers, in their own words, without a moderator.

I read the MD&A looking for three things.

The first is how management explains bad news. Every business has quarters or years where something goes wrong. The MD&A is where management has to account for it in writing. The quality of that accounting tells you a great deal. A management team that explains a revenue miss with precise language, identifying which product line, which geography, whether it was demand-driven or price-driven, and what they are doing about it, is operating with intellectual honesty. A management team that attributes a margin decline to “macroeconomic headwinds” and moves on quickly is telling you they either don’t know the real cause or prefer you not know it. The language is the signal.

The second is the gap between the narrative and the numbers. The MD&A is where management selects which metrics to highlight. Watch what they emphasize and what they don’t. A company whose revenue is growing but whose free cash flow is shrinking will often lead with revenue growth and bury the cash flow discussion later. That inversion, leading with the flattering metric and footnoting the concerning one, is a pattern worth recognizing.

The third is year-over-year language changes. This requires reading the prior year’s filing alongside the current one, which takes extra time but is often worth it. When a company stops discussing a metric they previously highlighted, or changes the way they describe a segment, or introduces new non-GAAP adjustments that weren’t in last year’s filing, those shifts are worth investigating. Businesses that are performing well tend to keep their disclosure language consistent. Businesses under pressure tend to change it.

2. The financial statements and footnotes

Most investors look at the income statement and stop. I look at the income statement last.

The order I use is: cash flow statement first, then the balance sheet, then the income statement, then the footnotes.

The cash flow statement first because it is the hardest to manage. Revenue can be accelerated with aggressive recognition. Earnings can be inflated with favorable depreciation schedules. But a company cannot fake cash leaving or entering the bank account. The operating cash flow line tells me whether the profits reported on the income statement are actually converting into real money. When a company shows growing earnings but stagnating or declining operating cash flow over multiple years, something is absorbing that cash, like inventory buildup, receivables growth, or capitalized expenses that should be running through the income statement. All of those deserve scrutiny.

The balance sheet before the income statement because it shows the cumulative result of all previous business decisions, not just the last twelve months. Debt levels, goodwill (the accumulated cost of past acquisitions, which tells me whether management has historically overpaid for deals), deferred revenue, pension liabilities, and off-balance-sheet commitments are all visible here. A company with a clean income statement and a messy balance sheet is a more fragile business than either statement alone suggests.

The footnotes are where the document becomes genuinely revealing, and most investors never reach them.

Footnotes are where companies disclose the accounting choices that determine how the numbers look. Revenue recognition policies, depreciation assumptions, lease capitalization methods, pension discount rates, goodwill impairment assessments, each of these involves judgment calls that management makes, and those calls can move earnings by double-digit percentages without changing the underlying business at all. A company that is consistently choosing the most favorable available accounting treatment in every footnote is a company I underwrite more conservatively. Not because they are necessarily doing anything wrong, but because their reported numbers represent the optimistic end of what is permitted, and the gap between reported and conservative reality is wider than it appears.

The specific footnotes I always read:

Revenue recognition policies. How does the company decide when revenue is earned? For software, is it recognized upfront or ratably over the contract term? For services, is it tied to delivery milestones? Companies that have changed their revenue recognition policy without a corresponding change in the underlying business model warrant close attention.

Goodwill and intangible asset impairment. This note tells you whether management believes prior acquisitions are holding their value. A company that writes down goodwill is admitting it overpaid for something. A company that has never written down goodwill despite years of declining performance in acquired segments may be carrying an asset value that the business no longer supports.

Related-party transactions. These are worth reading carefully in any company where insiders have significant ownership stakes. Transactions between the company and entities owned by management or board members are disclosed here. Most are routine. Occasionally, they are not.

Stock-based compensation. The total grant value and vesting schedule. For technology companies in particular, SBC can be 10–20% of revenue, and excluding it from “adjusted” earnings is one of the most common ways companies make their profitability look better than it is. The test I apply is simple: if I add back stock-based compensation to the expense line where it belongs, does the business still look as profitable as management’s adjusted figures suggest? Often it does not.

Contingent liabilities. Buried near the end of the footnotes, this section discloses legal proceedings, regulatory investigations, and other obligations that are probable or reasonably possible but not yet certain enough to book as a liability on the balance sheet. Most are routine. Occasionally a company will disclose a regulatory inquiry in this section that later becomes a material event. Reading it takes five minutes. Skipping it means occasionally being surprised by something that was visible all along.

The rest of this piece covers:

→ How to read across 3–5 years of filings to catch what a single 10-K hides

→ The segment note: the one place consolidated numbers can’t disguise what’s actually happening

→ The proxy — two tables that reveal more about management than any earnings call

→ Three specific passages that, more consistently than anything else, tell you what management isn’t saying

Paid subscribers read on below.

Read more

There is a question I come back to constantly when I look at a new stock: is this cheap, or is it a bargain?

Those words sound interchangeable. They are not. Confusing them is one of the most expensive mistakes a long-term investor can make, and it is far more common than most people admit, not because investors are careless, but because the two things look almost identical from the outside, at least at first.

A cheap stock has a low price relative to some metric: earnings, book value, sales, or historical price. A bargain is something different. A bargain is a business trading below what it is actually worth, where the gap between price and value exists because the market is temporarily wrong, not because the business is permanently impaired.

The gap between those two definitions is where most value traps live.

Why Low Numbers Are Not Enough

Let’s start with the metrics people reach for first.

A stock trading at 7x earnings looks cheap. A stock at 52-week lows looks beaten down. A 7% dividend yield looks attractive. A price-to-book ratio below 1.0 looks like you’re buying assets at a discount.

These numbers are not useless. They are useful starting points. But they are only starting points, and investors who stop there are doing half the analysis and drawing full conclusions.

Here is the problem: all of those metrics are backward-looking. They describe what the business earned or owned in the past. What matters to an investor is what the business will earn and be worth in the future. A stock trading at 7x earnings is only cheap if those earnings are stable, recurring, and defensible. If they are about to decline significantly, the actual forward multiple might be 15x, 20x, or higher. The company does not look cheap anymore once you see the full picture.

The same logic applies to dividends. A 7% yield looks generous until you realize the payout is funded by shrinking free cash flow and may be cut within eighteen months. When dividends get cut, stocks typically fall sharply. The investor who was attracted by the yield ends up with both a lower income and a capital loss.

This is what a value trap looks like: a stock that is statistically cheap by surface metrics but is cheap for a reason. And that reason usually turns out to be real, persistent, and visible to anyone willing to look carefully at the business rather than just the number.

The Psychology of the Beaten-Down Stock

Before going further, it is worth being honest about why this mistake is so easy to make.

There is something deeply satisfying about buying something that has fallen. It feels disciplined. It feels contrarian. It feels like you are getting a deal that everyone else is missing. When a stock drops from €80 to €30, the brain registers that as a bargain the same way it registers a jacket marked down from €200 to €75 in a sale.

But stocks are not jackets. A jacket at €75 is worth exactly €75 when you are wearing it. A stock is worth the present value of all the cash it will generate for its owners in the future. The price it used to trade at is irrelevant to that calculation. The price it trades at today is only relevant insofar as it reflects, or fails to reflect, that underlying value.

This is what investors call anchoring bias, and it runs deep. We anchor to the old price and assume mean reversion: that what was at €80 will eventually return to €80. Sometimes that is true. Often it is not. The company’s fundamentals may have changed permanently. The market may not have been wrong when the stock fell; it may have taken new information and priced it correctly.

There is also a quieter psychological trap at work: buying what has fallen feels safer than buying what has risen. The downside seems limited because so much has already gone. The loss feels bounded. This is almost always a mistake. A stock that has fallen 80% can fall another 80% from there. There is no floor created by prior losses. A stock can fall 80% and still be expensive, if the business behind it is deteriorating fast enough.

What a Value Trap Actually Looks Like

Value traps tend to share certain characteristics. They are worth cataloguing, because once you see them clearly they become easier to recognize.

Structurally declining revenue. Not a bad quarter. Not a temporary setback from a macro event. Revenue that has been shrinking for three, four, five years, and where the reasons for that shrinkage are not going away. Think of a print newspaper in 2015. Think of a department store in 2017. The market was not wrong about these businesses. It was right, and investors who bought them because they looked cheap on historical earnings lost money watching earnings fall further.

Excessive debt with deteriorating cash flows. Debt is manageable when cash flow is growing. It becomes dangerous when cash flow is shrinking and the business still owes the same fixed obligations. A company with high debt and falling earnings faces a narrowing corridor: every dollar that goes to interest payments is a dollar that cannot fund growth, reduce costs, or return capital to shareholders. The margin for error disappears.

Poor capital allocation. This one is less visible in the numbers but equally important. A management team that makes questionable acquisitions, pays itself excessively, or pursues growth for its own sake regardless of returns is destroying value even when the headline numbers look stable. Over time, poor capital allocation shows up in return on invested capital declining year after year, which is usually the earliest signal that something is wrong before it reaches the income statement.

Structural industry problems. Some industries face challenges that are not cyclical but secular. Technology makes their products obsolete. Consumer preferences shift in ways that are not temporary. New entrants with structurally lower cost models make the old economics unworkable. In these cases, the business is not facing a temporary headwind that will normalize. It is facing a permanent restructuring of its market. A low multiple on the old earnings structure is not a bargain if those earnings are in long-term decline.

What a Real Bargain Looks Like

A real bargain is usually not a business that looks cheap on every metric. It is a business that looks cheap because the market is applying excessive pessimism to a temporary problem in an otherwise durable company.

The combination you are looking for is specific: temporary market pessimism + durable business quality.

Both halves of that equation matter equally. The pessimism creates the price. The quality determines whether the price is actually low relative to intrinsic value.

What durable quality looks like in practice:

Positive and stable free cash flow. Free cash flow, the actual cash left after operating costs and capital expenditure, is the number that matters most. Accounting earnings can be massaged. Cash in a bank account cannot. A business consistently generating real cash can fund its own growth, return capital to shareholders, and withstand difficult periods without external financing.

Competitive advantage that is genuinely hard to replicate. The question is not whether a company has a moat but whether that moat will hold under pressure. Switching costs that compound with time, permanent or near-permanent licenses, geographic positions built on owned infrastructure: these are durable. A famous brand without structural reinforcement, a technology advantage that competitors are actively attacking, scale that exists in a market where scale does not create pricing power: these are weaker than they appear.

A balance sheet that gives management options. Net cash or low debt relative to earnings means a company can wait, invest counter-cyclically, and survive a prolonged period of market pessimism without being forced into damaging decisions. Companies with heavy debt loads are constrained at exactly the moments when flexibility would be most valuable.

Management that allocates capital honestly and well. Great businesses can be destroyed by poor management. Average businesses can be significantly improved by managers who know how to allocate capital thoughtfully and speak to shareholders honestly about risks and uncertainties. The track record matters more than the current presentation.

The Cigar Butt and the Compounder

There is a useful evolution in investing philosophy that is worth understanding here, not as abstract history but as a practical framework.

Benjamin Graham, the intellectual godfather of value investing, developed what is often called the cigar butt approach. The idea was straightforward: find businesses trading so far below their asset value that even one final puff of value creation was enough to generate a return. Buy statistically cheap, collect the discount as it closes, move on.

This approach worked. In Graham’s era, it worked consistently. The market was less efficient, information traveled slowly, and genuine statistical bargains were more common than they are today.

Warren Buffett started as a Graham disciple, and then, over decades of working alongside Charlie Munger, evolved substantially. Munger’s influence pushed Buffett toward a different question: instead of asking “what is the cheapest business I can find?”, start asking “what is the best business I can find at a price that is not unreasonable?”

The shift sounds subtle. The implications are enormous.

A cigar butt bought at a deep discount and sold at fair value generates one return. A high-quality business bought at a fair price and held for decades generates a compounding return that grows with intrinsic value. If a business earns 20% on capital and reinvests those earnings at similar rates, its intrinsic value doubles roughly every four years. An investor who buys it at fair value and holds it for twenty years participates in that compounding. An investor who buys it at a 30% discount and sells when the discount closes captures one transaction and misses the next fifteen years.

Value investing is not buying what fell the most. It is buying future cash flows at a discount.

Intrinsic Value in Plain Language

Intrinsic value is a concept that gets wrapped in complexity it does not deserve.

At its core, a business is worth the sum of all the cash it will generate for its owners between now and the end of its life, discounted back to today’s dollars to account for time and uncertainty. That is the definition. Every valuation model, every spreadsheet, every multiple comparison is just trying to estimate that number in different ways.

The practical implication is this: if a business will generate substantially more cash in ten and twenty years than it does today, and you are buying it at a price that reflects pessimism rather than confidence, the gap between what you pay and what the business is worth tends to close over time. That closing gap is the source of the return.

What this means for identifying bargains is that you are not looking for a low P/E ratio. You are looking for a business where the future is better than the current price implies. The P/E ratio might be low because earnings are depressed temporarily, in which case the real multiple on normalized earnings is much lower. Or the P/E might be moderate while future growth is significantly underestimated. Both can be genuine bargains. Neither is identified by the ratio alone.

A Tale of Two Stocks

Consider two companies, both trading at 9x earnings. Both have seen their stocks fall 40% in the past year.

Company A operates in a sector facing structural disruption. Revenue has declined for four consecutive years. The dividend consumes more cash than the business generates, funded partly by borrowing. Management has changed three times in five years. Return on invested capital has fallen from 14% to 6%. The balance sheet carries net debt of 4x EBITDA. The stock looks cheap.

Company B is a regional leader in a boring, stable industry. Revenue has grown modestly for a decade. The recent 40% drop came from a sector-wide selloff and one difficult quarter driven by input cost inflation, which management expects to normalize. Free cash flow margin is 12% and has been stable for years. The balance sheet is net cash. Management has owned the same strategy for a decade and speaks frankly about what went wrong in the recent quarter. Return on invested capital has held consistently above 15% through multiple economic cycles.

