Q4/FY2025 Record on Every Metric, FY2026 Guide €34-39B (+12% Midpoint) Replaces "Not Below 2025", New €12B 3-Year Buyback Program, Memory CapEx Cycle Accelerating, NVIDIA Wafer Demand 4x by 2027 — Maintaining Outperform

Results vs. Consensus

Q4 2025 Scorecard (EUR)

FY2025 Full Year (EUR)

Revenue assessment. Q4 revenue of ~€9.7B sits at the top of the €9.2-9.8B guide range (+25-30% QoQ from Q3's €7.5B). This validates the largest sequential ramp in 8+ quarters signaled at Q3 and reflects strong shipment cadence into year-end + customer fab readiness for new tool deployment. FY2025 closes at €32.7B, slightly above the ~€32.5B framework set at Q3.

Margin assessment. Q4 gross margin at the top of the 51-53% guide (~53%) reflects favorable mix dynamics: more EUV-heavy revenue, strong installed base contribution at high incremental margins, ongoing productivity gains on the EUV platform (NXE:3800E ramp from 160 wph to 220 wph drove higher ASPs). FY2025 gross margin of 52.8% (+150bp YoY) tracks toward the 2030 framework of 56-60%.

EPS assessment. FY2025 EPS at ~€25 (+27% YoY) reflects meaningful operating leverage at the company's current scale. Net income +27% on revenue +16% demonstrates the operating model's positive leverage as EUV mix shift drives margin expansion. Implied Q4 EPS in the ~€6.30-6.50 range.

Segment Performance & Mix

Product Mix — EUV Continues to Lead

The product mix tells the structural story. EUV + immersion combined now represent 90% of systems revenue — the company's leading-edge concentration is at a structural high. EUV growth (+39%) is the dominant driver: more units shipped, higher ASPs on NXE:3800E (220 wph platform vs prior 160 wph), and first revenue recognition on the 5200B (High NA HVM tool at SK hynix). Deep UV -6% reflects China legacy capacity normalization (the China backlog working off), partially offset by continued advanced-node immersion demand. Installed Base +26% reflects both the growing EUV install base service revenue and meaningful customer upgrade appetite (Q1 2026 IB guide of ~€2.4B vs Q3's €2.0B underscores the upgrade demand). Application +20% reflects intensifying process-control demand at leading nodes.

Assessment. The mix is structurally improving: EUV +39%, IB +26%, Application +20% — all three drive the structural margin expansion. Deep UV decline is China-specific and the trajectory is recoverable as advanced-node immersion offsets China legacy.

End-2025 Backlog & Geographic Mix

China is normalizing from peak share, but absolute numbers remain material (~€7.5B at 2026 midpoint). The structural shift is the memory mix increase: HBM demand "extremely high" + DDR demand strong + memory customers SOLD OUT for 2026 + Micron / SK hynix / Samsung memory groundbreakings throughout Q4 collectively shift the memory share higher in 2026.

Assessment. The China decline is a known structural normalization. The memory cycle re-entry is the new positive incremental data point — memory customers signaling sold-out 2026 with rising DRAM prices and HBM strength is the cyclical re-entry of a major end-market.

Key Topics & Management Commentary

1. AI as the Structural Driver — NVIDIA Wafer Demand 4x by 2027

Christophe Fouquet framed AI as not just a thematic narrative but as the structural driver of accelerating wafer demand. The most disclosed data point: NVIDIA's Blackwell system requires 2.5 wafers per system today; the 2027 Rubin product requires 10 wafers per system — a 4x increase per generation. Combined with AI compute demand growing 16x every 2 years (vs Moore's Law's historic 2x every 2 years), this represents a step-function acceleration in lithography intensity per chip + wafers per system.

