Kulicke and Soffa Industries - Earnings Call - Q3 2025

Executive Summary

• Q3 FY2025 revenue of $148.4M and non-GAAP EPS of $0.07 modestly beat S&P consensus ($145.5M revenue, $0.055 EPS), aided by stronger general semiconductor and memory utilization; GAAP EPS was $(0.06) as restructuring and equity compensation still weighed on GAAP results.*
• Gross margin expanded sharply to 46.7% from 24.9% in Q2, reflecting normalization post-EA wind-down charges and improved mix; operating expense normalized to ~$75.3M GAAP ($68.0M non-GAAP).
• Q4 FY2025 guidance: revenue ~$170M ±$10M, GAAP EPS ~$0.08 ±10%, non-GAAP EPS ~$0.22 ±10%, non-GAAP OpEx ~$68M; management expects sequential improvement, with December quarter (Q1 FY2026) “flattish” vs September given ongoing tariff-related caution.
• Strategic catalysts: momentum in Fluxless Thermal Compression (TCB) with first memory-system shipment targeted by end of calendar 2025, vertical wire adoption for low-power HBM/DDR applications, and broadening Advanced Dispense orders (auto OEM, IDMs, OSATs).

What Went Well and What Went Wrong

What Went Well

• Normalization of margins and execution: gross margin recovered to 46.7% and non-GAAP EPS reached $0.07; management highlighted “revenue above guidance” and disciplined cost control.
• Strategic product momentum: initial POs in Advanced Dispense from an auto OEM, several IDMs and multiple OSATs; expanding portfolio to be showcased at Semicon Taiwan.
• Advanced packaging traction: TCB progress with planned first memory shipment by year-end; confident roadmap and share gains across memory and logic applications.

What Went Wrong

• Automotive/industrial demand softness: sequential revenue pressure and lingering weakness driven by tariff/trade uncertainty and Southeast Asia-specific customer hesitation.
• GAAP profitability still negative: GAAP EPS of $(0.06) and GAAP operating loss $(6.1)M reflect non-core charges and equity compensation; non-GAAP profitability remains low (2.5% net margin).
• Limited capacity additions despite high utilization: customers running at elevated utilization (overall ~81%) but remain cautious on capex due to tariff uncertainty; auto utilization <70%.

Transcript

Speaker 6

Greetings and welcome to the Q3 2025 quarter results. At this time, all participants are in a listen-only mode. A question and answer session will follow the formal presentation. If anyone should require operator assistance, please press star zero on your telephone keypad. As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Joe Elgindy. Thank you. You may begin.

Speaker 0

Welcome everyone to Kulicke and Soffa Industries' fiscal third quarter 2025 conference call. Fusen Chen, President and Chief Executive Officer, and Lester Wong, Chief Financial Officer, are also joining on today's call. Non-GAAP financial measures referenced today should be considered in addition to, not as a substitute for, or in isolation from our GAAP financial information. GAAP to non-GAAP reconciliation tables are included within the latest earnings release and earnings presentation. Both are available at investor.kns.com, along with prepared remarks for today's call. In addition to historical statements, today's remarks will contain statements relating to future events and our future results. These statements are forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995 and are subject to risks and uncertainties that may cause our actual results and financial condition to differ materially from the statements made today.

For a complete discussion of the risks associated with Kulicke and Soffa Industries that could affect our future results and financial condition, please refer to our latest Form 10-K and upcoming SEC filings for additional information. With that said, I'd now like to turn the call over to Fusen Chen for the business overview. Please go ahead, Fusen.

Speaker 2

Thank you, Joe. Good morning, everyone. Over the coming quarters, we demand focus on extending our market access through ongoing technology transition, and we continue to be encouraged by gradual core market improvements. As discussed last quarter, uncertainty around global trade has clouded near-term industry visibility, although we continue to see steady core market improvement as expected. We do not expect near-term trade dynamics to materially affect our global business operations, although they do create additional uncertainty in customers' near-term capacity planning decisions. Despite this near-term headwind, we continue to remain cautiously optimistic as we continue to see ball bonder utilization improve in key regions. We also continue to work very closely with customers to support technology transition, including advanced dispense, vertical wire, and thermal compression bonding, or TCB, which I will provide an update on shortly.