Same multiple. Same drop. Completely different situations.

Company A is cheap and likely to stay that way. The market is probably pricing future deterioration correctly. Company B is a bargain, if the diagnosis about the temporary nature of the headwind is right. That diagnosis requires work. It requires reading the business carefully, understanding the industry dynamics, and forming a view about whether the problem is cyclical or structural.

That work is what separates investing from shopping.

A Practical Checklist Before Calling Something a Bargain

Before labelling any stock a bargain, work through these questions honestly:

1. Why is it cheap? Can you articulate a specific, credible reason the market has mispriced this business? “It fell a lot” is not an answer. “The market is applying a distressed multiple to what is a temporary operational issue in a structurally sound business” is an answer.

2. Is the problem temporary or structural? Is revenue declining due to a macro cycle, a one-time event, or management misstep that can be corrected? Or is it declining because the underlying demand for what this company sells is permanently shrinking? Be honest with yourself about which it is.

3. What does free cash flow look like, actually? Not earnings. Not EBITDA. Free cash flow. Is it positive? Has it been stable over multiple years? Does it cover the dividend comfortably? Is it being reinvested at attractive rates of return?

4. Is the competitive advantage real? Can you describe specifically what protects this business from competition? If the best answer you have is “it’s a well-known brand,” push harder. What reinforces that brand? What would it cost a competitor to replicate the market position?

5. Can management be trusted? Look at how they have allocated capital over five to ten years, not one. Do acquisitions create or destroy value? Do buybacks happen when the stock is cheap or when it is expensive? Do they speak honestly about failures, or does every annual letter describe only successes?

6. What does the balance sheet allow? A business with net cash can absorb a prolonged period of market pessimism without being forced into dilutive financing or asset sales. A business with heavy debt loses optionality at the worst possible time.

7. If I’m wrong about the recovery, what is the downside? A real bargain has limited downside because the business quality provides a floor. A cheap stock has no such floor. Understanding your downside scenario before committing capital is not pessimism. It is how durable returns are built.

Final Thought

The hardest part of this discipline is not the analysis. The analysis can be learned. The hardest part is resisting the emotional pull of a stock that has fallen sharply, the instinct that says the discount makes it safe, that so much bad news must already be priced in, that surely things cannot get worse.

Sometimes things do get worse. Sometimes a stock falls 80% and then falls another 80%. Sometimes the business that looked like a temporary mispricing turns out to have structural problems the initial analysis missed.

This is why the question “why is it cheap?” deserves more time than any other question in the research process. The market is not always right. But it is not randomly wrong either. There is usually a reason for the discount. Your job is to determine whether that reason is temporary or permanent, and to be honest about the answer, especially when you want it to be temporary.

A genuine bargain is not a stock that has fallen. It is a good business, temporarily misunderstood, available at a price that leaves room for error.

Those situations exist. They are worth the patience required to find them and the discipline required to hold them.

If this kind of thinking is useful to you, paid subscribers get deeper access to the methodology behind it: how I think about competitive advantages, what I actually look for before committing capital, and the reasoning frameworks I apply when markets get noisy. Consider subscribing if you want to go further.

It is not the loss from buying a bad business. Those hurt, but they teach a clean lesson: the company was broken, the thesis was wrong, move on. The loss that lingers is a different kind, the one that comes from buying a genuinely good business and still losing money. Or worse: making a modest return on something that should have compounded for years, simply because you got the price wrong at the start.

Valuation errors are the silent destroyers of long-term returns. They are harder to spot than business errors because they don’t announce themselves immediately. A bad business deteriorates visibly. A valuation mistake sits quietly in your portfolio, delivering returns that feel acceptable until you realize what was actually possible if you had paid more attention to the number on the entry ticket.

This is not about esoteric financial modeling. The five errors below are common, intuitive-sounding mistakes that intelligent investors make repeatedly, not because they lack information, but because the errors are embedded in how most people have been taught to think about valuation.

Error 1: Mistaking a low P/E for a margin of safety

The price-to-earnings ratio is probably the most widely used valuation shortcut in investing. It is also one of the most frequently misunderstood.

A stock trading at 8x earnings feels cheap. A stock at 30x feels expensive. This intuition is not wrong in every case, but it is wrong often enough to cause serious damage.

The problem is that a P/E ratio is a snapshot, not a verdict. It tells you what the market is paying for one year of earnings. It tells you nothing about whether those earnings are reliable, growing, shrinking, or about to fall off a cliff. A business trading at 8x earnings because its profits are expected to halve in the next two years is not cheap, it is trading at a forward multiple of 16x on earnings that may not be the floor.

This is how many investors end up in what value investors call value traps: businesses that look statistically cheap but are priced correctly for a declining future.

The more useful question is not “is the P/E low?” but “why is the P/E low?” If the answer is temporary pessimism about a fundamentally strong business, the low multiple may represent a genuine opportunity. If the answer is that earnings are structurally impaired, the multiple is not low at all, it is simply honest.

A real margin of safety requires more than a cheap-looking multiple. It requires confidence that the earnings being valued are durable, and that the business can maintain or grow them over time. Without that confidence, a low P/E is not protection. It is a warning label in disguise.

There is a further subtlety that compounds this error: cyclical businesses. Industries like mining, shipping, chemicals, and construction go through extended boom-and-bust cycles. During the boom, earnings are high and multiples look low, the P/E of a copper miner at the peak of a commodity cycle might be 6x or 7x, which looks obviously cheap to an investor trained to think that 15x is “normal.” But those peak earnings are not representative of the cycle. When prices revert, earnings collapse, and the investor who bought based on peak-cycle multiples is left holding a stock that is now expensive on normalized earnings. The lesson is that in cyclical industries, a low P/E can actually be a sell signal, not a buy signal, and understanding which kind of business you own is prerequisite to interpreting any valuation multiple at all.

Error 2: Anchoring to a stock’s previous price

This error is so common it has a name in behavioral economics: anchoring bias. And yet knowing it exists does not seem to stop investors from falling into it, over and over again.

The mechanism is simple. A stock that traded at €100 six months ago and now trades at €55 feels like it is offering a 45% discount. Investors who owned it at €100 feel the pain of that gap. Investors who didn’t own it at €100 look at €55 and see an opportunity. Both groups are anchoring to a number that the market has already decided is wrong.

The €100 price from six months ago is not the intrinsic value of the business. It was what buyers and sellers agreed on at a particular moment, under particular conditions, with particular expectations about the future. When those expectations change, when earnings disappoint, when a competitor emerges, when a product fails, when a market shifts, the price changes to reflect a different set of facts. The historical price is a data point from a world that no longer exists.

The right question is never “how far has this fallen?” The right question is “what is this business worth today, and what does the current price imply about the future?” Sometimes a stock that has fallen 45% is still overvalued. Sometimes a stock that has risen 45% from its recent lows is still cheap. The starting point should always be the business and its current economics, not what someone paid for it at a different time.

A particularly dangerous version of this error involves stocks that were once high-quality businesses at genuinely premium valuations. When they fall, they seem to combine the best of both worlds: a great business at a cheap price. But if the fall reflects a permanent change in the business, weakening moats, new competition, shifting consumer preferences, the premium valuation was priced on a reality that no longer holds. The cheapness is an illusion.

Error 3: Underestimating the leverage in discount rate assumptions

Of all the errors on this list, this one is the most technically subtle, and the most consequential.

Most serious investors use some form of discounted cash flow analysis to estimate intrinsic value. The model projects future cash flows and discounts them back to the present using a rate that reflects the risk of the investment. Change the discount rate slightly, and the estimated value changes dramatically.

Here is why this matters in practice: a one-percentage-point difference in the discount rate, say, 8% versus 9%, can shift the intrinsic value estimate of a long-duration business by 15 to 25%. Not because the business has changed at all. Not because the revenue forecast has moved. Simply because of one input, changed by one point, on a spreadsheet.

The reason the effect is so large is that most of the value in a DCF model lives in the terminal value, the estimate of what the business is worth beyond the explicit forecast period, which typically extends ten years or more. Terminal value often represents 60 to 80% of total estimated intrinsic value. And terminal value is exquisitely sensitive to both the discount rate and the assumed long-term growth rate, because those two numbers sit in a denominator together. Small changes in the denominator produce large changes in the output.

What this means in practice is that investors who build precise DCF models are often producing results that feel rigorous but rest on assumptions nobody can verify. The discount rate is frequently chosen rather than derived, most practitioners pick a round number between 8% and 12% and call it analysis. That choice, which takes five seconds to make, determines a large portion of the intrinsic value estimate.

The practical correction is to run valuation as a range, not a point. What does the business look like at a 9% discount rate? At 10%? At 11%? If the investment case falls apart above 9%, that is important information, it means the thesis depends heavily on the market agreeing to value this business at a low required return. That is a form of valuation risk that deserves explicit attention.

The first three errors are about how investors misread numbers. Errors 4 and 5 are about how they misread time, and in practice, those tend to be the more expensive mistakes.

Error 4 explains why the most sophisticated investors consistently overpay for businesses they genuinely admire, and the specific condition that makes a premium valuation dangerous rather than justified. Error 5 is the mistake that costs long-term investors the most in compounding terms, because it feels like discipline when it is actually the opposite.

Both errors include a concrete framework for catching yourself before you make them.

Subscribe to read the full analysis.

Read more

Every value investor talks about moats. The word shows up in earnings calls, analyst reports, and newsletter introductions so often that it has started to lose meaning. A brand is a moat. A patent is a moat. Being the biggest player in your market is a moat. At some point, almost anything can be called a moat if you squint hard enough.

But moats are not equal. Some protect a business for decades and compound quietly in the background while everyone argues about short-term earnings. Others look impressive until a competitor with better technology, a lower cost base, or simply more patience decides to test them. The difference between the two is not always visible from the outside, and that is precisely why so many intelligent investors overpay for the wrong kind of protection.

Over the past year, I have analyzed seven businesses in detail: Novo Nordisk, Nike, Palantir, Alten, Dino Polska, Pfizer, and Constellation Brands. Every one of them had something you could call a moat. Not all of them were equally protected. Looking across these businesses together, a clear pattern emerges about which types of competitive advantages hold and which ones quietly erode while you’re still holding the stock.

Here is what I’ve found.

The 4 Moats That Actually Last

1. Switching Costs - The Moat That Compounds Invisibly

Switching costs are underappreciated because they don’t look like moats at first glance. There’s no famous brand, no patent filing, no regulatory barrier you can point to. What there is, instead, is friction, the accumulated cost, disruption, and risk that a customer faces when they try to replace you.

When switching costs are high enough, customers don’t leave. Not because they love you, but because the cost of leaving exceeds the benefit of whatever the competitor is offering. That is a durable economic position.

Palantir is the clearest example (one company I analyzed). Once a government agency or large enterprise embeds Palantir’s Gotham or Foundry platform into operational workflows, intelligence analysis, supply chain modeling, operational decision-making, extracting it is not a software migration. It is a years-long, operationally disruptive, politically sensitive project. The CIA does not swap data infrastructure like it swaps office furniture. The switching cost is not just financial. It is institutional.

This is why Palantir’s revenue profile looks the way it does: multi-year government contracts, expanding commercial relationships, and a gross margin of 82% that reflects the pricing power you get when your customers are effectively locked in. The business is expensive at current valuations, and I have written about that at length. But the moat itself, the switching cost moat, is one the durable structure.

What makes switching cost moats particularly powerful is that they compound with time. Every year a customer stays embedded in your platform, they build more workflows around it, train more staff on it, and accumulate more data inside it. The cost of leaving increases every year without you doing anything additional. That is a moat that gets stronger as the business ages, which is the opposite of most competitive advantages.

The test for a switching cost moat is simple: if a competitor offered your customer a 20% price discount tomorrow, what would it cost the customer to actually switch? If the answer is “more than 20% of the contract value,” you have a real switching cost moat.

2. Cost Advantages - The Moat That Survives Price Wars

A cost advantage means you can profitably serve customers at a price your competitors cannot match without losing money. It is the most straightforward moat in theory and one of the hardest to build in practice, because it usually requires either structural decisions made years earlier or scale that takes decades to accumulate.

Dino Polska is the most instructive case I have analyzed. Dino operates grocery stores in small Polish towns and villages where Biedronka and Lidl are simply not efficient to deploy. But the cost advantage goes deeper than geography. Dino owns most of its stores outright, the land, the building, the infrastructure, in an industry where leasing is standard. That ownership creates a structural cost advantage of roughly 2 to 3% relative to competitors who pay rent every year. It also owns Agro-Rydzyna, a meat processing plant that supplies fresh meat directly to every store daily. Zero wholesale markup. Zero dependency on external supply chains. Every Dino store has a butcher counter; most competitors do not.

The result is visible when things get difficult. Poland’s grocery market entered a sustained price war in recent years, with Biedronka, Lidl, and Dino all cutting prices. Biedronka, operating with a leasing model and thinner margins, faces considerably more financial pressure in a sustained price war than Dino does. The company with the lower cost base can afford to wait. The one with the higher cost base eventually blinks.

This is what a cost advantage moat looks like in a stress test: it doesn’t necessarily show up in sunny conditions, when everyone is growing and margins are expanding. It shows up when conditions deteriorate and you discover that you can absorb the pressure while your competitor cannot.

Cost advantages are particularly durable when they come from physical or structural sources, owned real estate, captive supply chains, manufacturing scale, rather than from operational efficiency alone. An efficient competitor can close an operational efficiency gap. They cannot easily replicate a decade of real estate acquisition or a vertically integrated supply chain that was built at below-current-market costs.

