There is a machine that costs roughly four hundred million dollars. It is among the most complex objects ever manufactured by human beings, the product of tens of thousands of precision components, and it works by generating light of an exotic wavelength by blasting droplets of molten tin with a laser tens of thousands of times per second, then focusing that light with the flattest mirrors ever made to etch features into silicon a few atoms wide. Without this machine, the most advanced computer chips on Earth, the ones inside the AI revolution, the latest phones, the fastest processors, simply cannot be made.
Exactly one company in the world can build it. Not the best at building it. The only one. Its name is ASML, it sits in the small Dutch town of Veldhoven, and it holds something almost unheard of in modern business: a true, complete, hundred-percent monopoly on the single most important tool in the most important industry of the age.
This is the deepest chokepoint in the modern economy, and the story of how a scrappy Dutch underdog, born in a leaky shed and nearly bankrupt more than once, came to occupy it is one of the most extraordinary tales in business. Because ASML did not win this position by being clever or lucky. It won it by attempting, and sustaining for two decades, a technical bet so punishing and so uncertain that every rival on Earth eventually quit.
The World Before ASML
To understand what ASML achieved, you have to understand the central problem of making chips: how do you draw something almost unimaginably small?
A computer chip is made by a process called lithography: shining light through a patterned stencil onto a light-sensitive silicon wafer, etching the circuit pattern into the surface. The smaller the features you can etch, the more transistors you can pack onto a chip, and the more powerful it becomes. This relentless shrinking is the engine of the entire digital age. But there is a hard physical limit: you cannot etch a feature much smaller than the wavelength of the light you are using. For decades the industry used progressively shorter wavelengths of ultraviolet light, but everyone could see the wall coming. To keep shrinking chips, the industry would eventually need light of a far, far shorter wavelength, in the “extreme ultraviolet” range, about fourteen times shorter than what was then in use.
The problem was that extreme ultraviolet light is monstrously difficult to work with. It is absorbed by almost everything, including air and ordinary glass lenses, so it must be generated and steered in a vacuum, using mirrors of impossible perfection rather than lenses. For years, harnessing EUV to industrialize chipmaking was considered by many to be somewhere between absurdly hard and physically impossible. It was the kind of problem that could consume decades and billions of dollars and still fail. And that is precisely the problem ASML decided to stake its future on.
The Founder
ASML does not have a single charismatic founder; it has a humble, almost improbable origin. It was created in 1984 as a joint venture between the Dutch electronics giant Philips and a smaller equipment maker, ASM International, and christened ASM Lithography. It began, by the company’s own lore, in a leaky shed next to a Philips building in the Netherlands, with a handful of employees and modest ambitions.
And it began as an underdog in the worst sense. The lithography market in the 1980s was dominated by the Japanese giants Nikon and Canon, whose machines were the industry standard. ASML was a tiny, unknown European challenger with no commercially proven product, and many observers considered it a likely failure, its employees handed, as one account put it, a ticket to nowhere. The early years were genuinely precarious. The company nearly went bankrupt more than once, surviving at one point in 1992 only through an emergency loan from Philips of a few tens of millions of guilders. This was not a company that looked destined for greatness. It looked destined for the corporate graveyard.
What it had, instead of dominance, was a particular strategy and a particular kind of patience. Rather than dictating technology to its customers, ASML chose to co-develop it with them, working hand in hand with chipmakers to build the next generation collaboratively. This customer intimacy would, decades later, become a crucial part of its moat. But first, the underdog had to make a bet that would either destroy it or make it immortal.
Decision Point — the EUV moonshot
It is the 1990s. You can keep making steady, incremental improvements to existing lithography, the safe path your Japanese rivals are taking. Or you can bet the company on extreme ultraviolet light, a technology many believe is physically impossible to industrialize, that will take twenty years and billions of euros and may simply never work.
What do you do?
A) Take the safe, incremental path your competitors are taking.
B) Commit decades and billions to the EUV moonshot, and endure.
C) Hedge, dabbling in EUV without fully committing.
ASML’s rivals effectively chose A and C, and eventually abandoned EUV as too hard. ASML chose B and held it for twenty years. Their surrender is exactly what left ASML the only company on Earth that could build the machine. (This is a thought experiment, not investment advice.)
The Near-Death and the Moonshot
ASML’s defining act, the thing that makes it legendary, was the EUV bet, and it is one of the longest, most expensive, and most uncertain gambles in the history of business.
Beginning seriously in the 1990s, ASML committed to making extreme ultraviolet lithography work, to industrializing a technology that many believed could not be industrialized at all. The development stretched across roughly two decades. The early prototypes were, by all accounts, catastrophes: machines plagued by constant breakdowns, unable to maintain the speeds needed for real production, solving problems no one had ever solved because no one had ever before tried to harness EUV light for manufacturing. Through all of it, ASML poured a staggering share of its revenue into research, often fifteen to twenty percent, year after year, accumulating well over six billion euros in EUV research investment over the life of the program.
