ASML, analysed ten ways (10 PILLARS ANALYSIS) One company, one day, ten methods — written the day after its third-quarter report

1 Full stock analysis

What this method is for: to hold the whole company in view at once, so that no single strong number — a growth rate, a multiple, a chart — gets to decide the question by itself.

The business, and where the money comes from

Machines and the service of machines. ASML sells lithography systems to chipmakers and then maintains, upgrades and supports the ones already installed in their factories. Our data holds the two as one revenue line, so the split between a one-off machine sale and the recurring service income is not on this page (see the box below). What the figures do show is the scale and the shape, in the euros the company reports in: trailing twelve-month revenue of €39.47bn against €32.21bn in the prior twelve months, +22.54%; gross margin 54.34%, operating margin 38.55%, net margin 34.14%. A manufacturer of industrial machinery that keeps a third of its revenue as profit is not an ordinary manufacturer.

Competitive position, in one paragraph

The moat is technology, and at the leading edge it is complete: nobody else makes an EUV machine, so a chipmaker that wants to print the finest circuits buys from ASML or does not print them. Around that sits a switching cost of the heaviest kind — a chip factory's processes are built and qualified around its tools, and the tools are serviced by their maker for as long as they run. Scored on tab 3. What the moat does not protect is the timing of demand: it decides who gets the order, not when the customer places it. That is why the cycle, not a competitor, ranks first on tab 5.

Industry, at the level that matters here

Semiconductor equipment is a capital-spending business: its revenue is other companies' investment budgets. When chipmakers build, it grows quickly; when they pause, it does not, whatever the quality of the machine. Our own three years show both halves. In 2024 revenue grew 2.56% and net income fell 3.41%; in the third quarter of 2026 revenue grew 55.16% and net income 133.49% against the same quarter a year earlier. The industry argument of the moment is whether demand for the most advanced chips has made that cycle shallower. The figures on this page cannot settle it, because the number that would — the order book — is not in our data.

Financial health

Improving on every line we hold, from a soft year. Operating margin was 32.81% in 2023, 31.92% in 2024, 34.60% in 2025, 38.55% over the trailing twelve months and 44.36% in the third quarter of 2026 alone. Gross margin over the same steps: 50.21%, 49.82%, 52.83%, 54.34%, 57.45%. A business whose margins rise this fast with its revenue is one whose costs are largely fixed — which is also what makes the earnings fall faster than revenue when the cycle turns. The numbers, year by year, are on tab 2. What we cannot assess is the balance sheet.

The key risks, in one line each

• The cycle. Revenue is chipmakers' spending, and 2024 — revenue +2.56%, net income −3.41% — is the last soft year in our table. Ranked and argued on tab 5.
• Export rules and China. Governments decide which machines may be sold where; our data has no geographic split, so this page cannot size the exposure.
• A few large customers. Only a handful of companies can afford a leading-edge factory; our data does not say how much of revenue each one is.
• The multiple. 41.1 times trailing dollar earnings, a beta of 1.302, and the share already 17.50% below its June high while earnings rose.
• Currency. The earnings are in euros and the share is in dollars; a holder in dollars owns a euro profit stream.

Valuation, against the only comparison we can make honestly

The provider's computed P/E for this company is about 47. It is wrong, and not because of the provider's arithmetic: it divides a dollar price by euro earnings. Measured in one currency the trailing P/E is 41.1, and tab 4 shows the four numbers it is built from. We do not hold a peer set, so this page does not print an "equipment makers trade at 30×" it cannot source. What tab 4 does instead is turn the question around: given the price, what growth is already assumed? Every step of that arithmetic uses figures we hold, in dollars throughout.

Three cases, with what would have to be true

Twelve to twenty-four months: what to watch, not what will happen

This page does not forecast a price. What it can say is which number will settle the argument. The number ASML's own investors would name is order bookings, and our data does not carry it. Of the numbers we do hold, the one that decides between the bull and the bear is revenue growth against the same quarter a year earlier, reported quarterly — because a cycle announces itself there before it reaches the margin. The next report is 27 January 2027, the fourth quarter: EPS estimate $15.56, revenue estimate $16.10bn, which would be 38.71% above the $11.61bn of the same quarter a year earlier.