3. Intangible Assets With Permanence - The Moat That Doesn’t Expire

Not all intangible assets are created equal as moats. A brand can be an intangible asset. So can a patent. So can a perpetual license. These are very different things, and investors frequently conflate them.

The distinction that matters most is permanence. A brand requires constant maintenance, marketing spend, product quality, cultural relevance. A patent expires on a known date. A perpetual license, properly structured, is simply ownership in a different form.

Constellation Brands illustrates this with unusual clarity. In 2013, Constellation acquired the exclusive, perpetual US rights to import and sell the Modelo portfolio including Modelo Especial, Corona, Pacifico, and related brands, for $4.75 billion. That transaction was criticized at the time as expensive. Today, Modelo Especial is the number-one selling beer brand in the United States by dollar sales, having overtaken Bud Light in 2023.

What makes this moat genuinely durable is the word “perpetual.” This is not a distribution agreement with a renewal date. It is not a licensing deal that Anheuser-Busch InBev can renegotiate. It is permanent ownership of the US rights to those brands, backed by manufacturing infrastructure that Constellation has invested more than $5 billion building in Mexico. A competitor replicating this position would need to acquire brands with comparable recognition, build equivalent production capacity, and establish comparable distribution relationships in the world’s largest beer market. The capital required is north of $10 billion. The time required is measured in decades.

Tariffs, aluminum costs, and Hispanic consumer pullback are creating genuine near-term pressure on Constellation right now. I have written about those risks. But none of those headwinds touch the foundational asset. The permanent license still exists. The number-one beer brand in America still belongs to Constellation. That is the definition of a durable intangible moat: it survives the noise.

Contrast this with Pfizer, where the intangible assets, drug patents, are explicitly time-limited. Pfizer knows today that approximately $17 billion in revenue will be impacted by patent expirations between 2026 and 2028. Eliquis, Ibrance, Xtandi: the dates are on a calendar somewhere, and the business has to rebuild itself before that calendar runs out. That is a profoundly different investment situation from owning a perpetual license, even if both look like “intangible asset moats” on the surface.

4. Geographic and Structural Moats - The Moat Nobody Sees Coming

The most underappreciated moat type is one that doesn’t fit neatly into any standard category. I call it a geographic or structural moat: a competitive position that exists because of the physical or logistical reality of a market, rather than because of brand, technology, or regulation.

Dino Polska illustrates this too, though from a different angle than the cost advantage discussion above. The core insight behind Dino’s competitive position is that Biedronka and Lidl built their business models around urban and suburban populations of 20,000 or more. Their stores are too large, their cost structures too inflexible, and their product mix too broad to operate efficiently in villages of 1,000 to 5,000 people. Dino identified this gap early and has systematically filled it. As of the end of 2025, it operated 3,033 stores.

Critically, when Dino enters a rural village, it is not competing with Biedronka for the same customers. It is replacing the trip those customers previously made to a town 20 kilometers away. That is expansion into whitespace, not a price war. And because Dino owns the real estate in those villages, a competitor that later decides to enter the same market cannot simply open next door. The land is taken.

This type of moat is hard to see in a spreadsheet, which is partly why it is undervalued. There is no metric called “geographic whitespace” or “first-mover land ownership.” But the economic reality is clear: the cost for Biedronka to replicate Dino’s rural position is now enormous, years of capital expenditure acquiring land in markets Dino has already occupied, and the incremental revenue opportunity shrinks with every store Dino opens.

The 3 Moats That Don’t Last

1. Brand Moats - Durable Only With Conditions

Brand is probably the most frequently cited moat in value investing, and the most frequently overestimated. A brand is not a moat by itself. It is an asset that can function as a moat under specific conditions, and those conditions are less common than most investors assume.

The conditions are these: the brand must command genuine pricing power, meaning customers pay a premium specifically because of the brand. And that premium must be sustainable, meaning customers do not have a credible, easily accessible alternative.

Nike is the cautionary case I have spent the most time with. On paper, Nike has one of the most recognizable brands in the world. But “recognizable” is not the same as “defensible.” Nike’s brand has not prevented Hoka, On Running, and New Balance from taking meaningful market share in core categories. It has not prevented consumers from shifting preferences toward performance and authenticity over lifestyle and logo. The brand is real. The moat is not what it appeared.

The reason brand moats erode is that brand equity requires continuous maintenance and renewal, and the consumer always has a vote. Unlike switching costs, which get stronger as the relationship deepens, brand moats require constant investment just to hold position. Marketing spend, product quality, cultural relevance, influencer relationships, these are not one-time investments. They are ongoing costs of keeping the moat full.

Brand moats that are reinforced by something structural, Constellation’s Modelo, which combines brand with a permanent license and cultural authenticity built over 30 years, are far more durable than brands that stand alone. Brands that exist alongside high switching costs, like enterprise software with a well-recognized name, are also more durable. Brand alone, in a market where consumers have easy access to alternatives, is not a moat. It is a head start.

2. Technology Moats - The Fastest-Eroding Advantage

Technology moats are seductive because they feel definitive. This company has the best AI. This company has the most advanced GLP-1 molecule. This company built a data platform no one else has. These feel like insurmountable advantages because of how much they cost to build and how long they took to develop.

The problem is that technology moats are the most frequently disrupted. The competition does not stand still, and in most technology categories, the pace of innovation means today’s leading technology is tomorrow’s baseline.

Novo Nordisk is the live case study. Semaglutide, the molecule behind Ozempic and Wegovy, was a genuine technological breakthrough. When it launched, it was the most effective approved obesity treatment available. Novo Nordisk built an entire strategic thesis around its leadership in GLP-1 therapy. For several years, that thesis was correct.

Then Eli Lilly’s tirzepatide achieved 23.6% average weight loss in head-to-head trials, versus 20.2% for Novo’s next-generation CagriSema combination. The technological advantage Novo had built did not disappear, semaglutide is still an effective drug. But it narrowed, and in a category where efficacy is the primary purchasing criterion, a narrow technological advantage is a fragile one. The stock fell more than 50% from its peak.

This is the pattern with technology moats: they work until a better technology arrives, and in industries where better technology is the primary source of competitive advantage, pharmaceuticals, software, hardware, better technology always eventually arrives. The question is timing.

Technology moats can be durable, but only when they are accompanied by something harder to replicate: the data flywheel that makes better technology increasingly difficult to catch (Palantir’s accumulated operational data is part of this story), or the regulatory approvals and clinical infrastructure that slow down the pace at which competitors can deploy superior technology (which is part of why Pfizer’s scale still matters despite the patent cliff). Technology alone is not enough.

3. Scale Moats - Powerful, Until They’re Not

Scale is often described as a moat because large companies can spread fixed costs across more revenue, negotiate better terms with suppliers, and outspend competitors on R&D and marketing. All of this is true, and scale advantages are real. But scale is frequently confused with durability.

The problem with scale as a standalone moat is that it requires the market to cooperate. Scale advantages exist relative to competitors in the same market, serving the same customers, with the same business model. When any of those conditions change, the market structure shifts, a new entrant arrives with a different cost model, technology reduces the fixed costs that gave you the scale advantage, the moat that scale created can erode quickly.

Alten is an instructive case. Alten is a large engineering consulting business by industry standards, with tens of thousands of consultants deployed across major industrial sectors. It benefits from scale in terms of its ability to serve complex, multi-geography projects and maintain deep client relationships. But in consulting, scale does not create the kind of pricing power or client lock-in that scale creates in, say, cloud infrastructure or pharmaceutical manufacturing. Clients can replace one consulting firm with another relatively easily. Alten’s low switching costs and the commoditization risk in its sector mean that its scale advantage does not translate into durable pricing power.

Similarly, in pharmaceuticals, Pfizer has enormous scale advantages in manufacturing, distribution, and regulatory expertise. These are real and not easily replicated. But scale did not protect Pfizer’s revenue when its patent-protected drugs ran out of exclusivity. A generic manufacturer producing Eliquis biosimilars does not need Pfizer’s global distribution network. They need a patent to expire, and then the scale advantage collapses in that specific product line.

Scale is most durable when it creates infrastructure that is genuinely difficult to replicate, Constellation’s $5 billion brewing capacity in Mexico, Dino’s network of owned rural real estate, rather than simply being “bigger” in a market where size can be matched over time.

What This Means in Practice

The practical implication of this framework is that not all moats deserve the same confidence when you’re building a valuation or sizing a position.

Switching cost moats and structural/geographic moats deserve high confidence in durability, with valuations that can reflect a wide margin of safety across a long time horizon. These are the moats that hold during price wars, recessions, and competitive attacks.

Intangible asset moats, particularly permanent or near-permanent licenses, deserve similar confidence when the asset genuinely cannot be replicated. But the distinction between “perpetual license” and “expiring patent” is consequential. One is an asset you own. The other is a lease from the government with a known end date.

Brand moats, technology moats, and scale moats deserve scrutiny every time. They require asking: what reinforces this moat? What would have to be true for it to erode? How much of my valuation depends on this moat holding for 10 years versus 3?

The companies that have frustrated me most as an investor are the ones where I confused a temporary advantage for a structural one. Novo Nordisk’s semaglutide leadership looked structural because of how much R&D it took to develop. It turned out to be durable only until something more effective arrived. Nike’s brand looked durable because of how recognizable it was. It turned out to be vulnerable in the specific categories where consumer preferences were shifting most aggressively.

The businesses that have held up best in my analysis are the ones where the moat has a physical or contractual dimension that is genuinely difficult to replicate: Dino’s owned rural real estate, Palantir’s embedded government workflows, Constellation’s permanent license. These don’t just keep competitors out today. They get harder to challenge over time.

That is the definition of a moat worth owning.

Detailed valuation breakdowns and updated stock analyses, including the specific moat assessments that feed into my position sizing decisions, are available in the paid archive.

The bear case writes itself. Tariffs on Mexican imports. Aluminum levies on cans. A core Hispanic consumer base pulling back on spending amid immigration enforcement. A wine and spirits segment that required over $3 billion in goodwill impairments before management finally decided to divest the mainstream brands. Heavy capital expenditure commitments for a third brewery in Veracruz that the business may not need as urgently as it once expected.

But the bear case writing itself is usually a reason to look more carefully, not less. The most interesting investment opportunities rarely come with clean narratives. They come with real problems and a question underneath: are those problems temporary, or have they structurally changed what this business is worth?

That is what I want to work through here.

Understanding the Business

Constellation Brands is a beverage alcohol company headquartered in Rochester, New York. It produces and markets beer, wine, and spirits, but for practical purposes the investment case is almost entirely a beer story.

The beer portfolio consists of Mexican import brands licensed exclusively for the United States market: Modelo Especial, Corona Extra, Corona Light, Pacifico, Victoria, and the Modelo Chelada family. These are not produced in the US. They are brewed entirely in Mexico, predominantly at Constellation’s own facilities in Nava and Obregon, with a third brewery in Veracruz under construction.

As of fiscal year 2026 (ending February 2026), beer accounted for approximately $8.3 billion of the group’s $9.1 billion in total net sales, around 91% of revenue. The wine and spirits segment, having shed its mainstream brands through a series of divestitures (SVEDKA vodka to Sazerac, Woodbridge and Meiomi and others to The Wine Group), now contributes roughly $824 million in sales, down from nearly $1.7 billion two years earlier. What remains is a focused set of premium brands: The Prisoner Wine Company, Kim Crawford, Robert Mondavi Winery, High West Whiskey, and Mi CAMPO Tequila.

The revenue model is simple. Constellation does not brew beer in the United States. It holds an exclusive, perpetual license to import and sell its Mexican beer brands in the US, a right secured through the 2013 acquisition of the US rights to the Modelo portfolio from Anheuser-Busch InBev for $4.75 billion. That acquisition is arguably one of the best strategic purchases in US consumer staples history. At the time it looked expensive. Within a decade, Modelo Especial had become the best-selling beer in America.

The 2013 deal gave Constellation exclusive US rights to Modelo, Corona, and related brands, in perpetuity.

This is not a distribution agreement with renewal risk. It is permanent ownership of the US rights to the brands, secured at a price that now looks significantly below intrinsic value.

That permanence is the foundation of every valuation discussion about this company.

Durable Competitive Advantage

Constellation’s moat is a combination of things that rarely travel together: a permanently licensed, category-leading brand with no contract risk, demographic tailwinds from the fastest-growing consumer segment in the United States, and a production infrastructure that is genuinely difficult and expensive to replicate.

The brand architecture

Modelo Especial is the number-one selling beer brand in the United States by dollar sales. It held that position in fiscal 2025 despite the headwinds from Hispanic consumer pullback. Corona Extra is a top-five brand with genuine global recognition. Pacifico grew nearly 20% in fiscal 2025 and is establishing itself as a premium entry-level import.

These brands occupy a distinct and difficult-to-attack position in the market. They are not domestic mass-market beers, and they are not craft beers. They sit in imported premium beer, a segment that has grown consistently for twenty years while domestic mainstream beer (Budweiser, Coors, Miller) has declined or stagnated.

The key driver of that position is not advertising. Constellation spends approximately 9% of net sales on marketing, which is significant but not exceptional for consumer package goods. The driver is authentic cultural resonance, particularly with Hispanic consumers, and a genuine quality differentiation from domestic lagers that is recognized by both Hispanic and non-Hispanic drinkers.

That quality positioning is important for valuation. Brands that win primarily on price are vulnerable. Brands that win on quality and cultural identity are stickier. Modelo is the second category.

The manufacturing moat

Constellation has invested more than $5 billion in Mexican brewing capacity over the past several years. By the end of fiscal 2025 it had approximately 48 million hectoliters of capacity across facilities in Nava and Obregon. The Veracruz brewery is expected to bring total capacity to roughly 55 million hectoliters by fiscal 2028.

