Veeco Instruments Inc (VECO) Q3 2026 Earnings Call Transcript

for additional orders in the coming months, supporting their pilot lines and high-volume manufacturing needs. Turning to memory, we are in the early stages of building this business as customers begin adopting laser annealing technologies. We see significant runway for growth as adoption expands, supported by AI-driven demand for DRAM and NAND technologies. We continue to make progress with leading memory customers, including our LSA tool serving as the production tool of record at a Tier 1 high-bandwidth memory customer that accelerated their investment in 2026. We are advancing an LSA evaluation system at a second Tier 1 DRAM customer with potential follow-on orders in 2027 and 2028. And our engagement with a third DRAM customer could lead to an evaluation agreement in 2027. Our opportunity is also expanding into NAND. Several customers are exploring applications for our LSA and NSA platforms, and we expect to ship our first LSA system for a NAND application to a leading customer by year-end. As 3D NAN manufacturers increase layer counts and device complexity, controlling thermal exposure becomes increasingly critical. Together, these opportunities support our confidence in an annealing SAM of approximately $1.3 billion by 2030. Beyond annealing, we see another meaningful opportunity to expand our presence in memory through our IBD300 platform. Two systems are currently under evaluation at leading DRAM customers for advanced memory applications, including bit-line metallization, where low-resistance films are increasingly critical. These evaluations are progressing with strong customer engagement and program expansion in 2027. Lastly, we continue to be a market leader in ion beam deposition for EUV mass blanks, working closely with major customers to enable their technology roadmaps. Demand is supported by AI-driven scaling, increasing EUV layer counts at leading-edge logic nodes, and the expanding adoption of EUV and high-bandwidth memory applications. Our technology is a critical enabler of the multi-layer mirrors used in EUV mask blanks, and we are deeply engaged with customers and industry leaders as they advance their roadmaps toward high NA lithography and larger format masks. This close collaboration positions us at the forefront of enabling the industry's next generation of EUV technology. We are also expanding our EUV-related opportunity into pellicle applications, where we continue to win production business at a Tier 1 foundry and engage with new customers. Taken together, we see our ion beam deposition SAM in the front-end semiconductor market to be approximately $500 million by 2030. On the next slide, I'll discuss our compound semiconductor market and projected SAM through 2030. AI infrastructure is driving greater demand for high-speed optical connectivity and power-efficient semiconductor devices, creating attractive growth opportunities across our portfolio. In silicon photonics, we project a $700 million SAM by 2030. Primarily driven by indium phosphide lasers for high-speed optical connectivity. We are gaining momentum across the laser manufacturing process with multiple Lumina and Lumina Plus orders year-to-date, a market leader position in wet processing, and our differentiated Spectre IBD technology for laser facet coatings. I'll discuss this opportunity in greater detail on the next slide. In other photonics, We project a $550 million SAM by 2030, including applications in space-based solar, micro-LED displays, and other emerging photonics technologies. And in GAN power, we project a $250 million SAM by 2030, supported by AI data center power requirements, electrification, and other high-power density applications. Our Propel 300-millimeter GAN on silicon platform continues to progress with a leading power semiconductor customer as we work toward completing technical qualifications for future capacity expansion. Collectively, these opportunities position VICO to benefit from the growing requirements for advanced connectivity and power efficiency as AI infrastructure continues to scale. Next, I'll dive deeper into our role in silicon photonics. Within the compound semiconductor market, silicon photonics remains one of our most compelling AI-driven growth areas. As AI clusters scale, higher bandwidth requirements are accelerating the shift from copper to optical connectivity. This transition is driving adoption of advanced indium-phosphide lasers across... Higher-speed pluggable transceivers and, longer-term, near-packaged and co-packaged optics. Vico is uniquely positioned across three critical steps in the indium phosphide laser manufacturing process – epitaxy, wet processing, and laser facet coatings. Given our customer engagements, we continue to believe this represents at least a $2 billion opportunity over the next several years. In MOCVD, momentum continues to build for our Lumina and Lumina Plus platforms. We have received multiple orders year-to-date from leading photonics customers. We are installing systems at existing customers. Pursuing additional expansion opportunities and engaging with prospective customers across Epitaxi applications. We continue to win business supported by competitive advantages of our platform, including differentiated productivity, reduced cleaning requirements, and larger batch processing. In wet processing, our wafer etch and wafer storm platforms provide critical etching, cleaning, and surface preparation solutions for laser manufacturing. And in ion beam deposition, our specter tool addresses the critical laser facet coating step. As lasers move to higher power, the requirements for tighter film quality and greater reliability become increasingly stringent. Playing directly to the strengths of our differentiated IBD technology. Together, these capabilities give us broad exposure to the continued scaling of optical connectivity. We're investing to more than double capacity in 2027 to support our customers' unprecedented demand and capture this multi-year growth opportunity. With that, I'll turn the call over to John to review our financial results.

