Soitec SA (SOI) Earnings Call Transcript & Summary

Mr. Pierre Barnabe, CEO, to begin today's conference. Thank you.

Thank you. Good morning all, and welcome to Soitec's conference call dedicated to the publication of the revenue for the first quarter of our fiscal year 2026. This is a quarter covering the period from April to June 2025. I'm Pierre Barnabe, Soitec's CEO. Together with me on this call are Albin Jacquemont, our CFO; and Steve Babureck, our EVP, Strategy. I would like to take a moment to welcome Albin to Soitec. Albin has been with us for 2 months now. He brings his extensive experience across complex industrial and technology environment. He has a proven track record in financial transformation, working capital optimization and value creation. He is already making an impact, and I'm confident his leadership will further strengthen our financial foundation. Albin, would you like to say a few words?

Sure. Thank you very much, Pierre, for the kind words, and good morning to everyone. Very pleased to speak with you. Thank you for lining up on this call on what is a busy day today in Paris. First thing to say is, I'm very honored to join Soitec and very eager to take on this new challenge. As you may have seen, I have 30 years plus of experience in both listed companies and companies under private equity ownership. These include Carrefour, Darty, Altran, Saur and most recently, Inetum, a European leader in digital services with a team of 28,000 consultants across 19 countries, generating EUR 2.5 billion in revenue. After a couple of months with Soitec, I am already very impressed by the dedication of the teams and their commitment to preparing for the next phase of Soitec's expansion, planning well beyond the current temporary weakness of some of our end markets. If there is one thing you can be sure of, it is that the finance team will work hard and actively contribute to support these ambitions. I'm looking forward to having the opportunity to meet with each one of you.