2 Financial breakdown

What this method is for: to see whether the company is getting stronger or weaker over years rather than quarters, by reading growth, margin and earnings as one sequence instead of three headlines.

The sequence

Revenue: out of a flat year, and accelerating

From €27.56bn in 2023 to €28.26bn in 2024 is 2.56%; to €32.67bn in 2025 is 15.58%, a two-year compound rate of 8.88%. The trailing twelve months grew 22.54%, and the third quarter of 2026 55.16% against the same quarter of 2025. Each window is faster than the longer one before it. Read as a sequence it is a sharp acceleration; read as a shape — a flat year, then a surge — it is also exactly what one turn of an equipment cycle looks like. The sequence cannot tell the two apart on its own.

Earnings: the uncomfortable number runs in both directions

Net income was €7.84bn in 2023, €7.57bn in 2024 (−3.41%, on revenue up 2.56%), €9.61bn in 2025 (+26.91%, on revenue up 15.58%) and €13.47bn over the trailing twelve months (+42.39%, on revenue up 22.54%). EPS: €19.89, €19.24, €24.71, €35.00. In every period the earnings swing harder than the revenue: when revenue grew a little they fell, and when it grew 22.54% they rose 42.39%. That is operating leverage: a cost base that does not shrink when sales slow and does not grow as fast when they rise. It is why the last twelve months look so good, and it is the same arithmetic that took net income down in a year when revenue still grew.

Margins: the highest point in the table is the latest one

Gross margin is up 7.63 points from 2024 to the third quarter of 2026 and operating margin 12.44. One more line deserves a reader's attention. In the third quarter net income was 95.89% of operating income, against 85.03% for 2025 and 88.54% over the trailing twelve months — so something between operating profit and net profit was unusually favourable that quarter. Our data does not itemise the tax, finance and other lines between the two, so this page cannot say what it was, only that part of the quarter's net-income growth did not come from operations.

Why there is no rule of 40 here

The rule of 40 is a convention for subscription businesses; it does not apply to an equipment maker that sells machines into a cycle, and this page skips it.

Verdict of this method alone

Strong on every line we hold, and at the top of its own table. Revenue accelerating, margins at their highest, earnings growing nearly twice as fast as sales. This method also shows, in 2024, what the same structure does in a soft year, and it cannot tell you whether the price already knows all of that — that is tab 4's job.

3 Competitive advantage — the moat

What this method is for: to ask why the company's profits should still exist in ten years, since anything profitable and unprotected gets competed away.

The five sources, scored one at a time

Proprietary technology — exceptional, and most of the moat

At the leading edge there is no second supplier. ASML is the only company that makes EUV lithography machines, the tools that print the finest features on today's most advanced chips. The advantage is not a patent that expires; it is decades of engineering, a supply chain built around one product and the accumulated knowledge of making it work in a customer's factory. A competitor would need all three, and the first of them cannot be bought. This page describes the position in words because our data holds no market-share figure — but at the leading edge there is no share to divide.

Switching costs — strong

A chip factory's processes are designed, tested and qualified around the tools in it. Replacing a lithography supplier is not a purchasing decision; it is re-engineering the production line, and at the leading edge there is nothing to switch to. The installed machines are then serviced and upgraded by their maker for as long as they run, which turns every sale into a long relationship. Our data does not split that service income out, so the size of this source is described rather than measured.

Cost advantage — not the point

ASML's advantage is not that it makes machines more cheaply than a rival; at the leading edge there is no rival to undercut. What a monopoly supplier has instead is pricing room, and the margin sequence on tab 2 is consistent with it: gross margin from 49.82% in 2024 to 57.45% in the third quarter of 2026. Scored here as moderate, because the margin is evidence of the technology moat rather than a separate one.

Brand — moderate

Not a consumer brand, and it needs none. It is a brand in the engineering sense: customers commit the most expensive factories in the world to a roadmap they trust the supplier to deliver. That trust is real and earned, but it is a consequence of the technology, and scoring it separately would be counting the same thing twice.

Network effects — weak to moderate

Indirect only. Every machine in the field teaches the maker something about the next one, and suppliers and customers build their own processes around the platform. That is a learning loop and an ecosystem, not a network effect in the strict sense — one chipmaker's machine does not make another's more valuable.