This infrastructure is not incidental. It is a genuine competitive asset. A new entrant seeking to replicate Constellation’s US import beer position would need to acquire brands with comparable recognition, secure US distribution relationships, and build comparable production capacity in Mexico. The capital required for that is well north of $10 billion. The time required is measured in decades, not years.

The manufacturing investment also creates a secondary strategic advantage: it ties Constellation’s cost per hectoliter to its own capital, not to a third-party brewer. Gross margins on the beer business run at approximately 47–50% before marketing, high for beverage alcohol, and achievable in part because Constellation controls its own production economics.

The permanent license and what it means for intrinsic value

A business with a permanent, exclusive license to sell the best-selling beer in the largest beer market in the world does not typically trade at a mid-teens earnings multiple. The reason it does today is the cloud of near-term headwinds: tariffs, Hispanic consumer pullback, debt from the wine portfolio misadventures, and capex commitments that have compressed free cash flow.

The investment question is whether any of those headwinds have compromised the permanent value of the license. I will argue that none of them have, but that two of them are more serious than they initially appear.

The Two Risks That Actually Matter

There are a lot of things that could go wrong with Constellation Brands in the near term. Tariffs, aluminum costs, wine divestiture execution, debt levels, capex overruns. I want to be direct: most of these are noise. Two of them are not.

Risk 1: The Hispanic consumer is not a temporary headwind

Constellation’s CEO Bill Newlands has been unusually forthright about this. Hispanic consumers represent approximately 50% of Modelo’s US customer base. This is not a peripheral demographic. It is the foundational customer.

The mechanism through which immigration enforcement is affecting demand is specific and unusual. It is not primarily about income. The data Constellation presented to investors shows that the issue is occasion-based: social gatherings, restaurant visits, community events, the situations where beer is consumed, are declining because Hispanic consumers are reducing public exposure. Internal Constellation omnibus surveys showed that over 75% of Hispanic consumers remain concerned about the socioeconomic environment in the US. High-end beer buy rates among the demographic were declining month-over-month through at least July 2025.

In California, Constellation’s largest state market, construction employment declined both quarter-over-quarter and year-over-year through mid-2025. Constellation specifically tracks what it calls “4,000-calorie jobs”: physically demanding roles in construction and related industries that historically correlate with higher beer consumption. That employment base is under pressure.

This is a meaningful near-term demand problem. The question is whether it is temporary or structural.

My assessment is that it is primarily cyclical rather than structural, but with a longer cycle than most investors are currently modeling. Immigration enforcement is a policy choice, not a permanent demographic shift. The Hispanic population in the United States continues to grow. Consumption occasions will normalize when the policy environment stabilizes. The brand equity that Modelo has built with this demographic over 30 years does not evaporate in 12 months of difficult macro.

However, the recovery timeline is genuinely uncertain. If enforcement remains at current intensity through fiscal 2027, the earnings trough could be deeper and longer than the base case assumes. This is not a risk to dismiss in a valuation model.

Risk 2: The tariff structure creates a permanent cost increase, not just a temporary one

The 25% tariff on imported canned beer and the 50% tariff on aluminum represent a structural cost increase, not a transient one. Constellation has three responses available: pass costs to consumers through pricing, absorb them through operational efficiency, or reduce can volumes by shifting packaging mix.

Management has indicated a combination of all three. “Lightweighting” (reducing aluminum per can) and increased use of domestic scrap aluminum (largely exempt from the 50% tariff) reduce the per-unit impact. Some pricing has already been taken. The question is how much pricing the consumer will accept.

The aluminum tariff alone was estimated to impact the fiscal 2026 cost structure by approximately $20 million in the near term, with the full-year exposure running higher depending on packaging mix. For a beer business generating roughly $8.3 billion in sales, a $20–40 million cost headwind is manageable. But this is a floor estimate under current tariff rates, rates that have already been adjusted upward once and could move again.

The deeper concern is whether pricing actions to offset tariff costs will accelerate the demographic pullback. A consumer already cautious about spending who faces a 5–10% price increase on their preferred beer may trade down. That dynamic, tariffs compressing margins while simultaneously reducing volumes through pricing pressure, is the scenario that justifies the market’s current multiple compression.

The bear case is free. The investment thesis is what you’re paying for. What follows is the financial deep-dive: normalized free cash flow, the debt picture, what the wine divestiture actually changes about this company’s valuation, and the specific conditions under which the thesis holds or breaks down. If you’ve found this analysis useful, consider supporting it — it takes a long time to get these right.

Read more

People open a chart, scroll through recent headlines, get excited about a story, and then go looking for numbers to support the conclusion they’ve already reached. By the time they get to the actual business, confirmation bias is already doing most of the work.

My process is a bit different, though I’ll be honest about how it actually works, rather than how it sounds in theory.

I do look at price first. Not in depth, but enough to know whether the stock is even in the right ballpark. If something trades at 60 times earnings with no clear path to justifying that multiple, I don’t need to spend two hours on the financials to know it’s not for me. A quick check of a few relative valuation metrics, P/E, P/S, P/B against sector peers and historical averages, tells me whether there’s a plausible case to investigate further.

But that’s a coarse filter, not a conclusion. A cheap-looking stock can be a value trap. An expensive-looking one can still be worth understanding. So once something passes that initial price screen, I put the valuation aside and turn to the business itself.

That’s where these five numbers come in. Before I build a model, before I read the earnings call transcript, before I form any real view on what the company is worth, I run through five metrics. They take about ten minutes to find. And they tell me roughly 80% of what I need to know about whether a business deserves serious attention.

Here they are.

1. Free Cash Flow Margin

Free cash flow is the money left over after a company pays all its operating expenses and makes the capital investments needed to keep the business running. It’s what’s actually available to shareholders, not accounting profit, not adjusted EBITDA, not any of the other numbers companies prefer to highlight.

The margin, free cash flow divided by revenue, tells me how much of every dollar of sales actually turns into real cash.

I’m not looking for a specific number here. A software business running at a 35% FCF margin and a grocery retailer running at 5% can both be excellent investments. Context matters enormously. What I’m looking for is three things: Is the margin positive? Has it been stable or improving over the last three to five years? And does it make sense given the business model?

A company consistently burning cash, or one where margins are quietly shrinking year after year without explanation, raises a flag, not necessarily a reason to walk away immediately, but a reason to understand why before going deeper. Sometimes there’s a good answer: heavy reinvestment in growth, a transitional year, an industry with lumpy capex cycles. Sometimes there isn’t. Either way, I want to understand the cash generation story early, because everything else in the analysis rests on it.

2. Return on Invested Capital (ROIC)

If free cash flow tells me whether a business makes money, ROIC tells me whether it makes money efficiently.

ROIC measures how much profit a company generates relative to the capital invested in it, equity plus debt combined. A high and stable ROIC is one of the clearest signals that a company has a real competitive advantage. It means the business is earning more than it costs to run, which is harder than it sounds and rarer than most investors realize.

I look for ROIC above the cost of capital. Roughly speaking, if a business earns 15% on its invested capital and its cost of capital is 9%, it’s creating real value. If ROIC is below the cost of capital, or declining toward it, the business is destroying value even if it’s growing revenue.

This is why I’m skeptical of companies that grow fast but have mediocre returns on capital. Growth without returns is just an expensive way to run in place.

The other thing I look at is the trend. A business with 20% ROIC that’s been declining for three straight years tells a different story than one that’s held steady at 15% through multiple economic cycles. Falling ROIC is often the earliest signal that a competitive advantage is eroding, before it shows up in earnings, before it shows up in the stock price.

3. Gross Margin

This one is quick, but it’s important.

Gross margin is equal revenue minus cost of goods sold, divided by revenue, tells me something fundamental about the economic structure of the business. High gross margins mean the product or service is genuinely valued by customers and that the company has pricing power. Low gross margins mean the business is competing primarily on cost, which is a harder game to win long term.

A software company with 80% gross margins has a completely different investment profile than a consulting firm with 16% gross margins, even if both are growing at the same rate. The software company has enormous operating leverage, as revenue grows, costs don’t grow nearly as fast, and profits expand quickly. The consulting firm has to hire more people to grow, which limits how much margin expansion is possible.

I also use gross margin as a sanity check. If a company reports strong net income but thin gross margins, I want to understand exactly where that profitability is coming from, and whether it’s durable. One-off items, favorable accounting, or temporarily low expenses can flatter net income. Gross margin is harder to dress up.

4. Net Debt Position

This one is simple, and most investors underweight it.

Net debt is total debt minus cash and liquid equivalents. A positive net debt number means the company owes more than it holds. A negative number, what some call a net cash position, means the company has more cash than debt.

I care about this for two reasons.

First, debt amplifies risk. In a downturn, a heavily indebted company faces pressure from multiple directions at once: revenue falls, credit becomes expensive, and lenders get nervous. A company with no debt or net cash can take the same hit and wait it out. That asymmetry matters a lot in the kinds of extended holds I prefer.

Second, debt constrains capital allocation. A company spending a large portion of its cash flow on interest and debt repayment has less flexibility to invest in growth, make opportunistic acquisitions, or return capital to shareholders. The best businesses tend to have options. Heavy debt removes options.

I’m not dogmatic about this. Some businesses, particularly ones with highly predictable cash flows, can carry meaningful leverage responsibly. But when I see a company with debt exceeding five or six times annual earnings, I apply a higher level of scrutiny to everything else. The margin for error shrinks considerably.

5. Revenue Trend (5-Year Direction)

I save this one for last, not because it’s least important, but because it means something different in context.

I don’t screen for growth. I screen for direction.

What I want to know is simple: Has this business been growing revenue over the last five years, and has that growth been consistent? Not explosive, consistent. A business growing revenue at 7–10% per year, year after year, with few surprises, tells me something very different from one that had a great year, then a flat year, then a big year, then a contraction.

Consistency matters because it’s usually a signal that the business has some underlying demand driver that isn’t purely cyclical or dependent on one product or customer. It tells me there’s something structural going on, not just a favorable moment in time.

I also look for the type of revenue. Is it recurring? Contract-based? Or entirely transactional? Recurring revenue with visible renewal rates is worth a premium. Pure transaction revenue that shows up only when a customer decides to buy is more uncertain, not bad, just different.

Revenue trends also help me spot potential value traps. A company that looks cheap on earnings but has been shrinking revenue for three consecutive years deserves serious scrutiny. Often, the market is right about the direction, and the low P/E is not a bargain, it’s a warning.

What These Five Numbers Actually Do

Running through these five metrics doesn’t give me an answer. It gives me a starting point.

What I’m really doing is filtering for business quality before I spend any serious time on valuation. If a company has positive and stable FCF margins, strong ROIC, solid gross margins, manageable debt, and a consistent revenue trend, I’m willing to do the next level of work. If several of these look weak, I want a very clear reason to keep going, and that reason had better be compelling, not just a story I’ve talked myself into.

There are thousands of publicly traded companies in the world. Most of them are not worth analyzing. These five numbers help me narrow the list to the ones that are.

Then I go back to price, this time with enough context to know what I’m actually paying for.

Detailed valuation breakdowns, updated theses, and the full analysis behind every stock I look at are in the paid archive. If you find this kind of thinking useful, consider subscribing.

My approach is different. Before I form an opinion on the stock, I want to understand the business - what it actually earns, what protects it, and what could structurally change its earning power over time. Only then does price become meaningful.

When I do that with Pfizer, I find a company that is genuinely difficult to categorize. It is not a cheap stock pretending to be a compounder. It is not a broken business masquerading as a value play. It sits somewhere more uncomfortable: a high-quality franchise in the middle of a structural transition, burdened by a heavy acquisition balance sheet, facing a predictable revenue erosion event, and betting its next decade on a pipeline that has yet to prove itself.

That complexity is precisely why it deserves a careful look.

Understanding the Business

Pfizer is one of the largest pharmaceutical companies in the world, with annual sales of more than $60 billion. Its revenue comes from three broad sources: a diversified portfolio of branded drugs and vaccines (its core), a contract manufacturing and active pharmaceutical ingredient business (Pfizer CentreOne), and a smaller incubator platform for early-stage partnerships (Pfizer Ignite).

The core portfolio is the story. Key franchises include Eliquis (a blood thinner, co-marketed with Bristol-Myers Squibb), the Prevnar vaccine family, oncology drugs including Ibrance, Padcev, and Adcetris, the cardiovascular drug Vyndaqel, and the now-declining COVID products Comirnaty and Paxlovid.

The revenue model is straightforward: develop or acquire patent-protected drugs, sell them at premium prices during the exclusivity window, generate substantial cash, and reinvest in the next generation. When it works, pharmaceutical economics are exceptional, gross margins above 70%, minimal variable costs, and durable pricing power. Pfizer’s gross margin in fiscal year 2025 was 75.8%.

The business model is easy to understand. What is difficult to evaluate is whether the model will work for Pfizer specifically, over the next five to seven years, given the pressures it currently faces.

The COVID Hangover and What’s Left

To understand where Pfizer is today, you have to understand where it came from.

Pfizer reported more than $100 billion in revenue in 2022, at the height of pandemic demand for its COVID-19 vaccine and antiviral portfolio. By 2023, revenue had fallen to $59.5 billion, before recovering modestly to $63.6 billion in 2024. That collapse, nearly $42 billion in revenue gone in one year, was not a business failure. It was a normalization after an extraordinary event. But it left a mark.

To fill the gap and accelerate growth, Pfizer spent aggressively on acquisitions: the $43 billion acquisition of Seagen, which shifted the portfolio’s center of gravity toward oncology and antibody-drug conjugates. It also acquired Metsera, gaining clinical-stage obesity assets, and entered a major licensing deal with Chinese biotech 3SBio for a PD-1×VEGF bispecific antibody.

Full-year 2025 revenues totaled $62.6 billion, with non-COVID revenue growing 6% operationally. Strip out the declining COVID tail and the underlying business is growing. That is an important distinction that the headline revenue number obscures.