Thank you, Albin. Back now to Q1 revenue. And as usual, we'll start with a few comments on our figures. And after that, we'll open the floor to questions. Before we jump into our numbers, let me start with some general comments. At the macro level, we continue to see rising cautions from our customers, driven by increased volatility in end markets, as well as uncertainties around tariffs and geopolitical risk. What does that mean for us? In mobile, the RF-SOI customer inventory situation remains difficult. Inventory levels at foundries are still high with global smartphone sales forecast revised down from around 3% growth at the start of the year to around 1% today. Customers are focused on reducing excess of inventory. We are supporting our customers and proactively managing inventories as well as our own production. The objective is to restore a healthier balance between supply and demand. This should help reinforce our long-term competitive positioning. In automotive, while we start to hear some positive comments among our peers, the situation remains challenging. There is some excess inventory in the value chain through to a lesser extent than in RF-SOI, which is impacting Power-SOI sales. Beyond these short-term headwinds, our fundamentals remain very robust. Our segment shares are intact, and we maintain strong leadership across our core product families. Four technologies are now generating around or more than $100 million annually, up from just 1 3 years ago. And Photonics-SOI should get there soon. Our diversification is now firmly established. Our broader product mix is proving its strength across varying market cycles. Our innovation engine continues, remains active and focused. We are expanding the reach of FD-SOI and POI with new design wins and accelerating Photonics-SOI adoption with all key players, particularly in AI. We are embedding operational excellence across the company, optimizing fab loading, adapting CapEx and driving efficiency while protecting our ability to invest in R&D and scale as markets recover. And we continue to build and strengthen our strategic partnership across the value chain, building technology road maps with key decision-makers across our end markets. These strengths give us the confidence and agility to navigate the short-term headwinds while staying fully focused on long-term value creation. Now let's drive (sic) [ dive ] into our Q1 numbers. Following a strong Q4 '25 driven by seasonal inventory replenishment, Q1 '26 revenue pulled back, down 16% year-on-year on an organic basis, slightly better than our guidance. Our EUR 92 million revenue reflects a 16% decline at constant scope and FX, a negative currency impact at 5% and a negative scope effect of minus 3%, which is related to the divestment of Dolphin's Design business in the later part of calendar year 2024. As expected, Q1 revenue declined due to a combination of RF-SOI inventory correction, ongoing weakness in the automotive market and the planned phaseout of our Imager-SOI products. These dynamics were offset by continued strength in cloud infrastructure and AI-driven IoT application. Let's start with Mobile Communications, which generated EUR 43 million in revenue, down 7% year-on-year on an organic basis. Following a strong Q4 '25 driven by seasonal restocking, RF-SOI sales declined below Q1 '25 levels as customers continued adjusting inventories. 200-millimeter RF-SOI volumes fell year-on-year, reflecting that correction. 300-millimeter RF-SOI volumes increased through tempered by a slightly negative price and mix effect. POI wafer sales were stable with strong momentum in the U.S. and the temporary slowdown in Asia. POI is increasingly becoming the reference substrate for advanced surface acoustic wave filters. Finally, FD-SOI sales for 5G millimeter SoCs were significantly higher than in Q1 '25, reflecting growing adoption across key applications. FD-SOI adoption is progressing with first design wins for Wi-Fi 7 SoCs in premium Android smartphones. Moving on to Automotive & Industrial revenue fell to EUR 5 million, an 81% decline year-on-year on an organic basis. As anticipated, the Power-SOI inventory replenishment in Q4 '25 pulled demand forward, resulting in a very low Q1 and expected continued pressure in 2025. Meanwhile, we are accelerating the transition from 200- to 300-millimeter Power-SOI to support growing demand for Battery Management System. Automotive FD-SOI sales remained low, but ecosystem development continues to progress, supporting future adoptions across radar, microcontrollers and wireless connectivity. On SmartSiC, the combination of slower EV market dynamics and extended qualification cycles continues to delay the production ramp-up as previously communicated. Beyond automotive, we are seeing early signs of interest in adjacent applications, such as AI infrastructure and renewable energy, where SmartSiC could bring meaningful performance benefits. We are in the process of assessing the real events and timing of these opportunities. Finally, in Edge & Cloud AI, revenue reached EUR 44 million, up 13% organically despite the anticipated drop from the phaseout of Imager-SOI, which contributed significantly in Q1 '25. The momentum for our Edge & Cloud AI division remains very strong, driven by artificial intelligence with traction both at the Edge for low-power computing devices and Edge AI application and in the Cloud for data centers. Photonics-SOI posted another strong quarter as investment in AI data infrastructures accelerate. The technology is now a key enabler for high-speed, high-bandwidth optical links, including pluggable transceivers and CoPackaged Optics. We continue to strengthen partnerships with major AI and cloud players, as well as we advance our road map. FD-SOI sales were also above Q1 '25, reinforcing its role as the preferred platforms for AI-driven IoT applications requiring power efficiency, reliability and thermal performance. That wraps up our Q1 '25 review. Let's now turn to the outlook for Q2 '26. Looking ahead, we expect Q2 '26 revenue to grow by around 50% sequentially on an organic basis. This gradual improvement still represents a notable year-on-year decline versus Q2 '25, mainly due to ongoing RF-SOI inventory correction, continued weakness in automotive and the planned phaseout of our Imager-SOI product. Mobile Communications revenue will remain low despite nearly doubling from Q1 '26 as customers continue to work through excess RF-SOI inventory. Automotive & Industrial is expected to decline sharply year-on-year as end market demand remains soft and Q2 still reflects the pull forward of Power-SOI volumes from Q4 '25. While some analog players are starting to signal early sign of recovery inventory level across our customers base remain elevated, but not to the same extent as what we experienced with RF-SOI. Excluding Imager-SOI, Edge & Cloud AI is expected to maintain solid momentum and grow slightly versus Q1 '26, driven by continued strength in Photonics-SOI. The impact from the Imager-SOI phaseout will be less significant than in Q1 as it contributed around $7 million in Q2 last year. Finally, we confirm our fiscal year '26 CapEx cash out at around EUR 150 million, down from EUR 230 million in fiscal year '25. We remain disciplined on cost to protect our margins and sustain our investment in innovation and R&D. This ends my opening remarks. Thank you for your attention. We are now ready to take your questions.