Thank you, Bill. Revenue came in at $209 million within our guidance range. For the third quarter, our semiconductor revenue was $135 million, an increase of 4% from the prior quarter and comprising 65% of revenue. It was largely driven by laser annealing systems to leading-edge foundry logic and memory customers, along with web processing systems for advanced packaging. For the full year of 2026, we expect semiconductor revenue to grow by low to mid-teens compared to the prior full year. This performance is driven primarily by additional sales to leading memory, foundry logic, and EUV customers serving AI and high-performance computing applications. In 2027, building on this momentum, we expect revenue growth to meaningfully accelerate, led by advanced packaging. Compound semiconductor revenue for the third quarter totaled $40 million, a 96% increase from the prior quarter, totaling 19% of revenue. For the full year 2026, we expect compound semiconductor revenue to approximately double versus full year 2025. Growth is predominantly driven by our MOCVD and IBD systems for silicon photonics. Looking ahead, in 2027, we expect revenue growth to accelerate, driven by the significant backlog for these tools. Turning to data storage, revenue for the third quarter was $20 million, representing 10% of revenue. We expect revenue to approximately double in data storage year over year in 2026. Demand continues to increase driven by cloud computing and AI data centers. We have order visibility for 2027 and extending into 2028. We are aligned closely with our customer roadmaps as they adopt HAMR, increasing the need for our ion beam technology. Lastly, in Scientific and Other, the third quarter revenue is $14 million, representing 6% of revenue. Turning to the quarterly revenue by region, revenue from Asia-Pacific region, excluding China, was 45 percent, an increase from the previous quarter. Revenue from the U.S. accounted for 31 percent of revenue, an increase from the previous quarter, primarily from advanced semiconductor customers. Our China portion was 12 percent of revenue, a decrease from the previous quarter. EMEA and the rest of the world accounted for 12 percent of revenue as well. Turning to the third quarter non-GAAP operating results. Gross margin came in at 39 percent, below expectations due to higher supply chain costs, product mix, and the timing of revenue recognition on an LSA system under U.S. Government Export Review. Operating expenses total approximately $57 million, and income tax expense was $3 million. Resulting in an effective tax rate of approximately 12 percent. Net income was $24 million and diluted EPS was 36 cents on 66 million shares. Turning to the next slide to review the balance sheet and cash flow highlights, we ended the quarter with cash and short-term investments of $483 million, an increase of $53 million from the prior quarter. From a working capital perspective, our accounts receivable increased by $6 million to $154 million. Inventory increased by $16 million to $308 million and accounts payable increased by $11 million to $68 million. Importantly, customer deposits increased by $58 million to $169 million, reflecting strong customer commitment and demand. Cash flow from operations totaled $58 million, and CapEx totaled $4 million during the quarter. Lastly, I'll review our fourth quarter and updated full-year 2026 non-GAAP outlook. Demand across our key markets remains strong, and we continue to make deliberate investments to support our future growth. This includes adding capacity and expanding our supply chain to ensure we're well-positioned to meet customer needs as these opportunities scale. While there are certain industry-wide component constraints, we're actively working with our suppliers to address them. Fourth quarter revenue is expected to be between $230 and $260 million. Gross margin is expected to be between 43% and 44%, reflecting improved volume and more favorable product mix. We expect OPEX between $64 and $66 million, net income between $34 and $47 million, and diluted EPS between $0.50 and $0.70 on 66 million shares. We are updating our full-year revenue guidance to be between $790 and $820 million and expect our non-GAAP diluted earnings per share between $1.32 to $1.52 based on 66 million shares. Our third quarter performance and updated full-year outlook demonstrate our continued focus on delivering strong financial results while investing deliberately for the future. With a solid foundation, we remain well positioned to fund our capacity expansion, support our customers, and execute on the opportunities ahead. With that, we'll now open the call for questions.

