NextFin News - ASML raised its 2026 net sales outlook to €43 billion-€45 billion on July 15 and said it plans to add 30% to its 2026 low-NA EUV capacity for 2027, a sign that demand for advanced chipmaking tools is still outrunning supply planning even after a strong first half. The Dutch equipment maker had only recently guided to €34 billion-€39 billion for 2026, so the new range is not a cosmetic tweak. It is a larger signal: the company is telling investors that AI-linked spending, logic demand, and memory investment are still forcing a faster production ramp than its own earlier schedule assumed.
The immediate facts are straightforward. ASML said second-quarter 2026 net sales were €9.3 billion, gross margin was 54.0%, and net income was €2.9 billion. It also said Q3 sales should land between €11.0 billion and €12.0 billion, with gross margin between 55% and 57%, before the full-year outlook moved up to €43 billion-€45 billion from €34 billion-€39 billion. The same release said the company had booked enough demand in the first half of the year to plan a 30% increase in low-NA EUV capacity for 2027, from a 2026 base of around 65 units, and to study another 30% increase for 2028. That is the kind of manufacturing decision that usually follows not one quarter of strength but a longer queue of customer commitments.
That queue matters because ASML is not a generic capital-goods story. Its low-NA EUV tools sit at the center of leading-edge semiconductor production, where each capacity decision has a second-order effect on the rest of the chip supply chain. If ASML can only expand output at a measured pace, then foundries and memory makers have to plan node transitions around a constrained tool pipeline. If ASML can add capacity faster, that bottleneck loosens, but the benefit flows unevenly: the most advanced logic and DRAM customers get first access, while lagging fabs remain exposed to a tighter supply curve. The 2026 sales upgrade therefore says as much about who is still racing to secure tools as it does about ASML's own revenue line.
There is another reason the raised outlook matters. ASML had already told investors in January that 2026 would be another growth year, with net sales of €34 billion-€39 billion and a gross margin of 51% to 53%. It now sees €9 billion or more of extra annual revenue at the midpoint, which means the earlier guide was not simply conservative. The change implies that customer demand, especially for EUV-related systems and installed-base work, has turned into a stronger year-on-year bridge than ASML could see in January. That is a cyclical demand upswing for the quarter, but the decision to add capacity for 2027 looks more structural. Once a lithography bottleneck is entrenched, it does not clear itself with one good quarter; it requires tools, labor, logistics, and supplier investment.
For the market, the important question is not whether ASML can post a better 2026 number. It is whether the higher guide reflects a demand surge that still has runway or a temporary pull-forward from customers trying to secure tools before a later slowdown. If the former is right, ASML's revenue range may still prove conservative. If the latter is right, the 2027 capacity build could leave the company with more supply than demand once customers digest current spending. The company itself is leaning toward the first interpretation, but the market will need proof that the order book is broadening rather than merely accelerating.
Why The Guidance Change Matters More Than The Number
The bigger story is not the move from €34 billion-€39 billion to €43 billion-€45 billion. It is the mechanism that produced it. ASML said its order intake remained “extremely strong” in the first half of the year and tied that strength to ongoing AI-related investment, continued progress in AI technologies, and accelerated capacity-expansion plans by customers. In plain terms, chipmakers are not just spending on final demand for chips; they are spending on the machinery that will let them chase a larger AI-driven market later. That is why ASML’s own sales outlook can change so quickly. The company sits one layer upstream from the AI buildout, so a customer decision to add fab capacity today can show up in ASML bookings before it shows up in finished-chip output.
That transmission channel is powerful but not linear. A higher outlook at ASML does not automatically mean a clean, steady boom for the entire semiconductor equipment sector. It means the bottleneck has moved. When a company like ASML adds capacity, the binding constraint is no longer only customer demand; it becomes manufacturing execution, component availability, and the pace at which the company can hire, qualify, and ship complex systems. That is why capacity guidance is the more important line than the revenue guide. Revenue can rise because customers are eager. Capacity only rises if management believes the demand is durable enough to justify a bigger industrial footprint.
That is also why the move looks more structural than cyclical. Cyclical demand in semiconductor tools tends to come in waves tied to inventory replenishment, fab spending pauses, or memory upcycles. Structural demand, by contrast, comes when a technology shift changes the minimum capital intensity of the industry. EUV is in that second category. The need to pattern ever-smaller features is not going away, and the AI buildout has made advanced logic and DRAM more capital intensive, not less. ASML's own annual report pointed to long-term growth opportunity scenarios that contemplate 2030 revenue between approximately €44 billion and €60 billion, which is a sign that management already sees the business as anchored to a higher structural base than the pre-AI cycle.
That does not mean the whole move is structural. The order timing can still be cyclical, and the market can still over-earn the current year. But the capacity decision is the tell. Companies do not plan a 30% expansion in low-NA EUV output for a one-quarter pop unless they think the backlog is durable enough to justify multi-year capital and staffing commitments. The short-term revenue surprise is cyclical; the production build is the structural part.