[Operator Instructions] The first question comes from the line of Aleksander Peterc calling from Bernstein.

I'd just like you to come back a little bit more on the momentum among your 3 verticals in the second quarter. You mentioned that the Edge & Cloud AI would be up slightly sequentially. How about the other 2 divisions? Is automotive remaining as low as it was in the first quarter? And then kind of a follow-up to that would be you had nearly 80% sequential growth last year, Q2 on Q1. Why is this year so much worse at 50%, which means that your year-on-year declines are actually accelerating deeper into negative? So if you could give us a little bit of background on that, that would be helpful.

On automotive, clearly, we're going to be in the same territory as for Q1 due to Power-SOI strong orders in calendar Q1, meaning that there is no specific needs looking at the situation of the market in automotive for expecting a better Q2 than the Q1 we have experienced in automotive. As you understood, FD is for the moment, quite small and the SmartSiC is still in adoption, then there is a quite low Q2 expected for this time. On mobile, there is a clear cautiousness from our customers on RF-SOI, not compensated by FD-SOI or POI. And this cautiousness is making our customers accelerating the depleting of the inventories they have in excess. That's the reason why the decrease in RF-SOI is stronger than what we would have experienced last year, meaning that there is a low level of automotive clearly and accelerating in inventory depletion by our customers because of volatility of the market and instability of the market.

And so, I mean, to understand why this year is so much worse than last year?

Well, if we look at the level of inventories, they have enough to face a market that is finally -- is that going to be flattish this year compared to last year with 5G level of maturity with the content they're going to grow, but as you know, very, very slightly. And no specific change in these metrics. As you know, we are expecting the AI revolution to make phones -- existing phones obsolete. It is not coming. Then we have to face today a flattish market in smartphones in many domains. That's the reason why our customers are remaining very cautious.

Could you give us an idea of where your floor revenue would be? Because you now have -- you're going to have 5 business units that are at EUR 100 million, so that's EUR 500 million. Could you sketch out kind of a floor, an absolute floor of revenue? We're getting quite close to that now. So that would be helpful. And then just a housekeeping one. How much was inventory SOI in Q1 last year?

Well, for the inventory SOI, I will leave the floor to Steve to complement because we really enter into deep details today on the inventories, having accelerated the discussions and the talks with our customers. And Steve, perhaps you can give some precise flavors on how we are.

Yes. Regarding the inventory situation in mobile and the business outlook for RF-SOI, we told you in May that our customers, our foundry customers at the end of calendar year '24 were holding about 14 months' worth of inventories. And the market last year for RF-SOI, the addressable market for RF-SOI was around 2 million wafers in 200-millimeter equivalent. Going into calendar year '25, we were expecting, as we told you in May, that this level of inventories would fall with our foundry customers at around 11 months. We were anticipating a smartphone market in volume to grow in calendar year '25 at around 3%, 3% to 4%. That's what we told you at Mobile World Congress back in March. Since then, the outlook has deteriorated a little bit. We're now expecting between 0% to 1% volume growth for the total smartphone market in calendar year 2025. And at the same time, as Pierre mentioned during his introduction remarks, customers across the value chain, it's true for mobile, it's true for automotive are more and more cautious regarding the inventories that they're holding on their books, cash is king. So net-net, we will see. We see that there is a short-term impact on the RF-SOI business. We will see how we end at the end of this calendar year '25. But clearly, the inventories are going down in RF-SOI for our customer, and that's the positive signal we can communicate at the present time.

Okay. On the Imager in Q1 last year?

Well, on the Imager Q1 last year, as we said, we were generating $25 million last year in Q1, and we have generated $7 million in Q2, then this $32 million is going to go to 0, of course, for this H1.

The next question comes from the line of Robert Sanders calling from Deutsche Bank.