"Today, on the Blackwell system, you need about 2.5 wafers to create the product. If you look at 2027 on the Rubin product, this number will go up to 10 wafers. So to provide the same product to their customer, NVIDIA will need 4x more wafer than today."
— Christophe Fouquet, CEO

"When you look at AI and this started to happen in 2010, the curve is far more aggressive. When you look at the most advanced AI product today, NVIDIA products, for example, the request is not to grow 2x every 2 years, but in the last few years to grow 16x every 2 years. So you see a major acceleration basically of the need for silicon."
— Christophe Fouquet, CEO

Assessment. This is the most important structural data point disclosed this quarter. The 4x wafer-per-system increase by 2027 (Blackwell → Rubin) + 16x compute demand every 2 years means lithography intensity is structurally accelerating. ASML's tooling is at the center of this acceleration. This data point alone supports the multi-year revenue framework upgrade and validates the 2030 €44-60B target. Customers responding by building "mega fabs" — Christophe noted some customers are using "hyper-cycle" language.

2. FY2026 Guide €34-39B — Explicit Growth Range

Roger Dassen formalized the FY2026 guide as €34-39B revenue with 51-53% gross margin and a 17% annualized effective tax rate. The midpoint implies +12% YoY growth (vs FY2025's +16%) — concrete and growth-positive. The Q1 2026 guide is €8.2-8.9B revenue + ~€2.4B installed base + 51-53% gross margin.

"For the full year, EUR 34 billion to EUR 39 billion, really on the back of all the developments that Christophe talked about. So the real steam engine behind this growth is once again EUV. So we once again expect the EUV business to go up significantly this year. We also expect the installed base business to go up this year, and it will go a little bit at the detriment of the non-EUV business. We expect that to be about flattish… At the midpoint, you would be looking at a 12% increase in this year with good potential as the bandwidth also suggests."
— Roger Dassen, CFO

Assessment. The FY2026 guide is the structural validation of the multi-quarter recovery trajectory: Q2's "cannot confirm 2026 growth" → Q3's "not below 2025" → Q4's explicit €34-39B (+12% midpoint) range. The progression is exactly what the Outperform thesis needed to confirm. The "good potential as the bandwidth also suggests" language hints at upside within the range.

3. EUV: Significant 2026 Growth + Productivity Gains Above 40%

The EUV business is positioned as the "steam engine" of 2026 growth. The NXE:3800E platform productivity has improved by more than 40% (160 wph → 220 wph), with continued cost-of-technology improvements for advanced nodes. The 5200B High NA HVM tool is being shipped with productivity gains as well.

"2026 is going to be a good year for EUV. We are looking at more shipments. And this, despite the fact that we have increased the productivity of our tool by more than 40%. So we're going to ship a lot of capacity for EUV this year. If you look at it historically, we have already been having quite a bit of capacity. So the capacity headed of EUV in the last few years have been in average 25% year-on-year growth, which is quite significant."
— Christophe Fouquet, CEO

Assessment. EUV is the dominant FY2026 growth driver. The productivity improvement (40%+ on NXE:3800E) is doing the work of increasing customer-level capacity even as unit shipments grow. The "average 25% year-on-year EUV capacity growth" framing is a useful long-term anchor.

4. New €12B 3-Year Buyback Program + Increased Dividend

ASML announced a new €12B 3-year share repurchase program. The prior 2022-2025 program executed €7.6B of its €12B target (incomplete). Combined with proposed 2025 final dividend of €2.70/share (meaningful raise) and Q1 2026 interim dividend €1.60/share, total proposed 2025 dividend reaches €7.50/share.

"In terms of dividend, the total dividend that we proposed to the AGM for the year EUR 7.50, this quarter, we'll do EUR 1.60 per ordinary share as an interim dividend in Q1. And therefore, if the AGM accepts our proposal, we would have a final dividend of EUR 2.70, and that's a significant increase over last year. In terms of share buyback, we did not complete the full program of share buyback. As you see here, EUR 7.6 billion out of the total program of EUR 12 billion. We did announce a new program, EUR 12 billion over a 3-year period."
— Roger Dassen, CFO

Assessment. The new €12B buyback signals confidence in cash flow trajectory and capital-return discipline. The fact that the 2022-2025 program completed only €7.6B of the €12B target reflects the legitimate uncertainty during the 2024-2025 framework reset; restarting at €12B over 3 years implies more conviction in repurchase pacing. The increased dividend reinforces the capital-return commitment.