Durable Competitive Advantage

Pfizer’s moat is real but narrowing in places, which is the central tension of the investment case.

The core advantage in pharmaceuticals is patent protection: a government-granted monopoly on a drug for a defined period, during which the company can charge premium prices with minimal competition. Pfizer has built additional layers around this through scale, its global distribution network, manufacturing capabilities, regulatory expertise, and relationships with healthcare systems worldwide. These are not easily replicated.

The Seagen acquisition added a specific and valuable capability: antibody-drug conjugates (ADCs), a next-generation oncology technology that combines a targeted antibody with a toxic payload to attack cancer cells more precisely. ADCs represent one of the most active areas of oncology development, and Pfizer is now one of the leading players. This is a genuine long-term competitive asset.

The challenge is that the moat on Pfizer’s existing blockbusters is eroding on a known schedule. Pfizer’s CFO confirmed the company expects $17 billion in revenues to be impacted by patent and regulatory exclusivity expirations. The key drugs facing this pressure are Eliquis, the blood thinner that is among Pfizer’s top sellers with core patents extending into late 2026. Ibrance, a key breast cancer therapy; and Xtandi, a prostate cancer franchise. These are not speculative risks, they are predictable, dated, and quantified by management.

The moat is not destroyed. But it is actively shrinking on the legacy portfolio, and the new moat being built through Seagen and Metsera has yet to generate the revenues needed to replace what is being lost.

Financial Health

This is where the picture becomes genuinely complicated for a value investor.

Revenue and margins: Full-year 2025 revenues of $62.6 billion with a non-COVID portfolio growing at 6% operationally. Gross margin of 75.8% is excellent. These are the numbers of a high-quality pharmaceutical business.

The debt burden: As of fiscal year 2025, Pfizer had $1.1 billion in cash and equivalents together with $12.5 billion in short term investments against $63.9 billion in long-term debt. This is the balance sheet legacy of the Seagen acquisition. For a company generating roughly $9–10 billion in annual free cash flow, that represents a meaningful leverage ratio, and it constrains capital allocation flexibility.

Free cash flow and the dividend problem: Pfizer paid $9.8 billion in dividends in FY2025 against free cash flow of just $9.1 billion, meaning the dividend exceeded free cash flow by around $700 million. This is not a catastrophe, but it is a yellow flag. The company is essentially funding its dividend by drawing down reserves. Pfizer has increased its dividend for 16 consecutive years and has declared dividends for 349 consecutive quarters since 1937 - a streak management will defend aggressively. But the math needs to improve.

“Below: the full scenario analysis with base, bull, and bear case price targets and the specific price at which I’d actually consider buying.”

Read more

First version: 9% discount rate. Second version: 10%.

The intrinsic value changed by $18 per share. That was roughly 22% of the stock price.

I hadn’t changed a single thing about the business. Not the revenue forecast, not the margin assumptions, not the growth trajectory in year seven. I just moved one input by one percentage point, and my model handed me a completely different answer about what the company was worth.

That’s the discount rate problem. If you’re building DCF models, you almost certainly have it too. And the uncomfortable part isn’t that the math is wrong. It’s that the math is working exactly as intended, and we don’t like what it’s revealing.

What the discount rate actually is

In a discounted cash flow model, the discount rate does one job: it converts future money into today’s money. A business that earns $100 in ten years is worth less to you today than $100 right now, because of time, inflation, and the risk that the $100 never actually arrives.

The discount rate captures all of that, in a single number.

Higher discount rate: future cash flows are worth less today, intrinsic value falls. Lower discount rate: future cash flows are worth more, intrinsic value rises.

This is why that one-percentage-point shift above had such an outsized effect. And it gets more pronounced the further out you project, which matters a lot, because in most DCF models, the terminal value (everything beyond year ten) makes up between 60% and 80% of the total intrinsic value estimate.

In other words, the number you’re most uncertain about is also the one with the most leverage over your answer.

Where most investors get their discount rate

The textbook approach is CAPM, the Capital Asset Pricing Model. You take the risk-free rate (usually a government bond yield), add an equity risk premium, and multiply by beta, which measures how volatile the stock has been relative to the market.

This sounds rigorous. In practice, I think it’s largely backwards.

Beta measures historical price volatility. It does not measure business risk. A stock that fell 40% during a broad market panic may not be riskier as a business than one that held steady, it may just have been mis-priced or illiquid. Basing the discount rate on that number means you’re importing the market’s past mood into a model about a business’s future cash flows.

The equity risk premium is even more slippery. It’s an estimate of the return investors demand above the risk-free rate for holding equities in general. Depending on which academic paper you read, which time period you use, and whether you’re looking at realized or implied premia, it can range anywhere mostly from 3% to 12%. That’s not a measurement. That’s a range wide enough to drive a full investment thesis through, in either direction.

What this produces, in practice, is a discount rate that feels derived but is actually chosen. Most value investors, myself included, for longer than I’d like to admit, are picking a number between 8% and 12%, rounding it to a whole number, and calling it rigorous. The spreadsheet gives it an air of precision that the underlying logic doesn’t support.

The leverage problem at the terminal value

Here’s where this gets genuinely uncomfortable.

Most DCF models project cash flows for ten years and then calculate a terminal value, an estimate of what the business is worth in perpetuity beyond that point. The terminal value is typically calculated as a simple formula: final-year free cash flow divided by (discount rate minus long-term growth rate).

Take a business generating $100M in free cash flow in year ten, with a 3% long-term growth rate. At a 9% discount rate, the terminal value is $100M ÷ 0.06 = roughly $1.67 billion. At a 10% discount rate, it’s $100M ÷ 0.07 = $1.43 billion.

That’s a $240 million difference in terminal value, from one percentage point in the discount rate, before you’ve even discounted it back to the present.

Now add the discounting effect over ten years, and the gap compounds further. What started as a one-point assumption difference ends up as a 15–25% swing in your total intrinsic value estimate.

This is not a model flaw. This is how the mathematics of present value works. The problem is that most investors treat the output, the intrinsic value estimate, as if it were a measured quantity, when it is downstream of an assumption that nobody can verify.

A real example from my own work

I’m going to use Novo Nordisk, because I’ve written about it here and the numbers are already public.

When I built my base-case valuation earlier this year, I used a discount rate of 9%. Given the CagriSema setback and the 2026 guidance cut, I ran a bear case. At 9% discount rate, my bear-case intrinsic value came out around $58. At 10%, it came out around $50. At 8%, it was closer to $69.

Same business. Same assumptions about revenue decline, margin compression, and long-term market share. Three discount rates, three answers spanning a $19 range.

What discount rate is “right” for Novo Nordisk? That depends on how you weight: the competitive risk from Eli Lilly, the execution risk on next-generation pipeline, the regulatory risk across geographies, the currency risk, and the macro environment for healthcare spending. None of those factors produce a number. They produce a judgment.

I’m not saying the DCF is useless. I’m saying it has less precision than it appears to, and the discount rate is the single largest source of that gap.

What I’ve changed in my own process

The honest answer is that I no longer try to find the “correct” discount rate. I try to find a range that captures my actual uncertainty, and then I look at what the model requires me to believe at each end of that range.

In practice, this means running three scenarios for every valuation I build: a bear case at a higher discount rate (usually 10–11%), a base case at a mid-rate (8.5–9.5%), and a bull case at a lower rate (7.5–8.5%). The specific numbers depend on the business, a highly leveraged, cyclical company warrants a higher rate across all three than a company with a net cash balance sheet and decades of stable cash generation.

What this does is force me to confront the question I was previously avoiding: at what discount rate does this stock stop being attractive? If the answer is “it stops working at anything above 8.5%,” I know that my investment case depends heavily on the market agreeing to price the business at a low required return. That’s a different kind of risk than business risk, and it matters.

The other change: I’ve stopped treating the intrinsic value output as a target price. Instead, I use it as a map of assumptions. When the stock is trading at $50 and my model says fair value is $68, the useful question isn’t “should I buy at a 26% discount?” It’s “what does the market have to believe to justify $50, and do I disagree with that?”

Often, the market’s implied assumptions are more defensible than I initially thought, or defensible for a reason I hadn’t fully weighted. The model makes that visible in a way that a quick multiple comparison doesn’t.

The question that changes how you use a DCF

There’s a question I try to ask before I finalize any valuation now, and it’s surprisingly uncomfortable:

“If I’m wrong about the discount rate by one percentage point, does the investment still make sense?”

Not “does it still look cheap?” - cheap is a relative term. Does it still make sense as a long-term position, given the margin of safety I require and the risks I’ve identified in the business?

If the answer is yes, if I can stress the discount rate meaningfully and still find an attractive entry point, that’s a signal the investment case has real depth. The expected return isn’t all sitting in one narrow assumption.

If the answer is no, if the thesis only works at exactly 8.7% and falls apart at 9.5%, that tells me something important. It tells me the margin of safety I think I have may be mostly a function of how I chose one input on a spreadsheet.

Valuation is always part art, part science. I knew that before I started building DCF models. What took longer to internalize is that the art isn’t in the growth rate or the margin assumption, those feel uncertain, so we treat them as estimates. The art is also in the denominator. The number that feels most mechanical is the one that deserves the most skepticism.

What Buffett and Munger actually think about DCFs

This is where it gets interesting, because the two most celebrated value investors of the last century have both expressed deep skepticism about the tool that most value investors treat as their primary instrument.

Warren Buffett has said that he has never seen Charlie Munger use a DCF model. Not rarely. Never. Munger himself described DCF as something that “looks scientific but isn’t.” His view, stated bluntly at various Berkshire meetings over the years, is that if a business requires a spreadsheet to tell you it’s attractive, it probably isn’t attractive enough. The opportunity should be obvious. If you need ten years of projected cash flows and a precise discount rate to justify the investment, you’re likely working too hard to convince yourself of something that isn’t true.

Buffett’s own framing is different but arrives at the same place. He talks about intrinsic value as the discounted value of future cash flows in a conceptual sense, but in practice, his decisions are grounded in qualitative conviction about competitive advantage, management quality, and business economics. He’s famously said that he doesn’t use a formal DCF for most investments. What he does instead is ask: “Do I understand this business well enough to be confident it will generate substantially more cash in ten and twenty years than it does today? And am I paying a price that reflects reasonable caution rather than optimism?”

That’s a very different process from building a three-tab spreadsheet and trusting the output.

The lesson I take from both of them isn’t “don’t build models.” It’s that the model should confirm a conviction you’ve already formed through qualitative analysis, not create one. If you find yourself looking at the DCF output and thinking “I guess this works,” that’s a warning sign. The numbers should be reinforcing clarity, not manufacturing it.

DCF as a support tool, not a verdict

I still build DCF models. I find them genuinely useful. But I’ve stopped treating the intrinsic value output as the conclusion of my analysis. It’s one input among several, and usually not the most important one.

Here’s how it actually fits into my process now.

The real work happens before the model opens. That means understanding the business deeply enough to have a view on its competitive position in ten years. It means assessing management quality, capital allocation history, and balance sheet resilience. It means identifying the two or three variables that actually determine long-term outcomes for this specific company , and having a considered view on each of them.

Only after all of that do I build the DCF. At that point, the model serves three purposes. First, it forces me to make my assumptions explicit. Vague bullishness doesn’t survive contact with a spreadsheet, you have to commit to numbers. Second, it lets me stress-test my conviction by varying the discount rate and growth assumptions across a range. Third, it helps me understand what the market is pricing in, by working backwards from the current stock price to its implied assumptions.

What the DCF does not do, and cannot do, is tell me whether to buy. That decision rests on the qualitative analysis: the moat, the management, the margin of safety, and whether the downside scenario is one I can accept. The model maps the range of outcomes. The judgment call is mine.

This is, I think, closer to how serious long-term investors actually work, even when they don’t say so explicitly. A DCF that produces a number of $68 on a $50 stock isn’t a buy signal. It’s an invitation to ask: do I believe in the business enough that a range of $52 to $80, depending on assumptions, still makes this worth owning for the next decade?

If the answer is yes, the discount rate matters less than you think. If the answer is no, even a precise model won’t save you.

If this was useful, the paid archive goes deeper: valuation ranges, updated theses when facts change, and the analytical framework applied in real time to specific businesses. Every note includes the assumptions I’m making and, when I’m wrong, why.

My approach is different. I start by understanding the business, its model, its economics, its durability, before I ever look at price. And when I do that with Dino Polska, I find something genuinely interesting: a founder-led compounder with a differentiated position, growing rapidly in a large and fragmented market, now trading near multi-year lows after a sharp selloff.

This is not a perfect business. But it may be a good one at an improving price.

Understanding the Business

Dino Polska operates a network of medium-sized grocery supermarkets across Poland, primarily in small towns and rural areas. The typical Dino store is around 400 square meters, much smaller than a Biedronka or Lidl, and carries around 5,000 products, with a focus on fresh food and branded regional goods.

What makes Dino unusual in grocery retail is vertical integration. The company owns Agro-Rydzyna, a meat processing plant that supplies fresh meat to every store daily. Every Dino has an in-store butcher counter, which is a meaningful differentiator in Poland, where fresh meat is a priority for consumers. Competitors like Biedronka and Lidl largely offer pre-packaged meat.

Dino also owns most of its stores outright, the land, the building, the infrastructure. This is not common in retail, where leasing is the norm. Owning its real estate gives Dino a structural cost advantage of roughly 2–3% relative to competitors who pay rent, and it provides long-term stability in a sector where location economics matter enormously.

The revenue model is straightforward: foot traffic multiplied by basket size multiplied by margin. Dino does not run loyalty programs, does not advertise, and does not compete on price in the traditional sense. It competes on proximity. In small towns and villages, Dino often is the only modern grocery option.