Against the alternatives

The competition comes in two kinds. In older, less precise generations of lithography the company has competitors and the usual pressure on price. At the leading edge it has none, and the real long-run threat is a well-funded attempt to build an alternative — the kind of project only a government can afford, slow by nature and invisible in any financial statement until it works. The moat's actual limit is elsewhere: it decides who gets the order, not when the customer places it, and not which customers a government permits to buy. Those two risks are ranked on tab 5.

The score, and what it is worth

Nine, and the reasoning is the argument rather than the number. A sole supplier at the leading edge, whose customers cannot switch and whose installed base keeps paying for service, is about as strong a competitive position as an industrial company can hold. What keeps it below ten is that the moat protects ASML's share of chipmakers' spending and not the spending itself, and that a government can narrow the list of customers without any competitor lifting a finger.

4 Valuation

What this method is for: to separate the company from the price, because a good business bought at the wrong number is a bad investment and the two questions have different answers.

Where the multiple is

Forty-one times trailing earnings is a growth multiple, on a company whose earnings are currently growing at a rate that justifies one and whose history includes a year when they shrank. Three months ago it was 56.7 times: after the July report the share closed at $1,815.27 and the four reported dollar EPS figures summed to $32.01. Since then trailing EPS rose to $40.12, +25.34%, and the share fell to $1,650.00, −9.10%. A multiple compressing by more than a quarter in a single quarter in which the business did nothing but accelerate is the tension the rest of this panel measures. A dollar P/E for a year ago cannot be computed honestly: our calendar does not carry the three dollar EPS figures before October 2025.

The reverse question: what does the price already assume?

Rather than guess the future and discount it, hold the price still and solve for the growth that would justify it. Every figure in the table is in dollars. The arithmetic: pick a multiple a mature version of this company might trade at in five years, divide today's $1,650.00 by it to get the EPS that would be needed, multiply by the 383.5m shares to get the net income needed, then divide by today's net margin of 34.14% — a ratio, so the same in either currency — to get the revenue needed. Compare that with the $45.67bn of the last four reported quarters ($13.05bn + $10.67bn + $10.34bn + $11.61bn).

The two growth columns differ by about three tenths of a point because the calendar's dollar revenue was converted quarter by quarter, at each quarter's rate, while the margin is the euro statement's ratio. The EPS column is the cleaner of the two; neither changes the reading.

How to read that table. It does not say whether the price is too high or too low. It says: for the buyer at $1,650.00 to simply get their money back with the share trading at 30× in 2031, this company has to grow revenue by about 6.2% a year for five years and hold today's net margin. Over the last twelve months it grew 22.54% (in euros), and even the two years to 2025, which include the flat year, compounded at 8.88%. The table only turns demanding at 20×, where the price would need 15.2% a year for five years — below the current rate, above the two-year one. So unlike most growth stocks, the question here is not whether the growth exists today. It is how much of today's margin and growth survives the next turn of the cycle, because the table holds the margin at its highest point.

The comparison this page will not print

The method asks for an industry-average multiple. Our data does not carry a peer set for this company, so any "semiconductor equipment trades at 25×" here would be a number I remembered rather than measured, and the whole value of the table above is that it contains none of those. A reader who wants the peer comparison can build it in the app: open the chart's Key stats tab for two or three equipment makers and read the P/E line from the same source for each — and check, for any company that reports outside the United States, that the price and the earnings are in the same currency before trusting the ratio.

5 Risk

What this method is for: to work out what would have to go wrong and how much it would cost, because the size of a loss matters more than its likelihood when the position is large.

Ranked most dangerous first. "Dangerous" here means the product of how likely it is and how much of the share price it would take, not how frightening it sounds.