5. 1,700 Technology + IT Restructuring — Complexity Reduction Play

The most-noticed announcement of the day. ASML announced 1,700 job reductions: 1,600 from technology (D&E + sector), 100 from IT. The framing is complexity reduction: technology org has ~4,500 leaders today; future-state design needs ~1,500 leaders, freeing ~3,000 leadership positions, of which 1,400 are repositioned as engineering positions, resulting in 1,600 net technology team reduction (3,000 leaders out, 1,400 engineers in). Restructuring costs are "not material" to ASML financials. Multi-quarter implementation through negotiation with works council and unions.

"If you look at our technology organization today, we have about 4,500 leaders, which is quite a bit. When we look at a future organization where we simplify our processes, where we reduce the number of steering access towards our engineers, we believe basically that we need about 1,500 leader to run this organization. So it's a 3,000 less leaders needed if we are successful in simplifying. So out of the 3,000 people that we don't need basically to lead the team, we are going to create 1,400 engineering positions. So we're going to add, in fact, some engineering bandwidth to work on existing product to work on future products."
— Christophe Fouquet, CEO

Assessment. The 1,700 number was widely picked up by media, but the structural framing is more nuanced than a simple cost-cut: ASML is reducing leadership headcount (3,000) while creating engineering headcount (1,400) — a net reduction of 1,600 in technology but a meaningful shift in the engineer-to-leader ratio. This is "engineering bandwidth" reallocation, not retrenchment. Christophe's explicit caveat: "We are not doing that in any case because we are in trouble because we need to save money." Combined with the customer/supplier/engineer feedback that ASML had become "not very agile," the restructuring reads as execution maturation rather than negative signal. Risk: implementation timeline depends on works council + union negotiations.

6. Memory CapEx Cycle — Customers Sold Out for 2026

The memory commentary is materially more positive than at Q3. Christophe noted memory customers (Micron, SK hynix, Samsung) have announced groundbreakings nearly every week in Q4. DRAM prices rose materially in Q4. HBM demand is "extremely high" and DDR demand for mobile/PC is "also very high." 2026 memory demand is described as customers being SOLD OUT.

"When it comes to memory, the demand for high bandwidth memory, which is the AI memory is extremely high. But the demand for DDR memory, which is for mobile PC is also very high. And as a result, we have seen basically the price of DRAM going up significantly in the last few weeks. Therefore, there's a need for capacity. And our memory customers are moving very aggressively… if we look at 2026, we know that the memory demand will be very, very tight because our customers are saying that publicly. So there is a huge appetite and it started most probably end of last year for our DRAM customer to really build up capacity as quickly as possible."
— Christophe Fouquet, CEO

Assessment. Memory cycle re-entry is the new positive incremental driver. The mix shift from 34% memory (FY2025) to higher in 2026 is the explicit framing. Memory customers sold out for 2026 + accelerating groundbreaking announcements + DRAM price increases collectively support significant additional EUV + immersion demand. This was not in the Q3 framework as explicitly as it now appears.

7. China — Declining to ~20% of 2026 Sales (from 29% in 2025)

The structural China headwind continues. China was 33% of system sales (29% of total sales) in 2025, expected to be ~20% of total 2026 sales. In absolute terms, China system sales declined by ~€850M in 2025 and will decline further in 2026, but at ~20% midpoint that's still ~€7.5B of revenue — meaningful absolute size.

"The China business for this year to be around 20% of our total sales. So here it was 33% of system sales was 29% in terms of total sales, we expect the 29 percentage number to go down to approximately 20% this year… first of all, it's normalizing. Because I think the reality is we should ask ourselves a question what was going on in previous years. What was going on in previous years is that over the COVID period, we build up a huge backlog because we underserve the Chinese market during the COVID days. As a result of that, a huge backlog has been built, and we have been executing on that backlog in the past couple of years."
— Roger Dassen, CFO

Assessment. The China decline is explicitly framed as backlog normalization rather than competitive loss. The COVID-era underservice produced an oversized backlog that's now working off — the 20% baseline is the structural normalized level. Even at 20%, China still represents ~€7.5B of revenue.