Durable Competitive Advantage

Dino’s moat is geographic and structural rather than brand-based or technological.

The core insight is simple: Biedronka and Lidl, the two dominant grocery chains in Poland, are optimized for urban and suburban catchment areas with populations of 20,000 or more. Their store formats are too large, their cost structures too inflexible, and their product mix too broad to be efficiently deployed in villages of 1,000 to 5,000 people. Dino identified this gap early and has systematically filled it.

As Dino expands into rural Poland, it is not taking customers from Biedronka, it is replacing the trips those customers previously made to distant towns. That is a qualitatively different competitive dynamic. It is expansion into whitespace, not a price war.

The combination of land ownership, vertical integration in fresh meat, rural positioning, and zero advertising spend creates a cost structure that is genuinely difficult to replicate. A competitor entering this market would face years of capital expenditure building real estate, logistics, and supply chain infrastructure, and would be doing so into markets that Dino has already occupied.

The moat is real. It is not impenetrable, but it is durable.

Growth and Financial Strength

Dino’s track record of growth is exceptional for a grocery retailer.

Revenue has grown at a CAGR of roughly 19% over the past decade. In 2025, the company reported revenues of PLN 33.6 billion, up 14.9% year over year, while expanding its store network by 345 locations to reach 3,033 stores. Like-for-like sales grew 4.4% for the full year. Operating cash flow covered 122% of capital expenditure needs, reflecting the self-funding nature of the model.

The financial profile reflects the realities of grocery retail: margins are modest but competitive. The net profit margin runs around 5%, operating margins near 6–7%, and EBITDA margins close to 8%. These figures compare favorably against peers. Biedronka, for context, operates with lower margins despite its greater scale and market share.

Return on invested capital has historically been strong, above 20% at peak, though margin compression from a price war across the Polish grocery sector has weighed on returns in recent years. Free cash flow generation remains solid, and the balance sheet carries minimal net debt (net debt to EBITDA below 0.5x), providing both resilience and flexibility.

Risks

Dino is not without meaningful risks. A value investor must take them seriously.

Price war. The Polish grocery market is in the middle of an intense competitive battle between Biedronka, Lidl, and Dino. All three have been cutting prices to protect and grow market share. This has compressed margins across the sector and contributed to the recent share price decline. Dino’s share price fell more than 15% following the release of its 2025 full-year results, with investors reacting to margin pressure more than to the underlying store economics. The question is whether this is a temporary cycle, a battle Dino is well-positioned to survive given its structurally lower cost base, or a structural shift in the sector’s profitability ceiling.

Dino’s advantage here is meaningful. Because it owns its real estate and has lower per-store overhead than Biedronka, it has more room to absorb price reductions without destroying returns. Biedronka, operating with a leasing model and lower margins, faces greater financial strain in a sustained price war. This does not make Dino immune, but it does suggest the company is better positioned than its competitors to emerge from this period with market share intact or improved.

LFL growth deceleration. After years of strong same-store sales growth, like-for-like performance has slowed materially. In early 2025, LFL growth fell well below the inflation rate. Management expects a recovery as pricing normalizes and wage-driven consumer spending continues to strengthen. Poland’s minimum wage has risen substantially, and real household income growth remains supportive. But the short-term trend creates real uncertainty about the pace of earnings recovery.

Growth saturation. Dino’s strategy depends on continuing to find viable rural locations to build stores. Expansion into eastern Poland extends the runway, but the whitespace is finite. At some point, perhaps at 4,000 to 5,000 stores, organic expansion slows and the growth thesis changes. International expansion would then be necessary to sustain historical growth rates, and that introduces execution risk of a different order.

Geopolitical exposure. Poland borders Ukraine. A broader regional conflict could disrupt supply chains, consumer confidence, or the broader economic environment. This is a tail risk, not a base case, but it is not zero, and it is a consideration that does not exist for Western European peers.

Dino has fallen 40% from its highs, owns its real estate outright, and has compounded revenue at 19% annually for a decade. But the price war is real, like-for-like growth has stalled, and the FCF multiple looks alarming at first glance. Below I explain why I think the market is misreading one number in particular, and walk through what Dino is actually worth if store expansion continues on its current trajectory. For paid subscribers.

Read more

But in reality, many of the worst investments start exactly this way.

They don’t look like obvious mistakes at the beginning. In fact, they often look like smart, well-reasoned ideas. The valuation seems attractive, the downside appears limited, and the upside looks appealing. But over time, the story starts to fall apart, not because the market failed to notice something, but because the business itself wasn’t as strong as it seemed.

That’s what a value trap is.

A value trap isn’t just a bad investment. It’s a convincing bad investment. It looks cheap, but the low price is there for a reason. The market isn’t missing something, it’s pricing in real problems.

Learning how to tell the difference between a genuinely undervalued stock and a value trap is one of the most important skills you can develop as an investor. And it starts with looking beyond simple valuation metrics.

The Illusion of “Cheap”

At the center of most value traps is a misunderstanding of what “cheap” really means.

A stock trading at 8x earnings might look attractive compared to one trading at 25x. But that only makes sense if those earnings are stable and sustainable. And often, they’re not.

Earnings can be temporarily high because of favorable conditions, one-off factors, or peak cycles. When things normalize, earnings drop, and suddenly that “cheap” stock doesn’t look so cheap anymore.

This is why looking at valuation alone can be misleading.

The better question is not: Is this stock cheap?
It’s: Are these earnings reliable over the long term?

If the answer is no, then the valuation doesn’t really matter.

Why Value Traps Are So Easy to Fall Into

Value traps are tricky because they play into some very natural instincts.

One of the biggest is anchoring. If a stock used to trade at €100 and now trades at €40, it feels like a bargain. But that €100 price might have been based on overly optimistic assumptions or unusually strong conditions.

The market doesn’t care about where the stock used to trade. It cares about what the business is worth today, and what it will be worth in the future.

Another common trap is believing too strongly in mean reversion. It’s easy to assume that a struggling business will “bounce back” to its previous performance. Sometimes that happens. But sometimes the business has fundamentally changed.

Industries evolve. Competition increases. Technology disrupts old models. In those cases, there may be no “normal” to return to.

And then there’s the psychological comfort of buying cheap. It feels safer than buying a high-quality company at a higher price. But that sense of safety can be misleading. A weak business at a low price can be much riskier than a strong business at a fair price.

Cyclical vs. Structural Problems

One of the most important things to figure out is whether a company is going through a temporary setback or a long-term decline.

Some businesses are cyclical. Their performance goes up and down with the economy. During downturns, earnings fall, and the stock may look cheap. But when conditions improve, the business recovers.

In those cases, low valuations can be real opportunities.

But other businesses are facing structural problems. Demand may be shrinking permanently. New competitors may be taking market share. Technology may be making their products less relevant.

These businesses don’t just recover with time, they continue to decline.

The challenge is that, at first glance, both situations can look very similar. Revenue drops, margins shrink, and sentiment turns negative. The difference is what’s happening underneath.

Mistaking a structural decline for a temporary one is one of the most common ways investors fall into value traps.

Focus on the Business First

To avoid value traps, you have to shift your focus.

Instead of starting with valuation, start with the business.

Ask yourself: Is this company actually getting better or worse?

Look at the long-term trends. Is revenue growing or shrinking? Are margins stable or under pressure? Is the company generating solid returns on capital, or are those returns declining?

If the overall direction is negative, the low valuation might be justified.

On the other hand, if the business is still strong but going through a temporary issue, that’s where opportunities can exist.

Why Return on Capital Matters

Return on invested capital (ROIC) is one of the most useful ways to judge a business.

It tells you how efficiently a company turns its investments into profits. High and stable returns usually mean the company has some kind of competitive advantage. Declining returns can be a sign that something is going wrong.

A lot of value traps have one thing in common: falling returns on capital.

That’s a warning sign. It suggests the business is becoming less efficient, even if earnings haven’t dropped yet. Over time, that usually leads to weaker performance, and often a lower stock price.

Be Careful With High Dividend Yields

High dividend yields can look very attractive. They promise income and can make a stock seem safer.

But sometimes, a high yield is actually a red flag.

If the stock price has dropped significantly, the yield goes up automatically. The market may be signaling that the dividend isn’t sustainable.

If the company’s cash flow is under pressure, it may eventually have to cut the dividend. And when that happens, the stock often falls further.

So instead of focusing on the yield itself, it’s better to ask: Can this company comfortably afford to pay this dividend?

Competitive Advantage Isn’t Always Obvious

A strong business usually has some kind of advantage, something that makes it hard for competitors to take market share.

But that advantage isn’t always as strong as it seems.

A company might have a long history, a recognizable name, or a solid customer base. But if customers can easily switch to alternatives, or if competitors can offer similar services, that advantage may not last.

Many value traps come from overestimating how durable a business really is.

Management Matters More Than You Think

The quality of management can make a big difference, especially when a company is facing challenges.

Good management teams adapt. They invest wisely, make tough decisions, and position the company for the future.

Weaker teams often do the opposite. They try to maintain the status quo, make questionable acquisitions, or focus too much on short-term results.

These decisions don’t always show up immediately in the numbers, but over time they can have a big impact.

What a Real Margin of Safety Looks Like

A lot of investors think a low P/E ratio automatically means there’s a margin of safety.

But that’s not really true.

If the business is getting worse, its true value is probably falling too. In that case, the stock isn’t cheap, it just looks cheap.

A real margin of safety comes from owning a solid business at a reasonable price. You want something that can hold its value, or even grow it, over time.

Without that, the price alone doesn’t protect you.

What Good Opportunities Actually Look Like

Not every cheap stock is a trap. Some are genuinely undervalued.

But those opportunities usually don’t look as obvious.

They often involve short-term problems or uncertainty, while the underlying business remains strong. The company still has good economics, capable management, and a clear path forward.

The market may be too pessimistic, creating an opportunity.

The key difference is that the business itself is still intact.

Final Thoughts

Value traps are part of investing. You can’t avoid them completely, but you can get better at spotting them.

The biggest shift is learning to look beyond the surface. A low valuation doesn’t automatically mean a good opportunity. In many cases, it’s a warning.

Instead of asking, “Is this cheap?”, ask “Why is this cheap?”

And more importantly: Is that reason temporary or permanent?

That one question can save you from a lot of bad investments.

In the end, successful investing isn’t about buying what looks cheap. It’s about understanding what’s actually valuable, and being patient enough to wait for the right opportunities.

My approach is different. Like with any investment, I begin by analyzing the underlying business, its economics, its durability, and its ability to generate long-term value. Only after that do I assess whether the stock price offers an attractive opportunity.

Alten represents an interesting contrast to high-growth technology companies. It is a well-run, profitable business with strong client relationships and a conservative balance sheet. However, unlike software platforms, it lacks scalability, operates with lower margins, and faces structural limitations that impact long-term returns.

Using a value investing framework inspired by Benjamin Graham, this analysis will explain why Alten is a solid company, potentially undervalued, but not necessarily a top high-return compounder. For disciplined investors, the key question is not whether Alten is a good business, it is whether it is a good investment at the current price.

Understanding Alten’s Business

Alten is an engineering and IT services company that provides technical consulting to large organizations across a range of sectors, such as:

• Automotive

• Aerospace & Defense

• Energy & Utilities

• Telecommunications

• Finance & IT

Founded in 1988, the company has expanded globally and now employs tens of thousands of engineers. Its business model is straightforward: it supplies highly skilled consultants who work on client projects, often embedded within the client’s operations.

Unlike software companies that sell products or subscriptions, Alten generates revenue by billing for time and expertise. This distinction is crucial. It defines both the strengths and the limitations of the business.

Alten’s services are often mission-critical. Engineers may be involved in designing automotive systems, optimizing industrial processes, or supporting aerospace development. These are complex, high-value activities that require specialized knowledge, which gives Alten credibility and long-standing relationships with major clients.

Its revenue model is fundamentally billable hours x utilization x pricing power, not scalable products or SaaS. Unlike software companies whose costs to serve additional customers are low, Alten grows by expanding its workforce and deploying them on client projects.

This model is inherently less scalable than software. A software company can serve thousands of additional customers with minimal incremental cost. Alten, by contrast, must hire more people to grow revenue. As a result, its growth is linear rather than exponential.

Core Business Characteristics

To better understand Alten’s investment profile, it is useful to break down its core characteristics.

First, Alten operates in a highly competitive industry. Numerous firms provide engineering and IT consulting services, including global players like Accenture and Capgemini. While Alten has built expertise in specific sectors, it does not possess a dominant or exclusive position.

Second, the company benefits from long-term client relationships. Many contracts extend over multiple years, and clients often rely on Alten’s engineers for continuity and expertise. This provides a degree of revenue stability.

Third, the business is sensitive to economic cycles. When industrial activity slows, companies reduce external consulting spend. This can lead to lower utilization rates, slower hiring, and margin pressure.

These characteristics position Alten as a steady but cyclical business, rather than a structurally high-growth platform.

Growth and Financial Strength

Alten has demonstrated solid growth over the long term, particularly during periods of economic expansion. However, its growth profile is uneven and closely tied to macroeconomic conditions.

In recent years, the company experienced strong post-pandemic growth as industrial activity rebounded. However, more recently, growth has slowed and even declined slightly in certain regions, particularly those exposed to automotive and European industrial markets.

This highlights a key reality: Alten does not control its growth trajectory in the same way a software company might. Demand depends heavily on client budgets and broader economic conditions.

Despite this, Alten’s financial profile remains robust.

Profitability

Alten operates with moderate margins typical of consulting businesses. Operating margins are generally in the high single digits, while net margins are lower. This reflects the labor-intensive nature of the business.

Compared to high-margin software firms, this is a significant limitation. Lower margins mean less operating leverage and less capacity for exponential earnings growth.

However, the company compensates with consistency. It has demonstrated the ability to remain profitable across cycles, which is an important consideration for long-term investors.