1. The cycle — the largest risk, and the one the price is already arguing about ASML's revenue is chipmakers' capital spending, and capital spending pauses. Our own table has one soft year in three: in 2024 revenue grew 2.56%, net income fell 3.41% and operating margin slipped from 32.81% to 31.92%. Today's figures are the opposite extreme — third quarter revenue +55.16%, operating margin 44.36% — and the reverse table on tab 4 holds that margin constant for five years. The likelihood of another pause at some point is high because it is the nature of the industry; the cost is high because operating leverage takes earnings down faster than revenue. The trailing multiple falling from 56.7 to 41.1 while earnings rose is what the market's estimate of this risk looks like from outside.
2. Export rules, and China Lithography machines are subject to government export controls, and a change in which machines may be sold to which countries changes who the customers are, overnight and regardless of demand. Ranked second because the cost of a change can be large and its timing is political rather than economic. It cannot be ranked more precisely, because our data has no geographic split of sales: this page does not know what share of revenue is at stake.
3. A short list of customers The buyers of the most advanced machines are, by the nature of the product, the few companies that can afford a leading-edge factory. A decision by one of them to delay a factory moves this company's revenue. Our data does not carry customer concentration, so the size of that dependence is described here, not measured.
4. The multiple 41.1 times trailing dollar earnings and a beta of 1.302 — a share that has historically moved further than the market in both directions. It is 17.50% below its June high and fell 3.94% on the day of a 19.31% beat. Ranked fourth not because it is small but because it is mostly the first risk seen from the price side: if the cycle holds, the multiple has room; if it turns, the multiple and the earnings fall together.
5. Currency ASML earns in euros and this share is priced in dollars, so a dollar holder owns a euro profit stream. Our own data shows the effect over one year: the calendar's dollar figures imply about 1.16 dollars per euro in the third quarter of 2025 and about 1.12 in the third quarter of 2026, so the same quarter's revenue grew 55.16% in euros and 49.42% in dollars. Ranked fifth because it shaves the growth rather than reversing it.
6. The next machine The moat is renewed with each generation of technology, and each generation has to work in customers' factories on schedule. A delay would show first in the order book and only later in the revenue line. Ranked last because nothing in our data can see it coming, which is a reason to keep it on the list rather than to rank it higher.

6 Growth potential

What this method is for: to ask where the next stretch of revenue would have to come from, and whether there is a plausible place for it to come from.

What the growth has actually been

Accelerating, sharply, in euros. The two-year compound rate is 8.88%, the trailing twelve months 22.54%, the latest quarter 55.16% against the same quarter a year earlier. The caveat is the second cell: the two-year rate includes a year of 2.56%, and the latest quarter is compared with a quarter that was itself only €7.52bn of revenue — so part of the acceleration is the base, not the business.

Where more revenue would come from

• More advanced machines. Each new generation of lithography prints finer features and is a larger sale than the one it replaces. This is the engine the margin sequence on tab 2 is consistent with. Our data holds no price per machine and no unit count, so this page cannot split the growth into "more machines" and "dearer machines".
• The installed base. Every machine sold is serviced and upgraded for years afterwards, so the base of recurring income grows with every year of sales. It is the steadier half of the business and the half that softens a cyclical year. Our data does not separate it, which is the single most useful split this page does not have.
• Chipmakers' capacity. Demand for the most advanced chips is what makes chipmakers build leading-edge factories, and each factory is an order. This is the mechanism behind the current acceleration and also the cyclical one: it is somebody else's investment decision.
• More lithography per chip. As chips grow more complex, more layers need the most precise printing, so the same factory needs more of the advanced tools. Described, not measured — our data holds no figure for it.

The arithmetic of the next five years

Tab 4 computed what growth the price assumes; this is the same figure read as an operating question, in dollars. To reach $61.79bn of revenue by 2031 — the 6.2% compound rate the 30× row needs — the company has to add about $16.12bn to the $45.67bn of its last four reported quarters. In euros, the last twelve months alone added €7.26bn over the twelve before them, while 2024 added €0.70bn. That is the whole growth question in two numbers: the requirement is modest against the current rate and not modest against a flat year. The growth case does not rest on the next year; it rests on the average across the next cycle.

The one forward figure our data does hold is the consensus for the next quarter: revenue of $16.10bn, which would be 38.71% above the same quarter a year earlier and 23.34% above the third quarter just reported. It is an estimate, not a measurement, and it is reported on 27 January 2027.

7 The institutional view

What this method is for: to look at the share the way the people who move it do, since a private investor is trading against institutions and it helps to know what they are solving for.

A fund manager is not answering "is this a good company". They are answering "does this position help the portfolio I have to report on", and those are different questions with different answers.