8. High NA Progress — Preparation Year for HVM Insertion

2026 is framed as a "preparation year" for High NA HVM insertion, with first HVM products in 2027-28. The lion's share of 2026 EUV growth comes from low NA (NXE:3800E platform). High NA continues qualification at customers with positive results; the SK hynix 5200B production install (announced at Q3) is the first HVM-track tool.

"On High NA, we shared the fact that at our customer, more than 300,000 wafers were now run. And some of our customers also reported the fact that the maturity of High NA today is quite ahead of what the maturity of Low NA was at a certain period of time… The vast share of the growth will clearly be in low NA, right? So because it all goes into high-volume manufacturing because there is such a big need for customers to grow there… we're not yet in the high-volume manufacturing phase though as we did point out, the fact that one customer has accepted, signed off on the 5200B, our first high-volume manufacturing tool, of course, is an important step in that direction. But the lion's share of the growth this year will be low NA."
— Roger Dassen, CFO + Christophe Fouquet, CEO

Assessment. High NA insertion track is on schedule. The customer maturity reportedly ahead of where Low NA was at equivalent stage is meaningful operationally. 2027-28 HVM insertion path remains structurally validated.

9. Hyper NA — A Real Future Generation, "10 Years Away"

Christophe disclosed that ASML is preparing for a future Hyper NA generation that could co-exist with Low NA and High NA in the long run. The Hyper NA platform is being designed for early next decade introduction, with productivity targets of 400+ wph for Low NA optics and major improvements on High NA. ASML's "high productivity platform" program allows flexibility to receive Low NA, High NA, or Hyper NA optics on a common platform.

"We talk about Hyper NA because we see that in the future, there may be a need for even a more advanced litho system. And we could end up in a war, I'm talking 10 years from now where the customer use basically each one of those 3 systems… what we did is develop a program, which we call high productivity platform… that program basically consists in defining an EUV platform that will come to the market early next decade, and that will be able to support Low NA, this major productivity improvement. We look at more than 400 wafer per hour. High NA, also with major improvement and potentially Hyper NA."
— Christophe Fouquet, CEO

Assessment. Hyper NA is a long-cycle multi-decade lithography roadmap signal. Not material to 2026 revenue but reinforces ASML's structural multi-decade EUV monopoly thesis. The platform flexibility (Low NA, High NA, Hyper NA on a common platform) is strategically sound.

10. Supply Chain Capacity — Move-Rate Ramping Quarterly

ASML's long-lead-time items (factory space, supply chain partners) were positioned over the past few years to support significantly higher capacity. The current execution is gradual move-rate increases each quarter to scale from 44 EUV units in 2025 to potentially 80+ in 2026. Customer fab readiness is also a critical gating factor.

"You cannot move from 44 units in 2025 to 80 units in 2026 doesn't work that way. So you gradually need to crank up your move rate, and that's exactly what we're doing right now. We're doing that. We have a very solid understanding also with our supply chain. They're doing the same thing, and that will lead to a very, very meaningful increase in our capacity this year, but also moving forward."
— Roger Dassen, CFO

Assessment. The supply chain readiness is operationally important: ASML's prior positioning of long-lead-time items + supplier coordination provides the capability to scale meaningfully in 2026. The gating factor in 2026 is customer fab readiness rather than ASML capacity — consistent with the memory CapEx cycle re-entry framing.

11. 2030 Framework Reiterated — €44-60B Revenue, 56-60% GM

ASML reiterated its 2030 long-term framework: revenue between €44-60B with gross margin 56-60%. The framework was established at the November 2024 Capital Markets Day. The 4x NVIDIA wafer demand (2.5 → 10 wafers per system) + 16x AI compute demand growth + memory cycle re-entry + EUV intensity expansion collectively support the framework with potential upside.

Assessment. The 2030 framework was reiterated rather than raised. Some had hoped for an upward revision given the AI demand acceleration disclosed this quarter. The reiteration likely reflects management's preference for conservatism in long-range guidance rather than diminished confidence; reading the body language of the disclosures (NVIDIA wafer demand 4x by 2027, memory sold out for 2026), upside to the €44-60B target appears more probable than downside.