• Revenue: ~€4.119 billion (trailing 12 months)

• Gross Profit: ~€672 million

• Operating Income: ~€321 million

• Net Income: ~€151 million

• Gross Margin: ~16.3%

• Operating Margin: ~7.8%

• Profit Margin: ~3.7%

• Free Cash Flow Margin: ~9.6%

• ROE: ~7%

• ROA: ~4%

• ROIC: ~8%

Cash Flow

Alten generates steady free cash flow, supported by its relatively asset-light model. While it must invest in hiring and talent development, it does not require large capital expenditures.

This allows the company to fund its own growth, pay dividends, and pursue acquisitions without excessive reliance on debt.

Balance Sheet

One of Alten’s strongest attributes is its balance sheet. The company maintains a net cash position, providing financial flexibility and resilience.

From a value investing perspective, this is a significant advantage. A strong balance sheet reduces downside risk and provides a margin of safety during economic downturns.

• Net Debt: approximately neutral (modest debt vs cash)

• Operating Cash Flow: ~€409 million

• Free Cash Flow: ~€394 million

• Current Ratio: ~1.55

Durable Competitive Advantage

A key question for any long-term investor is whether a company has a durable competitive advantage, or “moat.”

In Alten’s case, the moat is moderate rather than strong.

Sources of Advantage

Alten benefits from several competitive strengths:

• Client relationships: Long-standing partnerships with large industrial companies create repeat business.

• Sector expertise: Deep knowledge in industries such as aerospace and automotive enhances credibility.

• Scale: Its global workforce allows it to serve large, complex projects.

These factors contribute to stability and help the company maintain its position in the market.

Limitations of the Moat

However, Alten’s competitive advantages have clear limitations.

• Low switching costs: Clients can replace one consulting firm with another relatively easily.

• Talent dependency: The company’s primary asset is its workforce. If employees leave, value leaves with them.

• Commoditization risk: Many consulting services are interchangeable, leading to pricing pressure.

Unlike software platforms, which can lock in customers through high switching costs and integrated ecosystems, Alten operates in a more fluid and competitive environment.

As a result, its moat is not strong enough to guarantee long-term dominance or pricing power.

Alten doesn’t have a dominant moat or software margins. But it trades at a forward P/E that most quality businesses haven’t seen in years, and the free cash flow picture is even more interesting than the earnings multiple suggests. Whether that’s a genuine opportunity or a value trap, and what I think it’s actually worth, is below for paid subscribers.

Read more

Investing isn’t about relying on the latest stock tip or trying to guess where the market is headed next. I’ve learned over the years that the real key to building wealth is much simpler: understand the business, know what you own, and put your money to work with conviction.

I want to share how I approach building a portfolio, how I decide which companies to own, how much to allocate to each, and why I focus on quality over quantity. These are lessons I’ve learned from managing real money over years, and they’ve worked whether the market is up, down, or sideways.

Own the Business, Not the Stock

The first thing I always remind myself: when you buy a stock, you’re buying a piece of a business. Not just a ticker, not just a number on a screen. That perspective changes everything.

When I look at a company, I ask questions like:

• Can it generate cash consistently, even when the economy isn’t great?

• Does it have something that keeps competitors away, a moat, a brand, or loyal customers?

• Does management make smart decisions with money?

If the answers aren’t clear, I usually move on. Understanding the business is step one. Without that, everything else, position size, portfolio allocation, doesn’t matter.

Size Positions Based on Conviction

Once I understand the business, I think about how much money to put in. This is where a lot of investors go wrong. Many spread their money across dozens of stocks, thinking it lowers risk. But over-diversifying often just waters down your best ideas.

I do it differently: position size reflects how confident I am.

• High conviction → larger allocation. These are the businesses I understand deeply and believe will compound value over many years.

• Medium conviction → moderate allocation. I like the business but I’m still watching closely.

• Low conviction → small or exploratory allocation. If I don’t understand it fully, I don’t bet much.

This keeps my portfolio focused on my best ideas.

Building a High-Conviction Portfolio

I structure my portfolio with three “layers”:

1. Core Positions (50–70%)
   These are my strongest ideas. High-quality businesses, predictable cash flow, and durable advantages. These positions I hold with the most confidence.

2. Satellite Positions (20–40%)
   Smaller positions in companies that are interesting or temporarily undervalued. They give me flexibility and optionality without risking the core.

3. Cash (0–10%)
   For small investors like me, cash is less about safety and more about opportunity. Unlike billion-dollar funds, we can take advantage of small or illiquid opportunities that they can’t. That’s why I generally keep the portfolio fully invested, holding cash for long periods means missing chances that move the needle.

Focus matters. I usually hold no more than 25–30 positions. After that, adding more stocks barely improves risk-adjusted returns. You end up with a portfolio that’s harder to track and diluted in performance. Fewer, well-understood positions let your conviction really matter.

The Margin of Safety

I never buy a business at any price. There has to be a margin of safety. That’s the buffer between what I think the business is worth and what the market is asking for.

This margin protects me if I miscalculate, if something unexpected happens, or if the market turns against me temporarily.

It’s not just about price, it’s also about quality. Businesses with strong moats, predictable earnings, and careful management give an extra layer of protection. Even if the market dips, these companies can keep generating value.

Active Management, Not Blind Holding

Conviction doesn’t mean setting it and forgetting it. I actively review my positions.

I ask:

• Has the business improved or weakened?

• Are new risks emerging?

• Are there better opportunities elsewhere that deserve more capital?

I don’t react to every market swing. But I do act when fundamentals change. Patience is key, but so is flexibility.

Lessons from Running a Portfolio

From managing real money over the years, some lessons have stuck:

1. Concentration pays off, but only if you know what you own. Large positions in your best ideas can drive long-term returns.

2. Small positions are optionality. They let you explore opportunities without risking your core.

3. Price matters as much as quality. A great business bought too high is a poor investment.

4. Patience compounds wealth. Markets are often irrational. Let the right businesses work for you over years, not days.

5. Being fully invested can be an edge. As a smaller investor, liquidity gives you access to opportunities that bigger players can’t take fully.

6. Keep the number of positions manageable. More than 25–30 stocks rarely improves results. Fewer, focused positions are easier to track and let your conviction matter.

Step-by-Step Framework

Here’s how I approach it in practice:

1. Understand the business – Know its competitive advantages, cash flow, and management quality.

2. Estimate intrinsic value – Conservative assumptions about growth and earnings.

3. Check the margin of safety – Only buy if price is meaningfully below intrinsic value.

4. Allocate based on conviction – Large for high conviction, medium for medium, small for exploratory.

5. Build the portfolio – Core positions, satellites, keep total ≤30, and stay fully invested.

6. Actively monitor and adjust – Review fundamentals, valuations, and opportunities regularly.

Final Thoughts

Investing isn’t about predicting the market, it’s about ownership, discipline, and patience. The strongest portfolios are built with fewer, well-understood businesses, fully invested in high-conviction ideas.

Small investors have an advantage: liquidity allows us to capture opportunities that big funds can’t. Use it wisely. Focus on quality, stick to your margin of safety, size positions based on conviction, and let time do its work.

Key Takeaways:

• Think of stocks as ownership in businesses, not just numbers.

• Allocate capital based on conviction.

• Build a focused portfolio: core, satellites, ≤30 positions, fully invested.

• Always maintain a margin of safety in price and quality.

• Actively monitor but be patient, let compounding work.

Investing well is simple but not easy. Stick to your best ideas, focus on quality, and let your portfolio do the work over time.

Over the past two decades, Palantir has built an impressive software platform that transforms massive datasets into actionable insights for governments and corporations. Their products, Gotham for government, Foundry for enterprise clients, and the new AIP platform for AI deployment, are gaining traction across industries.

Yet, strong business fundamentals do not automatically make a stock a good investment. Using my value investing framework, I will explain why Palantir is an excellent company but currently overvalued, and why disciplined investors should exercise patience.

Understanding Palantir’s Business

Palantir was founded in 2003 with the mission of helping organizations integrate and analyze large-scale data. Its early success came from deep government partnerships, including the CIA, Department of Defense, and other intelligence agencies. Today, Palantir is expanding aggressively into commercial applications, including healthcare, finance, industrial operations, and logistics.

This dual focus, government and enterprise, provides Palantir with two complementary revenue streams. Government contracts offer stability and predictability, while commercial growth provides scale and margin expansion as the company deploys AI solutions to more industries.

Core Platforms

1. Gotham: Designed primarily for government clients, Gotham enables intelligence, defense, and law enforcement agencies to process complex data, detect patterns, and make operational decisions. The platform is integral to national security operations, which gives Palantir a unique first-mover advantage and high switching costs.

2. Foundry: Targeted at commercial enterprises, Foundry integrates disparate data systems, provides analytics capabilities, and allows organizations to model operational outcomes. Foundry has been a key driver of Palantir’s commercial growth, especially as companies seek to leverage data for competitive advantage.

3. AIP (Artificial Intelligence Platform): Palantir’s new AI platform allows organizations to deploy AI models into operational workflows. By combining proprietary algorithms with operational data, AIP positions Palantir as a potential “operating system” for enterprise AI, which could create a long-term, sticky revenue stream if adoption continues to scale.

Palantir’s business model relies heavily on integration and customization, which creates high switching costs. Once clients embed Palantir’s software into workflows, replacing it becomes expensive and disruptive. This feature, combined with long-term government contracts, forms the backbone of a business with durable relationships and predictable revenue, which is highly attractive from a value investing perspective.

Growth and Financial Strength

Palantir’s revenue growth is impressive:

• Q4 2025 revenue: $1.41B, up 70% YoY

• 2026 guidance: ~$7.2B, up ~60%

Growth comes from both government and commercial clients, but the commercial side is accelerating faster, up 137% YoY. This suggests that Palantir is successfully scaling its enterprise offerings beyond government work, which is critical for long-term sustainability and margin expansion.

The company’s revenue profile demonstrates a rare combination for tech stocks: rapid growth with substantial recurring revenue. Contracts often span multiple years, providing visibility into cash flows and reducing execution risk relative to early-stage software firms.

Profitability

Unlike many high-growth software companies that prioritize market share over profits, Palantir is remarkably profitable:

• Gross margin: ~82%

• Operating margin: ~30%

• Free cash flow margin: ~40–50%

These metrics show that Palantir can generate substantial cash from operations. High margins allow the company to fund its own growth, reinvest in AI development, and pursue strategic initiatives without relying heavily on debt or dilutive financing. For long-term investors, this is a key factor: financial strength reduces downside risk.

Balance Sheet

Palantir has a robust balance sheet:

• Cash: ~$7.2B

• Debt: minimal

This provides flexibility to pursue new opportunities, invest in R&D, and withstand macroeconomic shocks without compromising the business. From a value investor’s standpoint, a strong balance sheet is a margin-of-safety factor, it ensures the company can endure downturns while still funding growth initiatives. However, there are not significant tangible assets

Durable Competitive Advantage

Palantir has several features that constitute a durable competitive advantage, or moat:

1. High switching costs: Once clients integrate Palantir into operational workflows, moving to a competitor is expensive and disruptive. This protects revenue and creates long-term client retention.

2. Government relationships: Multi-year contracts with intelligence and defense agencies provide stable revenue streams. These relationships also offer credibility when Palantir approaches commercial enterprises.

3. Proprietary AI and integration: Palantir’s AIP platform positions the company as a central hub for enterprise AI, which could strengthen client dependency and create an ecosystem effect.

4. Early mover advantage: Being in the market since 2003, Palantir has accumulated experience, data expertise, and a network of government and commercial clients that competitors cannot easily replicate.

These advantages suggest Palantir is likely to maintain its profitability and revenue growth for many years, which is exactly what long-term investors seek. However, as a value investor, it is critical to compare these strengths to the price being paid.

Palantir has 82% gross margins, $7.2B in cash, and AI contracts that competitors can’t easily replicate. On every business quality metric, it’s exceptional. But the valuation tells a completely different story, and the math on expected returns at today’s price is harder to ignore than most bulls want to admit. The full breakdown, my fair value estimate, and what price I’d actually consider buying are below for paid subscribers.

Read more

Over time, I’ve developed a simple framework that helps me evaluate companies and make long-term investment decisions. It’s not about chasing trends or predicting the market. It’s about finding strong businesses and buying them at reasonable prices.

My approach is heavily inspired by legendary investors like Warren Buffett and Benjamin Graham. Both emphasized the importance of treating stocks as ownership in real businesses rather than pieces of paper to trade.

Their ideas around patience, discipline, and rational decision-making form the foundation of how I invest today.

My investing framework

When evaluating a company, I focus on a few key principles.

1. Look for undervalued businesses

A great business can still be a poor investment if the price is too high. I look for situations where the market price is lower than what I believe the business is actually worth.

This disconnect between price and value creates opportunity.

2. Strong balance sheet

Financial strength matters. I prefer companies with manageable debt, consistent cash flow, and the ability to fund their own growth.

Businesses that rely heavily on borrowing or constant capital raises are often more fragile during difficult economic periods.

3. Durable competitive advantage

The best businesses have something that protects them from competition. This is often called a moat.

Examples include strong brands, network effects, high switching costs, or proprietary technology. A durable moat allows companies to maintain profitability over long periods of time.

4. Margin of safety

One of the most important principles in value investing is the concept of a margin of safety, originally popularized by Benjamin Graham.

Instead of paying full estimated value, I prefer buying at a meaningful discount. This creates a buffer against mistakes, unexpected risks, or changes in the business environment.

What readers will get

This newsletter is where I share how I apply this framework in practice.

Every two weeks, I will share:

• Undervalued stock ideas
• Detailed valuation breakdowns
• Investing lessons from real companies
• Insights on long-term investing

My goal isn’t to provide quick stock tips. Instead, I want to build a community focused on thoughtful investing and long-term decision making.