For Q4, we generated revenue of $148.4 million, GAAP loss per share of $0.06, and a non-GAAP earnings per share of $0.07. Our focus on efficiency, customer engagement, and execution has allowed us to exceed expectations. Next, I will provide an update on financial performance and the outlook shortly. As expected, the sequential revenue reduction into the June quarter was largely driven by order hesitation within the automotive and the industrial market. We mentioned last quarter this was focused uniquely with certain customers' production facilities in Southeast Asia, and we anticipate this will persist through the September quarters. We anticipate this softness is largely driven by trade uncertainty. It's broadly affecting global automotive and the industrial supply chain, and it will create a slight headwind over the coming quarters. Regardless, this key market is supported by ongoing technology transition and above-average growth rate.

Despite the near-term, automotive-driven technology transition provides a long-term set of growing opportunities. For example, EV charging infrastructure is driving new equipment opportunities, and we anticipate charging-related infrastructure will exceed a 20% CAGR over the next five years. Additionally, this ongoing growth is also driving the need for smarter and more efficient power semiconductor applications, which we are addressing with our growing base of pinwelder, advanced dispense, and clip attach. Similar to automotive and the industrial, but less pronounced, order hesitation was also apparent within the general semiconductor end market during the June quarters. More recently, we are encouraged by seasonal and cyclical dynamics, which are improving ball bonder evaluation improvements within core regions, and we anticipate both leading and high-volume markets to improve through the September quarters. We remain very focused on both core market recovery and also new product momentum.

We also experienced strong sequential demand increases in memory and are encouraged by improving conditions, recent price dynamics, and the emerging packaging formats. We continue to be focused on driving share gain and expanding our reach into DRAM applications by enabling new packaging for capabilities in high volume with our vertical wire solution and within leading edge applications in future versions of HBM. Next, I would like to provide a brief status update on broader technology transitions we are addressing through our advanced dispense, vertical wire, and the TCB portfolio. First, with advanced dispense, we are continuing to seek opportunities across key customers and end markets while maintaining an aggressive product development pipeline. We have received initial POs from an automotive OEM, several IDMs, and multiple OSATs, which highlight the broader diversity of our solutions.

The need for higher precision and a more capable dispense system is broadening across end markets. We look forward to expanding our portfolio to support this need, and we plan to introduce new advanced dispense capabilities in September at Semicon Taiwan. Next, within vertical wire, market expectations remain on track. We continue to plan for initial high-volume productions to begin in fiscal 2026, driven initially by an exciting technology transition within the memory market. Emerging on-device AI applications are demanding higher bandwidth. This market need is driving demand for transistor-dense vertically stacked, low-power dynamic memories. We have observed market references such as mobile HBM, energy-efficient HBM, or low-power I/O DRAM applications, which describe this new opportunity. This new format of a low-power HBM is anticipated to increase bandwidth by three to four times over existing low-power DRAMs.

Vertical wire interconnects are enabling an alternative, more cost-effective production process for those lower-power HBM applications. As a reminder, higher-power data center HBM utilizes more costly through silicon via diodes and thermal compression-based assemblies. We anticipate this new vertical wire-based HBM will be adopted in broader DRAM applications and eventually support higher transistor density requirements across broader general semiconductor applications. Finally, within TCB, we continue to focus on many different applications for both the logic and the memory. Our FluxLabs thermal compression, or FTC solution, continues to be best in class and is increasingly positioned to seamlessly integrate into a variety of applications and customers' production flows. We have recently demonstrated new physical and chemical-based inline material preparation capabilities, which further extend our leading FTC process. We are very proud of how our existing chemical-based solutions have performed, which have enabled us to lead the initial market transition to FluxLabs.

This solution has allowed us to be first to high-volume production, and we are currently supporting multiple large customers in mass production. With that said, we are now pleased to offer a tailored mix of physical and chemical-based processes within one solution to best suit the widest variety of customer process flows and market applications by adding inline die and wafer preparation capabilities to our leading chemical-based process. Initial customer feedback has been positive, and we expect this new capability to be a market enabler, which lowers barriers to entry as customers initiate new production or expand their FTC capabilities. In addition to this new FTC solution, we have also made progress within the high-power HBM market. We now expect to ship an initial FTC system by the end of calendar year 2025 to support the anticipated FluxLabs transition within the HBM space.

As we increase focus on emerging memory opportunities, we are confident our proven leadership in driving FTC adoption within leading edge logic applications provides a unique advantage to support leading memory applications as they transition to fine pitch FTC-based assemblies. We are confident we have the most robust and capable solution for leading edge thermal compression applications and remain positive on our engagement, recent progress, and long-term roadmap of this highly capable technology. We expect FTC solutions will outpace overall TCB growth, allowing us to extend our market shares over the coming years within both the memory and the logic markets. In summary, we continue to expand our market presence on multiple fronts and remain cautiously optimistic as key regions and the end market show signs of cyclical improvement.