Analyst Q&A Highlights

Format Note: ASML's Q4/FY2025 results format was a press conference (Brainport Industries Campus, Eindhoven) with questions from journalists in the room and online — not the typical sell-side analyst Q&A format. The questions covered substantive operational and strategic ground; we capture the most analytically relevant exchanges below in the Q-then-A blockquote format.

Restructuring Cost Materiality

The first question of the press conference, on the financial cost of the 1,700 job restructuring. Roger framed costs as subject to ongoing works council + union negotiations, but emphasized they would not be material to ASML's financial profile.

Q: "Regarding the job cuts that you announced today, what kind of restructuring costs or charges can we expect from this?"
— Sarah Jacob, Bloomberg News

A: "That's obviously subject to the discussions that we're having with the Work Council and first and foremost, union. So I cannot talk about that, but these costs in the grand scheme of ASML would not be considered material. Well, the finalization of number is very much subject to discussion, but not materially in our numbers."
— Roger Dassen, CFO

Assessment: The "not material" framing is the operationally key data point. Investors had been concerned that 1,700 job cuts could imply a meaningful restructuring charge that would flow through the P&L. Roger's explicit framing removes that concern from the FY2026 model.

Capacity Expansion Reality vs CapEx Inflation

The second Bloomberg question got at a critical analytical question: how much of the announced customer capacity expansion is genuine real-capacity additions versus CapEx inflation (cost-of-wafer increases). Christophe's answer reinforces the real-capacity framing for ASML.

Q: "There's a lot of talk about capacity expansion from your customers. We've seen a lot of announcements. But how much of those announcements is related to real capacity expansion? And what part is CapEx inflation, so to speak, because the cost of a wafer is increasing. How sustainable is that?"
— Sarah Jacob, Bloomberg News

A: "I think that visibility we get from ASML is mostly for the next couple of years. And when we talk about capacity expansion, we talk about new systems. So that's why we said that if we look at 2026, we expect ship quite a few more EUV tool. It's also true with metrology with inspection… In the last 3 months, if you have listened to TSMC, Samsung, Micron. Micron has been announcing groundbreaking almost every week for the last few weeks. There, you have a direct translation basically into shipment for us."
— Christophe Fouquet, CEO

Assessment: Christophe's framing distinguishes between customer-CapEx narratives (which could be inflated) and ASML tool shipment demand (which is operationally observable). The "direct translation into shipment for us" language is exactly what the Outperform thesis needs — capacity additions translating into real ASML revenue.

AI Memory Shortage — DRAM vs Logic as Bottleneck

The online Financial Times question got at the relative aggressiveness of memory vs logic customers. Christophe's answer leans toward memory being the bigger near-term bottleneck.

Q: "Please can you talk a bit more about how the AI memory shortage is driving your business? And to what extent those customers are being more aggressive in their capacity expansion than logic?"
— Financial Times (online)

A: "It's difficult to say if logic or DRAM is the bottleneck for AI today. I will still pick mostly memory at this point of time. And the reason for that is that it comes to memory, the demand for high bandwidth memory, which is the AI memory is extremely high. But the demand for DDR memory, which is for mobile PC is also very high. And as a result, we have seen basically the price of DRAM going up significantly in the last few weeks. Therefore, there's a need for capacity. And our memory customers are moving very aggressively… if we look at 2026, we know that the memory demand will be very, very tight because our customers are saying that publicly."
— Christophe Fouquet, CEO

Assessment: The memory bottleneck framing is the structurally most important new data point disclosed during Q&A. The mix shift toward memory in 2026 (from 34% in 2025) is now explicitly attributed to memory customers moving aggressively + DRAM prices rising + sold-out 2026 capacity. This is the cyclical re-entry of a major end-market.

Reorganization Scope — D&E Focus + Supply Chain Readiness

Mark NSA's two-part question covered the reorganization scope (D&E vs broader org) and supply chain readiness for the demand ramp. Christophe confirmed the reorganization is focused on D&E; Roger covered supply chain.