If you’re interested in learning how to evaluate businesses and think like a long-term investor, you’re in the right place.

Disclosure

This newsletter is published for educational and informational purposes only. Nothing written here constitutes financial advice, investment advice, or a recommendation to buy or sell any security.

I am not a licensed financial advisor, investment advisor, broker, or dealer. All analysis reflects my personal research, opinions, and framework as an individual investor. It may contain errors, omissions, or outdated information, and should not be relied upon as the basis for any investment decision.

Investing involves risk, including the possible loss of principal. Past performance, of any security, strategy, or analytical approach discussed here, is not indicative of future results. Markets are unpredictable, and even well-researched theses can and do go wrong.

I may personally hold positions in securities mentioned in this newsletter, either long or short, at the time of publication or at any point thereafter, without obligation to disclose changes. My interests may not align with yours. Always conduct your own independent research and consider your own financial situation, objectives, and risk tolerance before making any investment decisions. Consult a qualified financial professional if you need personalized advice.

This newsletter is not affiliated with, endorsed by, or associated with any company or security mentioned herein.

The CagriSema Trial Result: Strategic Implications

The 84-week REDEFINE 4 trial compared CagriSema (a fixed-dose combination of cagrilintide 2.4 mg and semaglutide 2.4 mg) against tirzepatide 15 mg. Patients on CagriSema achieved 20.2% weight loss versus 23.6% with tirzepatide. The study failed to meet its primary endpoint of demonstrating non-inferiority.

This outcome is significant for several reasons.

First, CagriSema was positioned as a key next-generation therapy designed to extend Novo Nordisk’s leadership in obesity beyond its current semaglutide franchise. Expectations were that the combination therapy would at least match, if not exceed, tirzepatide’s efficacy profile. The failure to demonstrate non-inferiority challenges that assumption.

Second, this result alters competitive dynamics. Obesity pharmacotherapy is increasingly viewed as a two-horse race between Novo Nordisk and Eli Lilly. In that context, efficacy differences matter. If tirzepatide maintains a consistent weight-loss advantage, prescriber behavior and payer negotiations may gradually shift.

Third, the company’s plan to initiate a higher-dose trial introduces additional uncertainty. While dose optimization may improve outcomes, it extends timelines and carries regulatory and safety considerations. The pathway to restoring perceived parity is no longer straightforward.

The pipeline remains active, but the probability distribution of outcomes has widened.

2026 Guidance: From Hypergrowth to Normalization

Compounding the trial miss, Novo Nordisk now expects a 5–13% decline in adjusted sales and profit in 2026. This marks a clear shift from prior years characterized by exceptional growth driven by Ozempic and Wegovy.

Several factors are contributing:

• Continued heavy investment in manufacturing capacity

• Competitive intensity in obesity and diabetes

• Pricing pressures across key markets

• Mix effects within the GLP-1 portfolio

While capacity investments are strategically rational given long-term demand, near-term financial compression changes the narrative from “structural hypergrowth” to “growth normalization under competitive pressure.”

Importantly, this is not a collapse in demand. The obesity and diabetes markets remain large and structurally expanding. However, growth rates appear more cyclical and competitive than previously assumed.

The market’s reaction suggests that investors are recalibrating expectations for both near-term earnings power and long-term terminal growth.

What Remains Intact

Despite the sharp share price reaction, several core elements of the thesis remain intact.

The semaglutide franchise continues to generate substantial cash flow. Wegovy maintains strong global demand, and the European Commission has approved a higher, more effective dose, reinforcing its relevance. Additionally, the oral version of Wegovy is launching with encouraging initial uptake, offering an alternative growth vector.

Novo Nordisk retains:

• Global distribution scale

• Manufacturing capabilities

• Deep regulatory experience

• Strong balance sheet and free cash flow generation

The company remains a central participant in a massive and underpenetrated therapeutic category.

However, the competitive moat is no longer expanding uncontested. It is now being actively challenged.

After the CagriSema miss and 2026 guidance cut, I’ve revised my assumptions across the board, lower market share, faster pricing normalization, higher risk premium. The updated intrinsic value range, what I think the market is still getting wrong, and whether I’m holding or reducing my position are all below for paid subscribers.

Read more

The mistake I keep repeating is selling great businesses once they look “fairly valued.”

As a value investor, I’m trained to look for discounts. I want a margin of safety. I want to buy below intrinsic value.

But once a stock approaches my estimate of fair value, I get uncomfortable.

If I bought it at $50 and I think it’s worth $80, what do I do when it hits $78?

Too often, I sell.

Not because the business is worse.
Not because management disappointed.
Not because the long-term opportunity shrank.

But because it no longer looks “cheap.”

The trigger is mental anchoring. I anchor to my original valuation model. I treat intrinsic value like a fixed number instead of a moving target. When price converges with my spreadsheet, I feel like the job is done.

But value investing is not about flipping discounts into fair value. It’s about owning businesses that can increase intrinsic value over time.

And that’s where I keep falling short.

A Real Example

Several years ago, I built a position in a high-quality consumer business trading at what I believed was a 30% discount to intrinsic value.

Strong brand. Consistent cash flow. High returns on capital. Rational capital allocation.

I estimated intrinsic value at roughly $90 per share and bought in the low $60s.

Within two years, the stock traded up to the mid-$80s. My model said it was close to fair value.

So I exited most of the position.

What I underestimated was not the business — it was its ability to keep compounding.

Revenue continued to grow mid-single digits. Margins expanded modestly. Free cash flow increased. Management continued repurchasing shares.

Five years later, intrinsic value was no longer $90. It was meaningfully higher. The stock had more than doubled from my exit price.

I made a good return.

But I missed a great one.

The mistake wasn’t buying wrong. It was thinking the opportunity ended once price met my initial estimate.

Why It Happens

This mistake has deep roots in traditional value training.

First, anchoring bias.

When I calculate intrinsic value, I treat it as precise. But valuation is a range, not a point estimate. Businesses evolve. Competitive positions strengthen. Capital allocation decisions change outcomes.

By anchoring to a static number, I ignore the dynamic nature of compounding.

Second, an outdated definition of risk.

Classic value thinking often equates higher multiples with higher risk. But a high-quality business reinvesting at high returns on capital can justify a higher multiple for a long time.

The real risk is not overpaying slightly for quality. The real risk is exiting a long runway too early.

Third, the psychological comfort of “discipline.”

Selling at fair value feels prudent. It feels like sticking to the rules. There’s a sense of closure: buy at a discount, sell at fair value, move on.

But the market doesn’t reward closure. It rewards endurance.

This is where I think about Warren Buffett.

Early in his career, Buffett followed a strict cigar-butt approach — buying statistically cheap companies and selling once they reverted to fair value.

But over time, influenced by Charlie Munger, he evolved. Instead of selling great businesses when they reached fair value, he held them as they continued compounding intrinsic value.

Look at Coca-Cola. Berkshire did not sell once it reached “fair value.” They held as earnings grew and intrinsic value expanded year after year.

The lesson wasn’t to ignore valuation. It was to recognize that intrinsic value can grow faster than you think — especially in high-quality businesses.

I understand that intellectually.

But I don’t always execute it.

Structural Fix

To correct this, I’m changing my framework.

First, I now separate “mean reversion plays” from “compounders.” If I buy something purely because it’s cheap relative to assets or earnings, I will sell near fair value. That’s the strategy.

But if I buy a high-return-on-capital business with reinvestment runway, the default is to hold — even when it looks fairly valued.

Second, I now model intrinsic value as a range and update it annually, not just at purchase. If the business continues to grow, the valuation target must grow with it.

Third, I’ve introduced a simple question before selling: Is the business likely to be materially more valuable in five years? If the answer is yes, I require a stronger reason to exit.

Value investing is not about extracting one-time discounts.

It’s about owning assets that grow in value over time.

I’m still learning that lesson.

Disclosure

This newsletter is published for educational and informational purposes only. Nothing written here constitutes financial advice, investment advice, or a recommendation to buy or sell any security.

I am not a licensed financial advisor, investment advisor, broker, or dealer. All analysis reflects my personal research, opinions, and framework as an individual investor. It may contain errors, omissions, or outdated information, and should not be relied upon as the basis for any investment decision.

Investing involves risk, including the possible loss of principal. Past performance, of any security, strategy, or analytical approach discussed here, is not indicative of future results. Markets are unpredictable, and even well-researched theses can and do go wrong.

I may personally hold positions in securities mentioned in this newsletter, either long or short, at the time of publication or at any point thereafter, without obligation to disclose changes. My interests may not align with yours. Always conduct your own independent research and consider your own financial situation, objectives, and risk tolerance before making any investment decisions. Consult a qualified financial professional if you need personalized advice.

This newsletter is not affiliated with, endorsed by, or associated with any company or security mentioned herein.

Revenue comes primarily from athletic footwear, apparel, and direct-to-consumer sales through its retail stores and online platform. Globally, it’s the market leader in athletic footwear, with strong brand recognition in North America, Europe, and emerging markets.

On paper, Nike looks attractive due to brand strength, product innovation, and consistent free cash flow. But several assumptions make me uncertain, particularly around consumer trends, supply chain pressures, and economic cycles.

Key Assumptions

• Nike maintains strong brand relevance and consumer demand in core and emerging markets.

• Direct-to-consumer (DTC) sales continue to grow profitably, offsetting pressures from wholesale channels.

• Supply chain disruptions, including shipping and labor issues, do not materially affect product availability or costs.

• Economic conditions in major markets allow consumers to continue discretionary spending on premium athletic products.

These assumptions must hold for the thesis to work.

Biggest Risks / What Could Go Wrong

• Shifts in consumer preferences toward competitors or unbranded products could reduce market share.

• Rising labor, material, or logistics costs compress margins and earnings.

• Economic downturns in key markets lower discretionary spending on apparel and footwear.

• Regulatory or geopolitical risks in emerging markets could affect sales or distribution.

I acknowledge that if any of these occur, my thesis may be invalid.

Nike trades at a discount to its own history and below peers on P/S. That should make it interesting. But there are three specific reasons I think the thesis could still be wrong, and one number in the valuation that keeps me up at night. Both are below for paid subscribers.

Read more

Most stocks are not worth analyzing.

Not because they’re bad businesses, but because time and attention are limited.

Before I build a valuation model or write a thesis, every company has to pass a basic filter.

My initial filter

• Can I understand how this business makes money in 10 minutes?

• Is the business likely to exist in 10 years?

• Are revenues and margins relatively stable?

• Is there a clear reason the stock might be mispriced?

• Can I reasonably estimate intrinsic value?

Reasons I usually stop

• Too much leverage

• Business depends on one unpredictable variable

• Accounting I don’t trust

• Valuation requires heroic assumptions

• Story is more exciting than the numbers

What happens if a stock passes

Only after a company passes this filter do I spend time on a full valuation and downside analysis.

Most ideas never make it past this stage.

The ones that do become full write-ups, with assumptions and updates when facts change.

Deeper analyses and updates live in the archive.

Over time, this becomes a public record of how I think, including when I’m wrong.

Disclosure

This newsletter is published for educational and informational purposes only. Nothing written here constitutes financial advice, investment advice, or a recommendation to buy or sell any security.

I am not a licensed financial advisor, investment advisor, broker, or dealer. All analysis reflects my personal research, opinions, and framework as an individual investor. It may contain errors, omissions, or outdated information, and should not be relied upon as the basis for any investment decision.

Investing involves risk, including the possible loss of principal. Past performance, of any security, strategy, or analytical approach discussed here, is not indicative of future results. Markets are unpredictable, and even well-researched theses can and do go wrong.

I may personally hold positions in securities mentioned in this newsletter, either long or short, at the time of publication or at any point thereafter, without obligation to disclose changes. My interests may not align with yours. Always conduct your own independent research and consider your own financial situation, objectives, and risk tolerance before making any investment decisions. Consult a qualified financial professional if you need personalized advice.

This newsletter is not affiliated with, endorsed by, or associated with any company or security mentioned herein.

Novo Nordisk A/S is a global Danish pharmaceutical company focused on diabetes care, obesity treatment, and related cardiometabolic diseases. It manufactures and markets products worldwide, with its most notable offerings being semaglutide-based therapies:

• Ozempic/Rybelsus for type 2 diabetes

• Wegovy for obesity

These drugs have driven significant recent growth, while the company also produces insulins and other hormones and maintains a presence in rare diseases and other therapy areas. NVO sells in over 160 countries and remains the world’s largest insulin maker.

Investment Thesis

1. Dominant Position in Diabetes & Obesity (Massive Addressable Market)
Novo Nordisk benefits from secular tailwinds in global diabetes and obesity trends. With diabetes affecting hundreds of millions worldwide and obesity rates rising, its core markets are structurally expanding. Semaglutide products have become defining therapies in both segments, unlocking high growth and premium pricing.

2. Innovation & Pipeline Depth
Beyond existing blockbusters, NVO is advancing next-gen therapies such as CagriSema and oral GLP-1 formulations for obesity/diabetes, alongside a diversified pipeline in rare diseases and cardiometabolic conditions. Collaborations with AI partners and acquisitions expand therapeutic breadth and long-term prospects.

3. Structural Competitive Moat
The semaglutide platform and brand equity (Ozempic/Wegovy) confer a significant moat versus rivals. Manufacturing investments and broad global distribution also support supply resilience and patient access.

4. Financial Strength & Cash Flow
Novo Nordisk generates robust free cash flow, enabling R&D investment and shareholder returns. Its profitability metrics remain strong compared to big pharma peers.

Novo Nordisk has fallen over 50% from its peak. Below, I walk through my DCF estimates, the revised intrinsic value range after the CagriSema miss, and whether I think the market is now overreacting. That’s for paid subscribers.

Read more