Thank you. As a reminder, everybody, given the pending merger with Excellus, vehicle management will not be addressing questions related to the transaction. Thank you. Now we will begin the question and answer session. So if you do have a question, please press the star followed by the one on your touchstone phone. You will hear a prompt that your hand has been raised. Did you wish to decline from the polling process? Please press the star followed by the two. And if you are using a speakerphone, please lift the handset before pressing any keys. One moment, please, for your first question. And your first question comes from Charles Shi from Needham. Please go ahead.

Hey, thanks for taking my question. John, I think I heard you talking a little bit about supply chain component constraints. Before you gave the outlook, but wondering if you can provide a little bit of additional color on how the supply chain situation looks like for you guys right now. And I noticed that you gave some early view in the 27 as well, but we all know you have very strong backlog, semi-compound semi-accelerating, but there's still a little bit range of possibility there. Demand, unconstrained demand is one question, and assuming you're still going to have some components that are constrained into next year, what is the more realistic expectation for next year? Hopefully, you can help us from that angle, demand versus supply constraint. Thank you.

Sure. I'll handle part of that question. I'll turn part of the question over to Bill as well as talk about demand for next year. So I would say, Charles, similar to what Peer Company has been reporting, we have seen supply chain challenges increasing. We've experienced longer lead times and certain component shortages. Our supply chain organization is working closely with suppliers. We're placing orders earlier, qualifying alternative suppliers, and making use of advanced payments to secure supply. And as you mentioned, we've been working to align supply with higher demand outlook. And providing our suppliers with greater forward visibility, placing orders against our 27 requirements for the full year, 2027 at this point, and expanding our relationships with contract manufacturers there as well. So, while there are constraints... And execution risk, we believe we've identified the key bottlenecks and have actions in place to address them and believe our ability to ramp supply chains, you know, to support the demand as we move through 2027.

And the second half of your question, Charles, this is Bill, was, you know. Are those demand constraints limiting your outlook for 27? I would say, as you just said, that, you know, we continue to see incredible demand just driven by AI infrastructure, etc. And I would say, if I look at where people are kind of targeting WFE in 27 and kind of the range of 20 to 40 percent growth year over year in 27. I would say, and this includes a view of the supply constraints, which is your question, you know, we see across the whole corporation, all key segments, we see our 27 growth at the higher range of those WFE estimates, It's kind of closer to that 40% range.

Thanks. Maybe, Bill, a quick follow-up, right? So, let's say you grow close to that 40 percent. Would that be a potential – I mean, is that the possible range still with some of the supply constraints that you're probably going to have in mind, or that's really just the unconstrained demand number right now? No, no, that's...

That's a view that takes into account potential supply constraints and bottlenecks, and is our kind of current estimate, factoring all that in.

Great. Well, that's very good, Carla. So maybe a follow-up question on maybe two product-level, small product-level questions. One, um, people have been, uh, I'm not expecting this Ghana silicon repeat orders from your... Top customers wondering, because you do have a long lead time, right, wondering what's the latest, when you think the volume ramp is going to happen, when the repeat order is going to come. That's one. One. The other, there is the hardest drive demand, and I did hear you talk about visibility extending now already into 28. Wondering if you can provide a little bit more help rather than just a young year, but on the quarterly runway, where are we heading to? I think your The bottom level was kind of like 10 million per quarter, right? That's a purely service for data storage, but you are now running just a 20. Maybe you're going to the 25 million per quarter, but wondering if we can ever get back to the prior peak level, which I believe exceeded probably 50 million per quarter. Want to get some thoughts on how you think about the ramp of data storage as well. Thank you. That's two questions rolled into one. Thank you.

Okay. I'll try to keep me honest here, Charles. On GANON silicon, we do have an eval at a large IDM for a 300-millimeter GANON silicon. Our understanding from close engagement with the customer is the machine is operating well. It's producing high-quality epitaxy. It's being used to Provide parts to their customers for qualification. We did announce, I believe, that we are in receipt of kind of a HVM order for shipment in 2027 to support our customer's pilot line. So I think it probably will be starting in the back half of 27, and depending on how their customer demand. Uh, evolves that would probably ramp in 28. That's that's Bill's guess of the customer's view. So there's no questions on GAN power. I'll move to data storage. I would say you're right that the kind of $10 million run rate is kind of a service only, and we have effectively doubled the business in 2026, up to an $80 million rate, and we do have visibility into 27 and into 28. I think the challenge that makes answering your question on a quarterly basis challenging is that the ASPs of these tools are kind of in the... You know, five, eight, $10 million each. So a tool moving around between one quarter and another doesn't really justify. But I would say if I look at 27 in totality, I would say it's probably going to grow in about the 20% range, probably half the rate of the rest of the company. But we do have very good visibility of that business.