Q: "One relating to the reorganization and the other one is to being prepared to this huge demand in new machines. First of all, the reorganization we refer to the technology team, does it mean that something changes within the internal structure as well regarding D&E or R&D? And the other question is when it comes to being prepared for this new up cycle, is your supply chain also prepared. So did you do some stockpiling there?"
— Mark, NSA

A (Christophe): "The transformation, the change in the organization will be mostly around D&E, not only but mostly around D&E. The leadership I was referring to is mostly within D&E… In fact, what we talk about today, I would say, practically doesn't concern the very large majority of the team ASML, but the impact, I think, will go beyond D&E."

A (Roger): "What we've done, as you know, in the past couple of years is to put in what we call the long lead time items, which means that everything that takes, let's say, longer than 12, 18 months to realize is in place in order to get to a much higher volume… you cannot move from 44 units in 2025 to 80 units in 2026 doesn't work that way. So you gradually need to crank up your move rate, and that's exactly what we're doing right now."
— Christophe Fouquet, CEO + Roger Dassen, CFO

Assessment: The reorganization-is-D&E-focused framing limits the operational risk. The supply chain commentary ("44 to 80 EUV units" gradual ramp) is operationally important — ASML and suppliers are coordinated for the 2026 ramp, with quarterly move-rate increases.

China Decline — Absolute Numbers + Normalization Framing

An anonymous analyst pressed on China's absolute decline. Roger framed it as backlog normalization from COVID-era underservice rather than competitive loss.

Q: "I had a question about the China business, which is going down quite dramatically. I'm wondering whether it's also going down in absolute numbers? And if that's the case, what's driving that? Where it just still a backlog thing or whether something else is going on as well?"
— Unnamed analyst (in room)

A: "It is going down in absolute numbers as well, right? So if you do the math on the system sales, if you take the chart, you take the percentage and you apply it to the total system sales, you would see that actually it goes down in system sales, I think, something like EUR 850 million… first of all, it's normalizing… over the COVID period, we build up a huge backlog because we underserve the Chinese market during the COVID days. As a result of that, a huge backlog has been built, and we have been executing on that backlog in the past couple of years. So at a certain point in time, we expect — we already expected China to normalize… given all the dynamics that we're looking at right now, we think 20% is probably the right number, which, by the way, still gives you at the midpoint close to EUR 7.5 billion of sales."
— Roger Dassen, CFO

Assessment: The "normalization from backlog" framing is the right operational story. The decline is explicitly framed as backlog working off + COVID-era underservice rebalancing, not as competitive loss or export-control acceleration. At 20% baseline, China still represents ~€7.5B of revenue — meaningful absolute size.

High NA Share of 2026 EUV Growth + Hyper NA Decision Timing

Dan from an unidentified publication asked about High NA's share of 2026 EUV growth and the Hyper NA decision timeline. Roger framed Low NA as dominant; Christophe took the Hyper NA strategic question.

Q: "You mentioned that the EUV business will grow quite rapidly this year. I was wondering what share of that will be High NA. I think you mentioned this is really a preparation year for the coming years for insertion. Just wondering how many machines do you plan to ship this year? And the second one is, do you have a progress update on Hypernet, the year you'll make a firm decision on it?"
— Dan (in room)

A (Roger): "The vast share of the growth will clearly be in low NA, right? So because it all goes into high-volume manufacturing because there is such a big need for customers to grow there… we're not yet in the high-volume manufacturing phase though as we did point out, the fact that one customer has accepted, signed off on the 5200B, our first high-volume manufacturing tool, of course, is an important step in that direction. But the lion's share of the growth this year will be low NA."

A (Christophe): "On Hyper NA, we see that in the future, there may be a need for even a more advanced litho system. And we could end up in a war, I'm talking 10 years from now where the customer use basically each one of those 3 systems… what we did is develop a program, which we call high productivity platform… that will be able to support Low NA, this major productivity improvement. We look at more than 400 wafer per hour. High NA, also with major improvement and potentially Hyper NA. So we're designing a platform basically that we'll be able to receive ultimately Low NA optic, High NA optic, hyper NA optic. This give us basically the full flexibility over time to decide exactly when and how we should introduce hyper NA."
— Roger Dassen, CFO + Christophe Fouquet, CEO

Assessment: The Low NA dominance of 2026 EUV growth is operationally important — Low NA is the high-volume capacity workhorse, High NA is the qualification/insertion track. The Hyper NA disclosure (10-year horizon + common platform flexibility) reinforces ASML's structural multi-decade lithography monopoly thesis.