Maybe you could just talk about your costs and your headcount. Obviously, you were scaling the business for a $2 billion medium-term sales ambition. That looks today a long way off. And you've obviously underestimated significantly the degree to which you have effectively overfilled the channel of customers with wafers. So what are you doing to resize the cost base to protect your profitability? Because as I see it, you're going to make a net loss in the first half. So it would be good to hear about what you're doing in this area.

Then, of course, we have not -- we are working on cost containment and reduction since we have been working for 3 years already. We have reduced -- we have limited, first of all, the hirings. And you're right, there were a trajectory 3, 4 years ago to reach $2 billion revenue in a quite short period of time. Rapidly after the drop -- dramatic drop of smartphone markets, we put in place some stringent discipline in cost, limiting, of course, SG&A, controlling the R&D investment while increasing it because, of course, we believe that our future is in the innovation, getting also additional subsidies and also taking care of our CapEx. We have reduced from EUR 280 million, EUR 230 million and now EUR 150 million. Then we are continuously reducing and containing our cost to be sure that we are following the right trends. Then looking at the headcount, we're going to continue to imply the same discipline and having a rigorous view on what do we need for what we have to do in H1. But we need also to prepare the future and the future after the depletion of the inventories in RF. And after -- and if we look at some encouraging signals for the future on automotive, we need also to be prepared for that, while we need also to accompany the growth of the other products, Photonics, FD, POI and preparing next-generation products we are cooking in innovation. Then, of course, we are really taking care of our cost. On a weekly basis, we look at what we can afford, what -- where we can reduce when we can cut. And we have protected so far the profitability by implying from the right beginning this discipline. Then we're going to continue. If we need to accelerate or if we need to go further, we will go further. But we're being sure that we will be -- we will have enough muscle to rebound when the products that are suffering today going to come back. That's very important to take that. But if you look at what we made, what we have done over the last years, you can count on us, and Albin is totally listening to that and totally aware and already engaged to have this discipline continuously and increasing it if necessary to protect the margin and also to protect the cash. Discipline is on the cost, but also discipline on the working capital to be sure that we have enough, let's say, reflections to what's going on, on the top line and enough gear and fuel to rebound or energy to rebound when it's going to come back to growth rate.

Got it. And just one quick follow-up. Just on reflection, when you look back to the $2 billion revenue ambition and what you're going to do now, which is probably 40%, 50% below that. What was the biggest delta? Was it over -- was it the millimeter-wave attach rate that you were overly optimistic on? Was it price erosion? Was it the smartphone market, unit volume estimates behind that $2 billion estimate? What was the kind of modeling error, if you want?

Well, to make a very long story short, first of all, keep in mind that we have an operating model that could drive us to double our revenue, as you know, in terms of structures of our manufacturing ability in terms of, let's say, drivers, in terms of customer relationships, product engagement and so on. We have an operating model that is still accurate to double our revenue in the future. And we keep, of course, active this operating model. To make a long story short, we had tailwinds, headwinds. It's a mix of, let's say, lower market than expected, particularly mobile. Remember the minus -- the big drop of volumes of smartphone in '22, 11% then again 7% drop in '23. That was not expected. You're right, the millimeter-wave was expected to go beyond the 13%, 14% market share we have today. Then there were other hypothesis that didn't come. But on the other side, we managed to make POI taking off a bit better than expected, to make Photonics becoming a standard for supporting AI development and also to create 4 incubators to prepare the future. Then it's also a way to be flexible, to be agile and in front of headwinds to react and to create other opportunities to face these difficulties. As Alex was mentioning, we have today 5 products in the family in the range of $100 million plus. This is something that is, of course, giving us some stable foundations. And now we need to work on fostering the products that are growing by continuously working with our customers and, let's say, managing properly the declining products, particularly by depleting as soon as possible the excess of inventories for RF and Power namely.

The next question comes from the line of Sébastien Sztabowicz calling from Kepler Cheuvreux.