All right, so it sounds like the HDD side of the business, I mean, compared with, let's say, the 2021 peak, previous cycle peak, is still pretty constrained at this point.

I don't know that it's constrained. Charles, this is John, constrained, or this is what the customer telling us demand is. They're saying capital intensity is like 4% to 6% of their revenues. We wouldn't be capacity constrained going into this level of business. So I don't think that the demand is there right now for going back to sort of peak levels our customers is giving us. I think, as you know, we've been working with these customers for a long period of time on a bill-to-order model. They're giving us their requirements for 2027 and highlighting sort of the beginning of 2028 at this point. And none of those demand signals from the customers would indicate we'd be going back to peak level. For those customers. Now, that could change if their demand signals change, but we're not receiving those type of demand signals from the customer.

And we are not supply constrained in this area, given the full lead time that we have visibility that we have.

Got it. Thanks for the update. Thank you, Charles. Thank you, Charles.

Thank you. And your next question comes from David Dooley from Steelhead. Please go ahead.

Yes, thank you for taking my question. I guess my first question is on your MOCVD business for Indian Phosphide. Laser production. You know, I think you're kind of the low market share guy there. And could you talk about, you seem to have indicated some progress on the conference call about winning customers, broadening out your customer base. I'm shipping more tools to current customers. Could you just talk about, you know, what your goals are for market share of that business and kind of where you're at now? And, you know, what sort of growth would we expect for that part of the, for MOCVD in 2027?

Steve, I would say, as I think you know, we've developed the Lumina and the Lumina Plus product line and recently launched it. And what we're seeing in the marketplace, whether that's in indium phosphide lasers or space-based solar applications or micro-LEDs, the... The product is being broadly accepted, and it seems that it's a pretty capable product in the marketplace, and so we are winning business this year and growing that business at a number of fronts, but more specifically to your question, indium phosphide as well. And so we are working very closely with a few of the leaders. We're doing demos with others, and we're knocking on the doors of others to try to get in. And so I would say some of the challenges that we face is because this is a new product, ramping it to high volume will take a little bit of – higher volume will take a little bit of time. So this is an area where we are constrained a bit in terms of our growth in 27, but I would expect to start gaining a bit more share in kind of the late 27, 28, and beyond timeframes.

Okay. And then you mentioned LSA revenue in 2027, I think, is going to be a significant grower. Could you kind of highlight, you know, both for FoundryLogic and for Memory, you know, how much you think you can grow next year in relationship to the growth rate you've given for the segment, I guess?

Yeah, and Yeah, in LSA, in 26, our advanced logic business has really ticked up with the tier ones. And the other thing is we've won one memory customer for high bandwidth memory, and they've grown substantially. So our memory segment has grown substantially in 2020. 26. The headwind we have in 26 is that our China business, as we've been forecasting, continues to decline. So our mix of kind of front end, back, or leading edge, trailing edge has kind of flipped back to more traditional kind of two-thirds, one-third during 2026 for LSA. You know, when I look at this year, I would, I think that, I'm sorry, when I look at 27, I would say, if I'm looking specifically at LSA, I would say. That business will continue to grow in the advanced segment as it has in 26, and we are expecting decline in 27 in China. So we expect China to become less relevant. And when I look at the semi-segment, including advanced packaging and all the other elements, I would say our Semi-segment will grow at about 40%, similar to what I said earlier, company-wide average. That being said, David, I think it's important for everyone to understand that silicon photonics, where we are seeing significant growth, sits in our compound semi-segment, not in our semi-segment. So if you were to add the compound semi with the semi, it would be growing faster than... The company average.