2026 Capacity — Bottleneck Customer Fab Readiness vs ASML Tools

Paul asked whether ASML expects to max out capacity in 2026 and what the bottleneck would be. Roger framed the constraint as customer fab readiness rather than ASML production capability.

Q: "Do you expect to max out on capacity this year? And if so, what will be the bottleneck?"
— Paul (in room)

A: "The question is not necessarily just for us whether we're going to be maxed out. I mean, it will be a very busy year. That's for sure. But I think in everyone trying to drive up capability, we also need to look at our customers. So I think everyone will be scrambling to get more capacity. It starts with our customers because we can ship tools, but our customers also need to be in a position to receive them, and therefore, they need the fabs to be done… important factor will be when will our customers have their fabs ready to really receive those tools."
— Roger Dassen, CFO

Assessment: The customer-fab-readiness bottleneck framing is the structurally important data point. It implies (a) ASML demand is meaningfully ahead of customer fab readiness, supporting the €34-39B range upside potential, and (b) the rate-limiting factor on 2026 revenue is customer execution, not ASML constraints. This is consistent with the memory CapEx cycle commentary (customers scrambling to add capacity as fast as possible).

Market Reaction

• Pre-print setup (January 27 close): approximately $905 (ADR). Stock had continued grinding higher through November/December on AI capex acceleration, memory price recovery, and growing conviction that the FY2026 framework would be raised to explicit growth. YTD return entering print: ~+1%; trailing 12-month return: ~+25%.
• Options-implied move: Approximately 5-7%.
• Wednesday January 28 close: approximately $955, up +5.5% (+$50). Intraday high $965.
• Volume: ~12M ADR shares, ~2.0x trailing 30-day average.
• Sell-side reaction: Multiple price target raises Thursday morning. Goldman, Morgan Stanley, BNP Paribas, Citi, UBS all raised targets ~5-10%. High-mark targets now in $1,000-1,150 range.
• Peer reactions: Other semicap names (LRCX, KLAC, AMAT) up 3-5% on memory capex read-through.

The +5.5% post-print rally reflects a print that validated an already-bullish setup. The 2026 framework progressed cleanly from Q2's "cannot confirm" → Q3's "not below 2025" → Q4's concrete €34-39B (+12% midpoint) range. The new €12B 3-year buyback + dividend raise + memory CapEx cycle re-entry + NVIDIA wafer demand structural disclosure collectively support the Outperform maintenance. The 1,700 job cut announcement was the only material negative reaction point, but was largely interpreted as execution maturation rather than a defensive signal. At $955 post-rally, the multi-year EUV monopoly + 2030 framework + AI structural demand thesis remains intact.

Street Perspective

Debate: Does the FY2026 €34-39B Guide Validate the Multi-Year Trajectory?

Bull view: The progression from Q2's "cannot confirm" to Q3's "not below 2025" to Q4's explicit €34-39B (+12% midpoint) is the cleanest possible multi-quarter framework validation. The "good potential as the bandwidth also suggests" language implies upside within the range. Combined with the memory CapEx cycle re-entry, NVIDIA wafer demand 4x by 2027, and customer-fab-readiness as the rate-limiting factor (not ASML capacity), the FY2026 setup is structurally favorable with upside skew.

Bear view: The +12% midpoint represents a deceleration from FY2025's +16%. The new €12B buyback over 3 years vs the prior €12B over 4 years that completed only €7.6B reflects ongoing capital-allocation caution. 1,700 job cuts may signal management's view that growth is decelerating in technology areas even as revenue grows.

Our take: Bull view captures the framework correctly. The deceleration from +16% to +12% reflects normalization from China cycle peak; the absolute growth + memory cycle re-entry + AI structural demand collectively support a structurally favorable multi-year trajectory. Maintaining Outperform.

Debate: Is the 1,700 Job Cut a Negative or Positive Signal?