And here is the decisive fact, the one that explains everything that followed: the competitors quit. The problem was so hard, so expensive, and so slow to pay off that ASML’s rivals, including the once-dominant Japanese giants, abandoned the EUV race. They made the rational short-term decision that the bet was not worth it. ASML made the opposite decision and held to it for twenty years. By the time EUV was finally production-ready, around 2017, ASML was the only company on Earth that had stayed in the game long enough to win it. There was no second place. The rivals had gone home.
This is the rarest kind of near-death story among legendary companies. ASML’s danger was not a market crash or a single crisis; it was the slow, grinding, decades-long risk that its moonshot would simply never work, that it would pour billions and twenty years into a technology that failed, and die. It survived not by avoiding the risk but by having the conviction and the patience to endure it longer than anyone else was willing to. The moat was the willingness to suffer the bet.
The Inflection: From Supplier to Chokepoint
When EUV finally worked, ASML’s position transformed from “leading supplier” to something far rarer and more powerful: an absolute chokepoint.
Because EUV is required to make the most advanced chips, and because ASML is the only company that can make EUV machines, every manufacturer that wants to build leading-edge chips, TSMC, Samsung, Intel, all of them, must buy their machines from ASML. There is no alternative. Not a more expensive alternative, not an inferior alternative, no alternative at all. The most advanced chip factories on Earth are, in a precise sense, assemblies of ASML machines. The company sits beneath even TSMC: if TSMC is the indispensable factory, ASML is the indispensable supplier to the indispensable factory.
And ASML kept extending the lead rather than resting on it. It strengthened the system by acquiring critical pieces of the puzzle, the light-source company Cymer and the metrology firm Hermes Microvision, building what it calls a “holistic lithography” capability that ties the whole ecosystem together. And as the industry pushed toward ever-smaller chips, ASML introduced the next generation, High-NA EUV, machines costing on the order of four hundred million dollars each, first shipped to Intel and TSMC in the mid-2020s, ensuring that the monopoly extends into the next era of chipmaking. It is, in the truest sense, extending its monopoly rather than defending it.
The Moat
ASML’s moat may be one of the single deepest anywhere, and it is worth being precise about why a rival cannot simply build a competing EUV machine.
First, there is the technology itself: two decades of accumulated knowledge in industrializing the hardest optical process ever attempted, knowledge that exists in no manual and was earned only through twenty years of failure and iteration. Second, there is the ecosystem: ASML does not make every part of the machine itself; it sits atop a vast, deeply integrated network of thousands of specialized suppliers (for the mirrors, the light sources, the precision components) that took decades to assemble and that no newcomer could replicate. Third, there is the capital and time: even a rival with unlimited money would need many years and many billions, with no guarantee of success, to attempt what ASML has already done, all while ASML moves further ahead. And fourth, there is the customer co-development: ASML’s machines are built in deep partnership with the handful of chipmakers who use them, a collaborative intimacy a newcomer cannot conjure.
Stack these together and you get something almost unique in modern business: not a wide moat, but a complete one. A literal monopoly, protected by technology, ecosystem, capital, time, and relationships all at once, on a product the entire world economy depends on. There is a reason no one has built a second EUV machine. It may be the hardest-to-replicate competitive position on Earth.
The one genuine vulnerability is not competition but politics and cyclicality: ASML sits at the center of the US-China technology conflict (its machines are subject to export controls), its customer base is highly concentrated, and the semiconductor industry is famously cyclical. But the long order backlogs, often eighteen to twenty-four months, give it a visibility into future revenue that most companies can only envy.
The Wealth Created
ASML’s investor story carries the most extreme version of a lesson that runs through every legendary company: that the deepest value can hide in the unglamorous infrastructure beneath the famous names, and that the longest, hardest bets can produce the most durable rewards.
An investor who understood, somewhere along ASML’s long climb, that this obscure Dutch equipment maker was quietly becoming the sole gatekeeper of all advanced chipmaking, and who held through the cycles, would have owned one of the great compounding franchises of the modern technology era. Its revenue climbed from the low tens of billions to far higher as EUV ramped, its market value grew into the hundreds of billions, making it for a time Europe’s most valuable technology company, and its monopoly economics, the ability to charge enormous prices for a product with no substitute, produced the kind of margins and pricing power that most companies can only dream of.
The lesson, in its purest form, is this: the most powerful position in any value chain is the irreplaceable chokepoint, the single point everyone must pass through. ASML found, and then defended, the ultimate chokepoint of the digital age. And it got there not through a flashy product or a brilliant marketing campaign, but through the least glamorous virtue imaginable: the patience and conviction to keep funding an impossibly hard bet for twenty years, long after everyone else had quit. The reward for being the last one standing on the hardest problem was a monopoly on the future.