How do you see your margins trending in H1 based on your guidance for Q2 and the revenue already achieved in the first quarter, you will have revenue declining 25% organically, if I'm correct? How do you see your EBITDA margin trending in the first part of the year? And the second question is on the POI, just to understand a little bit what is happening really in China business is slowing down? And how long will it take to go back to growth in China? How do you see POI trending in the next few quarters?

Albin, you take the first question.

Yes, I'll take the first margin. Great to speak with you, Sébastien. And I think it's a very good question. When we look at where we are, we have to strike a balance between working capital, revenue and cost. And so we're working a tight rope. Like Pierre said, the priority is on working capital. And within working capital, it is on inventory. And that's the main reason we don't guide for margin on H1. That being said, you know the dynamics which are at play. We -- the company has told you that within our process costs, 30% are fixed. So it's -- I think that would be very helpful for you to run the operating model. But the company said that it would guide on a quarterly basis on revenue back in May. And for the time being, we will stand to -- we'll keep on this.

Thank you, Albin. On POI, Séb, the situation in China is just temporary. And clearly, the adoption of POI is increasing. There are some cycle of design wins and new next-generation smartphones that are different compared to what we are experiencing in Western world. Then it's just a temporary pause, but we see, of course, POI growing in the future without any doubt. And this pause is compensated by a good traction in the Western world where we see an increasing adoptions by customers. And the good news that on POI, we are working on getting LTAs to secure business productions in the future with Chinese and Western world customers. Then POI on the short-, mid- and long-term is a promising product that is becoming a standard.

And how many customers do you have today in volume production and in, I would say, qualification at the end of first quarter?

Yes. Then we have 10 customers today. And we are in a qualification between 13 to 14 customers under qualifications today on POI. The main customers are in China and the qualified or under qualification customers are mainly in the Western world.

The next question comes from the line of Emmanuel Matot calling from ODDO BHF.

First, on CapEx, why have you decided not to further cut your budget for this year at a time demand is even weaker than expected? And second, are there any new developments in relation to the tax assessment procedure for more than EUR 300 million? What are the next steps with the French authorities? And my last question, could we have the level of the interest rate of your new EUR 200 million Schuldschein loan?

Emmanuel, I'm very happy to leave the floor to Albin for these 3 questions.

Thank you very much, Pierre. And thank you, Emmanuel, for bringing these 3 topics up. In terms of capital expenditures, so today, we keep EUR 150 million cash out for the year. There are 2 things to add on this. We do review -- we are reviewing our capital expenditures for the year, but it will not translate into significantly lower cash out this year because there is a time lag between when the CapEx is decided, the CapEx is recorded and the CapEx is paid. So you may infer from what I'm saying that capital expenditures for the next fiscal year would be lower if we decide to curb CapEx. But it's a great point, which is at the forefront of our discussion. That's for CapEx. For the tax audit, I have a couple of things to say. It's -- yes, it's a very significant amount. I was not in the company at the time. And I think it puts me in a good position to say that I'm comfortable with the substantiated response the company has made back in February and which -- and we got additional comfort from independent third-party experts. So I think we are on a strong footing. And then we will engage in the coming months with the French tax authorities, and try to strike a settlement, an amicable settlement. And for Schuldschein, Schuldschein, so we raised EUR 200 million. Average tenure is 4.1 years. The rate is based on Euribor 6 months to at 2.05% and the margin for the margin, you can factor 200 basis points. And of course, the Schuldschein is not secured. So it's a very flexible and convenient way of financing the company, and we are pleased with the response we got from the lenders we reached out to.

The next question comes from the line of Adithya Metuku calling from HSBC.

Firstly, just on the mobile inventory correction. I'm just surprised at the length of time this is going on for. I just wondered if there is anything structural changing with the amount of substrates needed for phones, maybe your customers are becoming more efficient. And so there is less square millimeters needed than previously -- than you previously expected. Is there anything like that, that we need to be aware of when we think about demand going forward? That's my first question. And I've got a couple of others.