While automotive headwinds are anticipated to linger into the September quarters, general semiconductor capacity digestion and expansion, driven by China and Taiwan, as well as memory technology transition and price improvements, are increasing our confidence in the outlook. It's continuing to be a unique time in semiconductor assembly with a wide set of opportunities to be addressed. I look forward to updating our progress over the coming quarters, and I will now turn the call over to Lester to discuss the financial and outlook. Lester?

Speaker 5

Thank you, Fusen. My remarks today will refer to GAAP results unless noted. Despite uncertainty regarding macro dynamics, we delivered revenue above guidance, continued to execute on those customer engagements, and maintained an ongoing focus on cost control. Gross margin came in at 46.7%, and we delivered $0.07 of non-GAAP earnings per share. Total operating expense came in at $75.3 million on a GAAP basis and $68 million on a non-GAAP basis. Tax expense came in at $3.2 million, and we continue to anticipate our effective tax rate will remain above 20% over the near term. We are also positive on recent U.S. tax law changes and are in the process of evaluating the future impact on our tax rate. We have continued to opportunistically reduce shares outstanding by repurchasing approximately 668,000 shares, equivalent to 1.3% of diluted shares during the June quarter.

Since the first fiscal quarter of 2024, we have taken a more aggressive approach to delivering shareholder value directly. Over this seven-quarter period, we deployed over $270 million in dividend payments and open market repurchase activities. This has allowed us to repurchase over 5 million shares, nearly 10% of shares outstanding, and currently offers a best-in-class dividend yield above 2%. Looking ahead, as Fusen mentioned, we are more positive on key end market improvements in general semiconductor and memory supported by regional utilization and pricing improvements. Automotive and industrial headwinds will likely persist over the coming months, although we anticipate this softness to be short-term. For the September quarter, revenue is expected to increase by approximately 15% sequentially to $170 million, with gross margins at 47%.

Non-GAAP operating expenses are expected to be $68 million, with GAAP earnings per share targeted to be $0.08 and non-GAAP earnings per share of $0.22. We are very focused to exit from this extended period a leaner and more growth-oriented organization. Today, we're either an incumbent leader or are aggressively taking significant market share in every key market served. We continue to ensure our highest potential opportunities are adequately resourced, and our customer development efforts are on a positive trajectory. Although we remain cautiously optimistic as we look into fiscal 2026, end market visibility remains unclear. We continue to anticipate gradual recovery, with some potential minor seasonality anticipated in the December quarter. We look forward to demonstrating progress on vertical wire, advanced dispense, and thermal compression opportunities as we prepare for the broader core market recovery. K&S is very focused on executing our strategic priorities.

We are confident with our capabilities and technology leadership, and we remain well prepared to navigate the near-term macro environment. This concludes our prepared comments. Operator, please open the call for questions.

Speaker 6

Thank you. We will now be conducting a question and answer session. If you would like to ask a question, please press star one on your telephone keypad. A confirmation tone will indicate your line is in the question queue. You may press star two to remove yourself from the queue. For participants using speaker equipment, it may be necessary to pick up the handset before pressing the star keys. One moment, please, while we pull for questions. Our first question comes from the line of Krish Sankar with TD Cowen. Please proceed with your question.

Speaker 1

Hi, thanks for taking my question, and nice to see sequential revenue improvements. I have three questions, Fusen or Lester. Number one, I think you kind of mentioned about FY2026 gradual recovery, minor seasonality. Is it fair to assume December quarter revenues have to be down sequentially and then we grow from there, but are we going to be bouncing around these $170 million levels for a few quarters or how to think about it?

Speaker 2

Okay, you asked three questions, so we do it one by one, right? First question is, I think at this moment, the industry utilization rate is quite healthy. Actually, it's a benefit improvement, the general semiconductor. The memory, I think, is also quite strong. We're still facing the ongoing headwind in the automotive. The Q3 revenue we just published, and we got $170 million for Q4. In our review, we do believe probably Q1 of 2026, first half always is weaker. Because we are high utilization, right now we feel, I think, Q1 of 2026 will be flat, flat-ish. I hope I answer your questions. For the 2026, again, 2026, we do believe it's in a better cyclical position with a potential significant improvement in the industry utilization rate and for the ball bonder. We also have new products, three new products. For example, clip attach, wire bonder, pinwelder.