Okay, and then I signed on just like a couple minutes late, and you're talking about doubling your capacity for advanced packaging, I believe. Um, and talked about a panel level when I could just was wondering, you know, kind of would you just take a step back, you know, what you kind of think, you know, I imagine your co-op business grew quite substantially here in 2026 with all the wafer starts that are being added, but there's a lot of wafer starts being added next year too. So maybe just talk, you know, parse the, what you kind of think the co-op of that packaging, you know, kind of revenue outlook is, and is doubling capacity enough to meet demand in 2027.

Yeah, I would say, Dave, as you know, in 25, we doubled the business on the back of an advanced packaging, really in our wet processing business. And in 26, and that was for a lot of co-op capacity that you're now seeing waper starts in 26. So that capacity was put in at 25. 426. That business continues at an elevated level at, you know, I would say it's probably flat-ish to maybe up a little bit in 26. But when I look at advanced packaging in 27, we're definitely seeing that business grow substantially based on our backlog, not only in wet processing, but also in lithography for heading towards doubling from these kind of elevated levels that we're operating at now. So to that end, I'm not sure how much you caught on the call, but we are adding two contract manufacturers. We've pretty much run out of. Space in our current factory. We're adding a contract manufacturer in Southeast Asia. Their first shipments will be coming online in Q1, and a domestic contract manufacturer that's been building sub-assemblies is now going to start building whole systems for us. But the majority, I would think, is heading towards Southeast Asia.

And then just inside that question was a commentary about panel. Yes, I'm sorry. Can you just talk about panel adoption?

Yeah, so we have shipped a panel tool to a leading foundry, and You know, just given the size of these packages and dye, that, you know, a lot more dye fit on a square than fit on a circle, effectively. And so, without a large change in Footprint, but geometry, you know, our customer can add substantially more output of devices. So that's been progressing. It's doing well, and we're – I think we're in a very strong position there for – in wet processing for that panel tool.

Thank you.

Thanks, Dave.

And your last question comes from Gus Richard from Northland. Please go ahead.

Yes, thanks for taking the questions. Just real quick on the IDD opportunity at the DRAM guys, could you talk a little bit about what you need to finish to get one of these customers over the line to order tools?

Yes, sure, Gus. As you know, we have two evals at leading DRAM customers. I would say we've been making a lot of progress. The team has been making a lot of progress this last quarter or so. Our feedback from the customer has always been, wow, these are very high quality films. The resistance is really spectacular. We really love the film quality. But what we've been able to do is to really make some substantial progress in demonstrating to our customers the robustness of the system, which is really kind of what's been holding us back. And so our expectation is we're going to expand our program in 2027, and what I mean by that is customers are pulling for additional tools now to support their pilot line, HVM, that we expect to ship in 2027. So we're kind of in early stages of discussion of that, and we're working through what that looks like and how we execute that.

Got it. And then you mentioned, you know, you've got an opportunity in Pellicles, you know, beyond your initial customers. Just wondering if you could talk about, you know, is that logic? Is it memory? You know, is it multiple customers? Is the existing customer going to order more? You know, sort of what's the state of that piece of the IBD business?

Yeah, so what we've said is we have one foundry customer who are in the pellicle business for their own business, and then what we're seeing is the ecosystem is such that A lot of customers don't want to make pellicles, and so they're working with their suppliers. And so now we're not only working with this foundry customer, but another level down in the supply chain to help them as well, to work with them as well, to support the other major customers. And I would say largely those are more logic customers today.

Okay, and just to clarify, that would be like a mask blank vendor or someone that makes pellicles for DQV?

Exactly, exactly. Existing, the existing supply chain, yes.

Okay, I got it. Thank you so much. Yep.

Thank you. At this time, we have no further questions. I would like to turn the call back over to Mr. Bill Miller for closing remarks.

Thank you. Vico delivered another solid quarter, and we continue to build momentum across our key growth areas. AI investment and increasing chip complexity are driving strong customer engagement, order momentum, and greater visibility into 2027 and beyond. We're investing to support this growth while advancing our differentiated technologies, including IBD300, to capitalize on our long-term growth opportunities. We also continue to progress toward completing our merger with Excellus and remain confident in the compelling value creation potential of the combination. We're poised to benefit from the significant tailwinds within the industry and remain focused on execution. Thanks to our shareholders for your continued support and to the entire VECO team for your hard work and dedication. Have a great evening.

Ladies and gentlemen, this concludes your conference call for today. We thank you very much for your participation, and you may now disconnect. Have a great day, everyone.