“Given the business dynamics discussed above, we now expect total net sales for 2026 to be between €43 billion and €45 billion, with a gross margin between 54% and 56%.”
Christophe Fouquet, ASML President and Chief Executive Officer
That line matters because it ties the change to business dynamics rather than a one-off booking anomaly. It is the clearest official signal that the company believes customer spending is broadening enough to support a higher revenue base, not merely spiking for a quarter.
What The Capacity Plan Says About 2027 And Beyond
The capacity plan is the more revealing part of the announcement because it shifts the discussion from revenue recognition to industrial constraints. ASML said it is planning to add 30% to its 2026 low-NA EUV capacity of around 65 for 2027 and is investigating another 30% increase for 2028. That means the company is not just reacting to higher current demand; it is signaling that 2027 supply should be meaningfully higher than 2026 supply. Even if the exact unit count changes, the direction is clear: ASML expects customers to keep pulling on the EUV pipeline well into the next cycle.
This is where the second-order effects appear. If ASML successfully expands low-NA EUV capacity, the immediate beneficiaries are leading-edge foundries and memory makers that need more tools to move to the next nodes. But the broader industry effect can be paradoxical. More ASML output can lower the risk of delayed node transitions, yet it can also accelerate the competitive gap between the companies that secure tools early and the companies that do not. In other words, more supply from ASML does not democratize advanced manufacturing; it can widen the divide between the fastest buyers and everyone else. That is a different kind of concentration effect, and it is one reason the market cannot read the capacity increase as a simple good-news story.
The market has already priced a lot of this. Investors know AI infrastructure spending is feeding into the semiconductor equipment stack, and they know ASML is the most strategic bottleneck in that stack. What they had not yet fully priced was how aggressively management would translate demand into capacity expansion. The difference matters. A higher sales guide can lift the stock for a session or a week; a bigger capacity plan changes the earnings durability story into 2027 and 2028. That is the second-order question here: not whether 2026 is better, but whether the company is building an industrial base that can keep compounding after 2026.
The strongest counter-thesis is that this is still mostly a cyclical uplift, driven by a narrow wave of AI capex and memory spending that could soften once customers finish current fab ramps. Under that view, the raised 2026 outlook would be a timing shift, not a regime change, and the 2027 capacity build would risk overshooting if semiconductor spending cools. That is a serious objection, because ASML's own 2025 annual report still described 2026 non-EUV revenue as similar to 2025 and called out EUV as the main growth engine. If non-EUV stays flat and EUV demand merely pulls forward, the current enthusiasm could fade faster than the market assumes.
The falsifying signal is simple: if ASML's order intake or backlog meaningfully softens in the next two quarters while management still pushes capacity higher, the structural-demand case weakens. More concretely, if bookings stop supporting another step-up in 2026-2027 capacity or if 2027 low-NA EUV output plans are cut back from the 30% increase, the current read-through becomes too optimistic. Until then, the burden of proof sits with the bears.
Who Benefits, Who Is Exposed, And What To Watch Next
In the short term, the clearest beneficiaries are ASML's own revenue and margin expectations, which now sit well above the company’s January guide. In the medium term, customers at the front of the EUV queue benefit from a more visible tool pipeline, especially advanced logic and DRAM producers that need capacity to support AI-related demand. In the long term, the biggest beneficiaries are the semiconductor ecosystems that can use better access to lithography to keep node shrinks and performance gains on schedule. The exposed parties are the laggards: chipmakers that cannot secure tools early enough, suppliers that depend on a smooth fab buildout, and investors who assumed the AI capex cycle would plateau before 2027.
The next checkpoints are already visible. ASML's third-quarter 2026 guidance of €11.0 billion-€12.0 billion will test whether the first-half demand momentum is flowing through to shipments at the pace management expects. Backlog progression, bookings mix, and the pace at which low-NA EUV capacity turns into shipped systems will tell the market whether this is a one-off guide raise or the start of a longer industrial re-rating. The July 15 update also sets up a new debate around 2027: if customer commitments stay “extremely strong” into the next two quarters, the 30% capacity increase may still look conservative. If they do not, the market will remember this announcement as the high-water mark of the current wave.
The base case is that ASML remains the key bottleneck in the advanced-node supply chain and keeps converting AI-related demand into higher revenue through 2026 and into a larger 2027 capacity base. The upside case is that bookings and installed-base activity stay strong enough to force another step-up in 2028 as well. The downside case is that the current AI and memory cycle cools before the expanded capacity is needed, turning today’s plan into tomorrow’s overbuild. That is the real tension in the story.
ASML is not just raising a forecast. It is telling the market that the AI chip race has become an equipment race that now stretches into 2027.
Explore more exclusive insights at nextfin.ai.