Adithya, this is Steve. I'll take this question. On the fundamentals, no big change. If you open the phones, especially flagship products, you can see that the technology share of RF-SOI remains extremely high, number one. Number two, our segment share in RF-SOI is also very high, between 70% to 80% depending on the product line. And our innovation engine puts us in a very good position when implementing new solutions for our customers. We've made a few announcements over the last 12 months with UMC, Tower, GlobalFoundries. Then, of course, the smartphone market has been disappointing. Clearly, Pierre was reminding us the down years that we've had in 2022 and 2023. This year, again, we thought the industry was expecting 3%, 4% volume growth. And in fact, it's going to be flattish. So the volumes have been disappointing. There's a bit of shrink in the die size as we -- some of our customers are moving from 200-millimeter to 300-millimeter. That is maybe eating up a few percentage points of content increase. But overall, yes, we can only agree with you that it's taking a long time. Clearly, through COVID and post-COVID, the supply chain overpurchased parts all across the supply chain. And on the wafer side, it's taking a while to digest. Note that this situation is not unique to Soitec and to RF-SOI. You see that just in bulk silicon wafers in other areas. But yes, it's taking a while. The good news is that, it continues to go in the right direction.

Understood. And maybe just following on from that, your quarterly volatility in your business has increased a lot recently. If your auto revenues, they're down 80%, they've grown in the double digits previously. So just at this level of volatility, it becomes a bit difficult for investors to think about where estimates for the year might end up, especially as you've now stopped giving full year guide. And I'm sure you've seen your price-to-book ratios, which were at the lowest level since 2015. So I just wondered maybe for the new CFO, how are you thinking about full year guidance going forward? Are you planning to reinstate that maybe with a bigger range just to give people some idea of where numbers might go and where your assessment of the inventories in the channel are, et cetera? And finally, I've got a housekeeping question just on the hedging of the FX exposures. If you could give us the latest details on hedging, that would be helpful.

Yes. No. So we said that we will not guide for the year, and we will remain on that stance for now. And there's several reasons behind this. First, the markets are choppy. So we prefer not to guide if there is -- if the visibility is not good enough. But the main reason behind this is that, like I said, the priority that Pierre and the Board gave to me is to work on the working capital, reduce costs, reduce inventories and enhance cash generation. And we are working on that. We have made much progress. And I think the company is very -- is already mature in terms of thinking as to what it will do in the coming days, not weeks, days. And what we will do may impact EBITDA and the EBIT? So in the normal course of business, we would be happy to guide on the margin, but the priority for now is working capital. So this is the main reason, yes, and the cash, of course, is the priority.

Most priority.

Understood. And just on the FX exposure, if you could give us the details?

Yes. It's also a very good -- a very fair question given the trend of the dollar. You remember that for fiscal year 2025, our blended rate was 1.08. The policy of the company is to hedge its net exposure. That's probably 50%, which means that we translate 50% of our revenue at hedged rate and 50% at spot rate. For fiscal year '26, 85% of our net exposure is hedged at 1.10 per dollar per euro. And depending on where the euro-dollar is trending between now and the end of the year, we could see a bit of headwind on the top line. I think the main thing to keep in mind is that, from a cost perspective, a change of EUR 0.05 in the euro-dollar triggers the EBITDA 150 basis points downwards EBITDA and EBIT margin. That's the way you should model Soitec, the impact of the dollar and Soitec for the coming months.

The next question comes from the line of Oliver Wong calling from Bank of America.

I guess, just because the smartphone demand forecasts have been revised down from roughly 3% to 4% for the calendar year to now flattish 0% to 1%. I was wondering if you could give us a bit more quantification in terms of -- previously, you had expected for RF-SOI customer inventories to go down to 11 months at the end of this calendar year. Do you have a sense of -- given -- now assuming 0% to 1%, what kind of inventory level it would look like at the end of this calendar year now? And also, just because it's quite a significant cut to demand. Could this now mean -- previously, it was assumed perhaps this would last for another a year or so. Could this now mean that this would be a multiyear sort of situation?