This is for the wedge bonder family. For high-power semiconductor, we also have improvements of advanced dispense, vertical wire, and TCB growth. TCB growth was due to industry FluxLabs technology transition and potential again win in the HBM. We do believe probably the first half of 2026 will be flat to Q4 of our 2025. This industry already has a downturn facing the third quarter. We do believe Q3 and Q4 will be up. I hope I answered your questions.

Speaker 1

Yeah, that was really helpful, Fusen. Around $170 million for the next couple of quarters. I have two other quick questions. The second one is, I'm just kind of curious how to think about the impact of Intel CapEx cuts, because you do have some pretty good FluxLabs position there, some of the copper-to-copper, et cetera. I'm just wondering, is that a headwind or have you seen any progress there, or do you think that potential opportunity in the future for FTC is probably minimized with Intel? I have one last follow-up.

Speaker 2

I think the engagements are still quite healthy, but I can only answer the, of course, the revenue compared to our previous year will be down. That's all I can say.

Speaker 1

Got it, got it. The last final question, you said you're going to be shipping TCB for HBM end of this calendar year. Is this for HBM4 or is this just one customer or how to think about it?

Speaker 2

Okay, we do believe the next version of HBM is going to be FluxLabs will play a very important role. Currently, I think we have two customers we are working with, but one of the customers I think we will intend to ship a system end of calendar year 2025.

Speaker 1

Thank you very much, Fusen. Appreciate that.

Speaker 2

Okay, thank you. Thank you, Chris.

Speaker 6

Thank you. Our next question comes from the line of Tom Diffely with D.A. Davidson. Please proceed with your question.

Speaker 3

Thank you for taking the questions. I'm curious, you know, what are the utilization rates this quarter at your kind of core customers? What are the areas that's driving this incremental growth quarter over quarter here in the fiscal fourth quarter?

Speaker 5

Hi, Tom. It's Lester. Utilization rate overall is about 81%. I think in the end markets, general semi is around 83%. Memory is about 80%. We see growth in utilization in all the major end markets, even auto, even though auto is still quite soft. Utilization is below 70% there. As far as what's driving it, it's basically general semi as well as memory. They've shown some of their highest rising utilization rates. I think that's basically what's driving particularly general semi, our sequential growth in the revenues.

Speaker 3

Great. Thank you. Fusen, I was hoping you could just dig a little deeper into the thermal compression business. You know, how big was it historically with FluxLabs and going into memory going forward? How big could it be, you know, a year or two down the road?

Speaker 2

Okay, I will repeat a few times already. This year, previous quarter, you know, we targeted 25, you know, as a shipping. You know, we are talking about, you know, 60 to 70. Next year, I think originally we target $100 million. With the FluxLabs transition, and we really work hard on HBM, you know, we believe we will have some upside. As the market grows, our target actually, the memory market actually is higher, bigger than logic right now. We intend, 2028, we think the market will reach about $1 billion. We actually plan, we have a plan and target about $250 to $300 million in 2028. We believe at that time the market will have sized about $900 to $1 billion.

Speaker 3

Okay. Is the main advantage here on the cost side? You mentioned that obviously you don't do three silicon vias with this technology. Is it really for lower cost applications where you are benefiting the customer, or is there a technology advantage as well?

Speaker 2

I think it's really technology. We actually start with chemical-based. There are two technologies, and we are the leader. Actually, we can tell you at this moment, all the high-volume production, high volume, actually, we are the only technology in the high-volume production. As you mentioned, as I mentioned in my script, we are adding more capabilities. We are quite positive about the transition to FluxLabs. We are number one, and we intend to capture big markets. After focusing on logic, we will focus on as well as HBM from here.

Speaker 3

Great. I appreciate the extra color.

Speaker 2

Thank you.

Speaker 6

Thank you. Our next question comes from the line of Charles Shi with Needham & Company. Please proceed with your question.

Speaker 4

Hi, Fusen and Lester. Thanks for taking my question. The first one, I want to get a better sense on your soft guidance on December quarter, meaning you don't have an official guide, but you did provide the direction. This December quarter, how do I think about the confidence level you have? Is it that you're basically guiding December maybe flattish? Is it based on orders on hand, or is it based on forecasts that maybe you don't have all the orders on hand, but things may be coming along nicely? I want to gauge a little bit of your confidence there. Thank you.