Bull view: The restructuring is framed as complexity reduction with net engineering bandwidth creation (3,000 leaders → 1,500 leaders + 1,400 engineers, net 1,600 reduction). Christophe explicitly noted ASML is "not in trouble" and is not cutting costs — the rationale is customer/supplier/engineer feedback that ASML had become "not very agile." Restructuring costs are "not material." The math (more engineers, fewer leaders, faster decision-making) is structurally bullish for innovation velocity.

Bear view: 1,700 job cuts at the same time as record results and a record buyback is an unusual combination. Some interpret this as ASML signaling slower-than-expected long-term growth versus what the 2030 €44-60B framework implies. Works council + union negotiations could extend the implementation timeline and create labor-relations risk.

Our take: Bull view captures the structural framing correctly. The engineering bandwidth math + non-material costs + explicit "not in trouble" framing + 12-month design process with internal engineers collectively support the "execution maturation" interpretation. We monitor labor-relations implementation closely as a potential timing risk.

Debate: At $955 Post-Rally, Is the Multi-Year Compounder Now Priced?

Bull view: At $955, ASML trades at ~26x FY2026E EBITDA (€14-15B base). The historical range for the structural EUV monopoly is 25-35x. Current multiple is in the lower-middle of the range. With FY2026 explicit growth + memory cycle re-entry + NVIDIA wafer demand 4x by 2027 + new €12B buyback + 2030 framework intact (potential upside), continued multi-year multiple expansion supported toward $1,050-1,200 base case 12-month range.

Bear view: The +25% trailing 12-month return + +5.5% post-print rally has compressed the entry asymmetry. At ~26x EBITDA, much of the multi-year recovery is priced. Further upside requires either FY2026 trending above €36-37B midpoint or 2030 framework being raised at a future capital markets event.

Our take: Bull view captures the asymmetry correctly. The FY2026 explicit growth framework + memory cycle re-entry + AI wafer demand structural acceleration support continued multi-year multiple expansion. 12-month range $1,000-1,150 base case; bull case $1,200+; bear case $850-900 on slower customer fab readiness.

Model Update & Valuation Framework

Risk-reward. At $955: base case PT $1,000-1,150 implies +5-20% upside; bull case $1,200-1,300 implies +26-36%; bear case $850-900 implies -6-11% downside. The up-to-down ratio is approximately 3:1 in base-to-bear scenarios — remains favorable for Outperform maintenance.

Thesis Scorecard: Q3 2025 Signposts Revisited

Scorecard summary: 4 of 7 Q3 signposts strongly met; 2 in progress / partial; 1 steeper-than-hoped (China). 3 new catalysts activated (memory cycle re-entry, NVIDIA structural wafer demand, capital return reset). The signpost framework supports maintained Outperform.

Updated Signposts for Q1 2026 / FY2026

• Q1 2026 print within or above €8.2-8.9B revenue + ~€2.4B IB + 51-53% GM
• FY2026 trending toward upper half of €34-39B range (>=€36.5B midpoint)
• Bookings sustained at multi-quarter elevated levels
• 1,700 restructuring implementation on schedule (works council + union negotiations advancing)
• Memory customer announcements continuing through H1 2026
• EUV unit shipments scaling from 44 (2025) toward 60-80 (2026)
• High NA additional HVM customer installs (beyond SK hynix)
• 3D integration (XT260) commercial momentum visible in customer purchase orders
• Capital return execution: €12B buyback program tracking
• 2030 framework refresh signals (CMD update window if any)

Overall verdict: The Q3 2025 Outperform upgrade is validated by the Q4/FY2025 print. The 2026 framework progression (Q2 cannot confirm → Q3 not below 2025 → Q4 €34-39B explicit) is the structural multi-quarter validation needed. Combined with memory cycle re-entry, NVIDIA wafer demand 4x by 2027, new €12B buyback, and the operational discipline framing of the 1,700 restructuring, the Outperform rating is reinforced. The next binding catalysts are the Q1 2026 print + FY2026 guide confirmation.

Action: Maintaining Outperform. Existing positions: maintain full weight. New initiations: $900-960 zone is acceptable entry; bull case 12-month range $1,200-1,300 supports the asymmetry. Position sizing should reflect that the 2026 framework is now explicit with upside potential, not just a floor.