Oliver, no, I mean, I'll just repeat what we said earlier, and we're guiding now for the September quarter. So the December quarter is still has yet to be seen. And obviously, that will make also a difference on where we land and where our customers land in terms of inventories at this end of this calendar year. So it's too early for us to share with you. We have some internal analysis and forecast, but too early to share with you. I will also remind you that we were at 14 months at the end of calendar year 2024. This is high compared to anywhere in the supply chain and compared to what we had pre-COVID, which were between 6 to 9 months. So it's not that surprising that given the current market uncertainties and all the volatility on the macro level that our customers are trying to drastically reduce their inventories and, of course, protect their balance sheet. In terms of, yes, sentiment, everyone is being more cautious across the board to say the least, despite the fact that the fundamental level, mid- to long-term, the outlook, the technology development and the need for new functionalities for semiconductors bring a high level of demand for new products at the engineered substrate level. But near-term, cautious is the name of the game.

And my other question is, you flagged the temporary weakness in POI driven by customers in Asia. Have you seen any other areas of weakness from your Asian customers, whether it be on a sort of absolute basis or on market share?

In other products, you mean?

Yes, exactly.

In other products. No. I was a few -- 1 week ago in China where I spent more than 1 week. And I can tell you that there is, of course, cautiousness because instability is everywhere on the planet. But there are also good dynamics and willingness to develop semiconductor industries, including in the smartphone world, and we are benefiting from, of course, this market. You remember that 3 years ago, China was 3% of our revenue. On fiscal year '25, it's 18%. Then we expect continuing to grow and to take our shares in this market for smartphone, for automotive and devices, generally speaking.

The next question comes from the line of Daniel Schafei calling from Citi.

On mobile inventory...

We don't hear you very well, Daniel.

Is this better?

It's better now.

Perfect. Sorry. Yes, I just wanted to clarify on mobile inventory digestion. Basically, you've mentioned that customer inventories were at 14 months end of 2024, and you have the goal to go to 11 months by end of calendar year '25. Given kind of, yes, 2 quarters left, could you give us a number of months where we're standing right now? And did it accelerate as planned? Or is it slower than that? That would be helpful to understand.

Well, we don't want to give too much details right now because, obviously, getting all the data in the middle of the calendar year is not too easy. But clearly, when we -- the market -- there are 2 trends. The smartphone market is growing slower than anticipated. That's on the negative side when it comes to volumes. But at the same time, the business outlook near-term for RF-SOI is also lower than we had anticipated. So yes, again, net-net, everybody is going to have to be a little patient and wait until the end of this calendar year. Keep in mind that we shared with you forecast. We have daily discussions with customers. But at the end, they make their decision and they decide the level of inventories that they want to hold on their books. So we have, to some extent, limited control over what they decide to keep on their balance sheet when it comes to inventories.

Okay. Understood. And just to understand also your guidance going forward. What level of tariff headwinds do you already factor into these kind of numbers? And do you see any order pull-ins or something like that given the kind of overhang in the second half?

Daniel, as we said, our exposure to U.S. is around 8% of our revenues and quite limited. And amongst this exposure, a large part of the contract are covered by customers in case of changes of tariffs. And it's a quite limited impact for us, and we're going to manage it. And that's the way we see it. The problem of tariff is not for us directly, it's for the overall world indirectly. It is creating a lot of instability and cautiousness. That's the reason why our customers are so present today. It's one of the reasons why our customers are so present today.

There are no further questions. So I will hand you back to your host to conclude today's conference. Thank you.

Then thank you very much for your interest in Soitec and for the depth and the quality of your questions. The next date in our agenda will be the release of our Q2 '26 revenue and our H1 '26 results on November 19 after market close. This ends our call for today. Thank you. Have a good day.

Thank you for joining today's call. You may now disconnect.