What Everyone Got Wrong
Mistake #1: Writing off the underdog.
Reality: ASML emerged from Philips as a likely failure, a tiny European challenger to Japanese giants, nearly bankrupt more than once. The market that dismissed it missed the company that would become the single most indispensable supplier in technology.
Mistake #2: Believing the EUV bet was a money pit.
Reality: for two decades EUV looked like a bottomless investment that might never pay off, which is exactly why competitors quit. The willingness to keep funding it was not folly; it was the moat being built, one impossible year at a time.
Mistake #3: Assuming a monopoly that lucrative would attract competition.
Reality: normally enormous profits invite rivals. But the barrier, two decades of compounded technical knowledge, an irreplaceable supplier ecosystem, and astronomical capital requirements, is so high that the monopoly has only deepened. Some moats are too wide to cross even when the prize is this large.
The Alternative Timeline
A counterfactual, clearly hypothetical.
Picture the world where ASML, around the turn of the millennium, makes the same rational decision its competitors made and abandons the EUV moonshot as too hard, too expensive, and too uncertain.
In that timeline, EUV lithography arrives years later, or in some halting, fragmented form, because no single company is willing to absorb the decades of pain required to industrialize it. The relentless shrinking of chips stalls against the physical limits of older lithography. The most advanced chips of the 2020s, the ones powering the AI revolution, are slower to arrive or never arrive in their current form, because the machine that prints them was never perfected. The entire timeline of computing slows, and the AI boom, which runs on chips that can only be made with EUV, is delayed or diminished. Or, in another version, a different company eventually cracks it and occupies the chokepoint ASML now holds, capturing one of the great monopolies in history.
It did not happen that way, because one stubborn Dutch company decided to keep attempting the impossible for twenty years after everyone else gave up. The lesson is the most counterintuitive of all: the very difficulty that made everyone else quit was the source of the prize. The harder and longer and more uncertain the bet, the fewer competitors will endure it, and the more complete the monopoly for the one who does. ASML’s moat is made of other people’s surrender.
Why This Matters to Investors
The Greatest Companies Thesis
Every legendary company begins with an idea that looks improbable.
Every one survives a stretch where failure looks inevitable.
Every one eventually reaches a point where success looks obvious.
The opportunity exists only in the space between the second and third.
ASML is perhaps the purest embodiment of the thesis anywhere. The improbable idea: industrializing extreme ultraviolet light, a technology many believed was physically impossible to make work. The stretch where failure looked inevitable: two decades of breakdowns, near-bankruptcies, and a moonshot so punishing that every rival abandoned it. And the point where success looks obvious: a hundred-percent monopoly on the most important machine in the world economy. The opportunity lived, exactly as the thesis demands, in the long, dark middle, in the twenty years when EUV looked like it might be a catastrophic mistake rather than the foundation of a monopoly.
The reason to study ASML is that it shows the asymmetry that makes legendary companies legendary in its most extreme and instructive form. The payoff, a complete monopoly on a critical chokepoint, was so large precisely because the bet was so hard and so long that almost no one was willing to make it. The greatest opportunities almost always looked terrible, or impossible, before they looked inevitable, and ASML’s EUV gamble looked, for twenty years, like the most expensive way imaginable to go bankrupt. The deepest rewards consistently go to those with the conviction to endure the longest, darkest middle, and ASML endured one of the longest of them all.
Lessons in Order of Depth
On the surface — the move
Own the chokepoint. ASML does not compete for share in a crowded market; it is the single point the entire industry must pass through. The trader’s analogue is seeking the irreplaceable position in a value chain, the one node that captures value no matter who else wins around it.
Below the surface — the Money
Endure the long, expensive bet that others abandon. ASML spent twenty years and billions on a technology that might have failed, funding it through fifteen-to-twenty-percent-of-revenue R&D when rivals quit. The investor’s version is the willingness to make and hold a high-conviction position through a long, uncomfortable period when it is not yet working and everyone else has given up.
Below that — the Mind
The difficulty is the moat. ASML’s deepest insight, lived rather than spoken, is that the very hardness of the problem is what protects the prize, because hardness is what makes competitors quit. The trader’s parallel is the recognition that the most durable edges are usually the uncomfortable, difficult, patience-demanding ones, precisely because comfort and ease attract crowds that compete the edge away.
At the deepest level — the question
ASML’s monopoly was built from other people’s surrender, from being the last one willing to endure an impossible problem. So the deepest question ASML poses is about the relationship between difficulty and reward itself: are you willing to attempt the thing that is so hard, so long, and so uncertain that everyone else quits, knowing that their quitting is exactly what would make your success unassailable? Most people, and most companies, optimize for the achievable, the bet that pays off soon enough to be comfortable. ASML’s example suggests that the very largest and most durable rewards are reserved for those willing to suffer the bets that are almost too hard to bear, because the difficulty itself clears the field. The deepest edge is not intelligence or capital but the rarest form of endurance: the willingness to keep going, alone, on the hardest problem, for as long as it takes, after everyone else has gone home.