Why a large fund would own it

• It is investable at size. A $635.94bn market cap with daily volume averaging 1.42m shares over thirty days — about $2.34bn a day at today's price — means a large position can be built and, more importantly, sold.
• There is no substitute. A manager who wants exposure to the leading edge of chipmaking equipment has one company to buy for the part of it that matters most. Scarcity of that kind is why a position gets held through a bad quarter.
• The earnings momentum. Trailing net income up 42.39% in euros; third-quarter EPS up 126.52% in dollars, from $6.41 to $14.52. A growth mandate is built to find exactly that.
• The year. Up 67.82% against SPY's 16.21% over the same one-year window, a lead of 51.61 points. The leg since the November low at $946.11 has run 223 sessions, closed up on 121 of them, and traded on average 28.92% more volume than the 22 sessions before the low — a rise that was bought, not drifted into.

Why a large fund would avoid it

• A cyclical at a growth multiple. The position's outcome depends more on where the cycle is than on anything the company does next quarter, and a fund that buys at the top of an equipment cycle has to explain it for the length of the downturn.
• The last three months. Over the last 63 sessions the share fell 7.56% while SPY rose 2.51%, a gap of 10.07 points, despite the largest earnings beat in the table. A manager measured against the index has been paying for this position since July.
• The tape. Below the 20-day average ($1,762.01, −6.36%) and the 50-day ($1,741.30, −5.24%), still above the 200-day (+5.00%). RSI(14) at 38.1 from 39.4 twenty sessions ago; MACD at 3.67, below its signal at 17.99. A quantitative sleeve reads that as a weakening trend and does not care about tab 3.
• Risks it cannot size. A risk committee asks what share of revenue sits in a country whose export rules could change. Our data does not hold the answer; a fund would get it from the company, and the size of that number decides the position size before anything else on this page does.

The catalysts a professional would have in the calendar

• 15 October 2026 — the session after the report — not yet traded when this page was written. On the record of tab 9, the second session has as often reversed the first as extended it.
• 27 January 2027 — the fourth quarter and the full year. Estimates: EPS $15.56, revenue $16.10bn. The line that matters is revenue growth against 38.71%, and the order book, which our data does not hold and the release will.
• The $1,614.02–$1,642.35 band — the 20-day low and the 150-day average, on which the price closed. A close under it puts the $1,571.41 200-day average in play.
• The $1,873.50 line — the 20-day high, reached in early October before the slide into the report. A close back above it is what a trend-following buyer waits for.

The thesis, stated as an institution would state it

8 Bull versus bear

What this method is for: to force the strongest version of the argument you disagree with, because the case you cannot state is the one that costs you money.

Both analysts below are arguing from the same figures on this page. That is the exercise: identical data, opposite conclusions, and the reader watching where the disagreement actually is.

Where the disagreement actually is

They agree on every figure. They disagree about one question: are these earnings the top of a cycle or the base of a new level? The bull says the moat and the acceleration make this cycle different; the bear says every cycle looks different at its peak, and 2024 is two years old. Neither has proof, because the evidence is the order book — which our data does not hold — and a series of quarterly revenue growth rates that has not been reported yet. That is why tab 1 nominates revenue growth as the number to watch, and why an investor who cannot say which side of that question they are on is holding a position in the cycle's mood rather than in the company.

9 The last earnings report

What this method is for: to read a quarter as evidence rather than as news — what was expected, what arrived, and what the price did about the difference.

Q3 2026, quarter to 30 September, reported 14 October 2026

The headline below is the earnings calendar's, in dollars. The figures under the headline are the company's statements, in euros.

A large beat on earnings and a small miss on revenue — the largest EPS beat in the five reports our data covers, by $2.35 a share, on revenue $0.09bn short of the estimate. On the headline, an exceptional quarter with one blemish.

What was underneath the headline

Revenue up 55.16% against the third quarter of 2025; net income up 133.49%; EPS of €12.93, up 135.95%; operating margin 44.36% against 34.60% for the whole of 2025. Underneath the beat was the best quarter in our data on every line, with one qualification the headline does not show: net income was 95.89% of operating income, against 85.03% for 2025, so part of the jump in net income came from below the operating line, and our data does not say which line. In dollars the same quarter grew less — revenue +49.42%, from $8.74bn to $13.05bn; EPS +126.52% — because the dollar figures imply fewer dollars per euro than a year earlier. And the market, which read all of it, took the share down 3.94% on 2.02 times its 30-day average volume.