Speaker 5

Hi, Charles. It's Lester. It's a combination of the things you mentioned, right? I mean, Fusen already mentioned earlier about utilization rates being quite high, right? I think over the last couple of quarters, we've seen utilization rates at a healthy level, and it's increasing. Obviously, in normalized times, there would have been a ramp ready. As we've mentioned several times, the tariffs are kind of creating a little bit of hesitation in our customer supply chain. As far as our confidence on why we think that, again, we're not guiding, but as Fusen said, we think December quarter will be flat to our Q4, which is our September quarter. The reason is, yes, we do see higher order intake coming in. We do see a lot more increase from customers.

We're seeing, again, recovery in both end markets as well as regions that were a little slow over the last couple of quarters, right? I think a combination of all those things makes us think that even with, usually in December quarter, we have a little bit of seasonality, but we believe that with all these factors, it should be, again, relatively flat from the September quarter.

Speaker 4

Thanks, Lester. Maybe a second question. I want to ask a little bit more about the TCB. Maybe this is more about technology. Fusen, I think you mentioned about preparation, physical or chemical preparation, but kind of wanted to ask you what that means. Is it in situ preparation, or is it a separate tool? I think related to that, there has been a debate on TCB, the two approaches, the chemical approach, which is the approach you guys have, and the plasma approach. Are you, by saying physical preparation, talking more about the plasma approach you're developing, or what is that? On HBM, high bandwidth memory, which technology you are proposing or you are going to evaluate with customers? Thank you.

Speaker 2

Okay. As I mentioned, the industry, you know, we have eight customers, and they are all in production. The whole industry, the only technology in high volume production actually is chemical-based, right? We are very proud. All the customers we have, the most production actually is chemical. The end-to-end situation actually is we are also developing physical preparation technology. This is really very old technology, probably tens of years prior in a semiconductor fab. Actually, both technologies have a marry. I'll give you an example. I think for the physical one, you need to clean like soda oxide, actually have a faster rate. For the chemical, we are very confident. I think for the surface preparation, especially interface, interface integrity quality is the most important for the year. We are strongly confident. I think the chemical one is the best. Right now, to answer your question, we have a capability.

We have done it and demonstrated it and get the feedback. It's positive. We do it in situ all together, integrate all this capability together in a system. We're doing that. Our customer is so convenient to easily integrate their integration and enable us to actually fight for the market shares in various kinds of chips, on wafer, on substrate, and also maybe in the future, also HBM. Hope I answered your questions.

Speaker 4

Thank you. That's all from me. Thanks.

Speaker 6

Thank you. As a reminder, if anyone has any questions, you may press star one on your telephone keypad to join the queue. Our next question comes from the line of Christian Suab with Craig Hallam. Please proceed with your question.

Speaker 1

Great. Thanks for taking my question. Fusen, at the end of the prepared comments, you talked about being well-positioned, taking market share. Can you elaborate on what products that you plan on gaining market share, auto, industrial, memory, et cetera? Just give a more clear answer to that, please.

Speaker 2

Okay. The first one I mentioned actually is called clip attach and also pinwelder. This is in our wedge bonder family. This is especially for high-power semiconductor products, right? This will be very, very important. High-power semiconductor will be very, very important to show significant growth in the next many, many years. The second one is an advanced dispense solution. Dispense actually is quite a big market. It's like $1 billion to $2 billion. We are focused on appreciation, you know, dispense. Actually, three years ago, we acquired a company, and we do believe technology is differentiated and has a micro-dispensing capability with inline inspection and also has a good availability, right? We put this one actually in the common platform of our company, common platform. We do believe our cost is very competitive and have differential products. We will show actually the growth next year over here.

Then vertical wire, I think we talked about vertical wire actually probably close to five, six quarters. I think finally this will take off. When I mentioned, I mentioned this product will be implemented into a DDR5, you know, SP game. When people do this, a lot of people are encouraged because it's a stacked DRAM die, and the bandwidth is actually not like a high-power, you know, HBM. It has a much wider bandwidth. This one can actually provide 3 to 4x of a bandwidth increase compared to our current DDRs. The last one I mentioned, I think TCB, I think a lot of discussion, and we are quite focused, and we do believe we can make a traction on TCB.

Speaker 1

Great. Fantastic. Thank you. No other questions.

Speaker 6

Thank you. We have reached the end of the question and answer session. Therefore, I'll turn the call floor back to Joe Elgindy for closing remarks.

Speaker 0

Thank you, Shubneesh, and thank you all for joining today's call. Over the coming quarter, we'll be presenting at several conferences. As always, please feel free to follow up directly with any additional questions. This concludes today's call. Have a great day, everyone.