The Legendary Scorecard
| Category | Score | Notes |
|---|---|---|
| Founder Vision | 9 / 10 | Saw EUV as the future when most thought it impossible |
| Innovation | 10 / 10 | Industrialized the hardest optical process ever attempted |
| Execution | 10 / 10 | Twenty years of endurance through breakdowns and near-bankruptcy |
| Moat | 10 / 10 | A literal 100% monopoly; perhaps the deepest moat anywhere |
| Capital Allocation | 9 / 10 | 15–20% of revenue into R&D for decades; strategic acquisitions |
| Wealth Creation | 9 / 10 | From a shed to Europe’s most valuable tech company |
| Durability | 9 / 10 | Near-unassailable; risks are geopolitical and cyclical |
| Historical Importance | 10 / 10 | The chokepoint that makes all advanced chips possible |
| Overall Legendary | 9.5 / 10 | The monopoly built from other people’s surrender |
Scores are an editorial verdict on the standard eight-category scale used across the Greatest Companies series. The overall is a judgment, not a weighted average.
Company Timeline
- 1984 — Founded (April 1) as ASM Lithography, a Philips/ASM International joint venture in Veldhoven; starts in a shed
- 1980s — Scrappy underdog to the Japanese giants Nikon and Canon
- 1992 — Nearly bankrupt; rescued by a ~36-million-guilder Philips emergency loan
- 1996 — IPO on NASDAQ and Euronext Amsterdam
- 1998 — First major production tool; becomes independent of Philips
- 2000s — TWINSCAN dual-stage system and immersion lithography
- 1990s–2010s — The ~20-year, €6B+ EUV moonshot; Nikon and Canon drop out
- 2017 — Production-ready EUV
- 2023 — First High-NA EUV shipped to Intel (Dec)
- 2024–2025 — High-NA to TSMC; sole EUV supplier; market cap ~$345B
Key Numbers
| Founded | 1984 (Veldhoven, Netherlands) |
| Origin | Philips / ASM International joint venture |
| The bet | ~20-year, €6B+ EUV moonshot |
| Market position | 100% monopoly on EUV machines |
| Machine cost | ~$370–400M per High-NA EUV unit |
| Customers | TSMC, Samsung, Intel, SK Hynix, Micron |
| Market value (2025) | ~$345B (Europe’s largest tech company) |
Related Reading
More Greatest Companies
- TSMC: The Most Important Company You Have Never Thought About (the foundry that depends on ASML’s machines)
- Nvidia: The 90% Drawdown That Built the AI Age (whose chips exist only because EUV does)
- Apple: Ninety Days From Death to the World’s Most Valuable Company (at the end of the chain ASML begins)
Lesson Hubs
- Competitive Moats (the deepest moat of all: a complete monopoly)
- Innovation Cycles (the twenty-year R&D bet that rivals abandoned)
Across the Library
- The AI Boom (Market Mayhem — the demand wave ASML sits at the very bottom of)
This article is part of the Greatest Companies series, and adapted from the forthcoming book on how the greatest companies were built. Explore the framework in The Complete Trader’s Edge.
Frequently Asked Questions
What does ASML make?
ASML makes photolithography machines, the equipment used to print circuit patterns onto silicon wafers. It is the only company in the world that makes extreme ultraviolet (EUV) lithography machines, which are required to manufacture the most advanced computer chips.
Why is ASML a monopoly?
ASML spent roughly two decades and over €6 billion industrializing EUV lithography, a problem so hard that its competitors, including the Japanese giants Nikon and Canon, eventually abandoned it. By the time EUV was production-ready, ASML was the only company left that could build the machines, giving it a 100% monopoly.
How does ASML relate to TSMC?
ASML sits one layer beneath TSMC. TSMC manufactures the world’s most advanced chips, but it can only do so using ASML’s EUV machines. If TSMC is the indispensable factory, ASML is the indispensable supplier to that factory, and the same is true for Samsung and Intel.
Why can’t a competitor just build an EUV machine?
Because the barrier is enormous: two decades of accumulated technical knowledge that exists in no manual, a supplier ecosystem of thousands of specialized partners built over many years, astronomical capital requirements, and deep co-development relationships with chipmakers. Even with unlimited money, catching ASML would take many years with no guarantee of success.
What are the main risks to ASML?
Not competition, but geopolitics and cyclicality. ASML’s machines are subject to US-China export controls, its customer base is highly concentrated among a few chipmakers, and the semiconductor industry is cyclical. Its long order backlogs, often 18 to 24 months, give it unusual visibility into future revenue.