The market's answer, five reports running

"Close on the day" is the close of the report-day session, and the market's first answer — the 14 October session traded 2.02 times its 30-day average volume. "By the next close" is cumulative from the close before the report.

The rest of the year says the same from the other side. The five largest one-day moves in our series — +9.53% on 11 June, +8.78% on 2 January, +8.77% on 8 April, −7.82% on 23 June and −7.36% on 1 July — all fell on days without a report. The price of this share is being set between the reports, by whatever the market believes about chipmakers' spending, and the reports are read against that belief rather than against the estimate.

The size of the surprises is worth reading as a sequence on its own: +0.79%, −5.42%, +8.42%, +8.77%, +19.31%. After the miss in January, each beat has been larger than the last — the estimates have fallen further behind the business every quarter. A company whose analysts cannot keep up is accelerating; a market that sells the largest beat in the table is no longer paying for the acceleration. Whether that is because it expects the acceleration to end, or because of something in the release our data does not carry, this page cannot tell.

Next — 27 January 2027

Wednesday 27 January 2027. The fourth quarter and the full year. Estimates: EPS $15.56 (81.99% above the $8.55 of the same quarter a year earlier, and 7.16% above the third quarter's $14.52), revenue $16.10bn (38.71% above $11.61bn, and 23.34% above the third quarter). Based on the table above, the estimate is the least informative thing about it. What the market will be reading is whether the revenue growth rate is still rising, and what the order book says about the year after.

10 The decision

What this method is for: to put the other nine together and reach a decision — which, on this site, means reaching your decision. This page does not issue a verdict, and the next paragraph explains why that is a position rather than a dodge.

What each method concluded, in one line

The short term

Two numbers bound it, and both are on the chart. $1,614.02 is the 20-day low, with the $1,642.35 150-day average just above it — the band the share closed on, 2.23% above its floor — and the last line before the $1,571.41 200-day average. $1,873.50 is the 20-day high, reached in early October and lost on the slide into the report, and the last line before the $1,999.96 high of 30 June. Between them, the argument continues, with RSI(14) at 38.1 and MACD below its signal. The first scheduled test is the session after the report, untraded as this is written; the next is 27 January 2027.

The long term, five years and more

It reduces to one question, and tab 8 established that both sides already agree it is the question: are today's earnings the top of a cycle or the base of a new level? If they are a base, the 30× row on tab 4 asks for less growth than the company is producing, and the multiple's fall since July is the market being slow. If they are a top, the reverse table is built on a margin that has existed for one quarter, and 2024 is what the next soft year looks like before the multiple is applied to it.

The catalysts, dated

• 15 October 2026 — the first full session after the third-quarter report. In the table on tab 9 the second session reversed the first as often as it extended it.
• 27 January 2027 — the fourth-quarter and full-year report. Watch revenue growth against 38.71% first, the order book second, the EPS headline last.
• Any quarter in which revenue growth against the year before falls back toward the single digits — the shape of 2024 — is the bear case's first real evidence.
• A close below $1,614.02 — the 20-day low and the 150-day average fail together and the 200-day average at $1,571.41 is the next reference.
• A close above $1,873.50 — the 20-day high is regained and the slide since early October is at least questioned.

The questions to answer for yourself

These are the four the nine panels above cannot answer for you, because the answers are about you rather than about the company:

1. Do you believe this is the top of a cycle or the base of a new level — and what would change your mind? If your answer does not survive the 2024 row of the table on tab 2, you have a preference rather than a view.
2. Can you hold something that falls 17.50% from its high while its earnings rise? It did exactly that between 30 June and today. With a beta of 1.302, the question is not whether it can happen.
3. Is your horizon longer than the cycle? The question on tab 8 is settled by a series of quarterly growth rates and order books, not by one report. A position with a shorter horizon than its own thesis is a bet on the report, not on the company.
4. Do you own it in dollars or in euros? The same quarter grew 55.16% in the company's currency and 49.42% in the share's. If you have not decided which one you are measuring, every ratio you compute is at risk of the error on tab 4.