Good day and welcome to today's Exosens H1 2025 results conference call. My name is Sergei, and I'll be your coordinator on today's event. Throughout today's recorded presentation, all participants will be in a listen-only mode. Later, we'll conduct a question-and-answer session. You may register for questions at any time by pressing star one on your telephone keypad. It is my pleasure to introduce your host for today, Mr. Laurent Sfaxi, to begin today's call.
Thank you.
Thank you.
Thank you. Good morning, everyone, and thank you for joining us today. I'm Laurent Sfaxi, Head of Investor Relations at Exosens, and I'm joined today by Jérôme Cerisier, CEO, and Quynh-Boi Demey, CFO. We'll present you Exosens H1 2025 results and our outlook. This presentation will be followed by a Q&A session, so we can take your questions. I will now hand over to Quynh-Boi.
Thanks, Laurent. Good morning and good afternoon to all. Our full year 2024 results, in our first year following our IPO, were very strong, as you recall. We outperformed the guidance provided at the time of the IPO. Now, the first half of 2025 has also started with strong momentum, and in a very dynamic defense market environment, we remain fully on track to meet our full year 2025 guidance. Let me share with you a few highlights. Revenue first. We are targeting a high teens revenue growth for the year, and in H1 2025, we delivered EUR 224.5 million of revenue, which is up 20% versus H1 2024. Our organic growth remains also robust at +13%. Profitability. We are targeting low 20s EBITDA growth for half year 2025. Our adjusted EBITDA came in at EUR 69.5 million, up 24% year-on-year.
This represents a 30.9% EBITDA margin, which is a 92 basis points improvement versus H1 of last year. On free cash, we generated nearly EUR 24 million during the period, with a cash conversion ratio of 76%, slightly above our guidance of 70% to 75%. Finally, our balance sheet remains solid. With a leverage ratio of 1.3 x our EBITDA and strong operating cash flow generated from operations, we have the financial flexibilities to support our organic growth for R&D, but also for capacity increase, and also to fund our bolt-on M&A operations. I will now hand over to Jérôme, who will briefly recap who we are and what we do, and the key trends that shape our markets. After that, I'll return to provide deeper insights into our financial performance.
Thank you, Quynh-Boi. Good morning, everyone. First, let me summarize briefly what we do at Exosens. That will be very brief. At Exosens, we make use of physical and chemical phenomena that are invisible to the human eye to create sensing data information useful to our OEMs and end users. We hold leading positions in our niches that are part of four value-added fast-growing verticals: defense, industrial control, licenses, and nuclear. We are a global player, as our customers can be OEMs or armed forces anywhere around the world. We focus on applications more than on geographies, and our technology is used globally. We are a technology-driven company with 7.7% of our revenue invested in R&D. We have an active portfolio, a historical portfolio of over 230 patents.
We achieved, just as a reminder, last year revenues of close to EUR 394 million, active in two segments that are amplification and detection and imaging, both reflecting our high-end technology-differentiated positions on fast-growing markets. I assume for now, most of you know a lot about Exosens already, so I propose that we focus on the current understanding of our markets. If you need more information, you'll be able to find it in the presentation that we will not cover entirely today. Let me perhaps focus on our markets, starting with the markets around amplification. The night vision market benefits from a strong underlying trend in defense budget.
It was made official that now NATO countries, while not all of them had reached a 2% threshold, that was a pre-existing threshold before the conference in June, a new target has been set for NATO countries of 3.5% of GDP, with additional 1.5% for, let's say, security-related, but we'll retain here 3.5% defense-related investments as a percentage of the GDP. This 3.5% should be reached by the mid-2030s, and it will fuel the increase, will fuel the short, the medium, and the long-term growth in defense. In amplification, we provide massive, effective, and differentiated capacities to armed forces. As a reminder, we are the sole sizable non-ITAR provider of image intensifiers, and we have by far the fastest growing capacity in the world, as we have demonstrated over the last few years. Next page, we see massive demand in the coming years.
With the growing budgets, not everything is set yet. We see already some armed forces considering changing their format, adding new troops, or talking about adding new troops. For the time being, already, we see over 1 million navigation devices coming over the next few years, with three main areas: North America, where the penetration rate today is estimated to be close to 100%, but it's mainly a replacement rate with more modern, more advanced equipment, and more performing devices. Europe, where the night vision penetration rate is increasing and going to continue to increase from a 30%- 50% as a target to a 50%- 70% in some countries. Certain countries, like Germany, already targeting 100% of the troops being equipped. In the Middle East and Asia, where the penetration rate is more in the range of 10 %- 30%, it's far from being reached as of today.
Furthermore, in NATO countries, we start seeing consideration around reserve forces. We do not talk anymore about only armed forces and land forces, but really about global forces, and reserves are now being considered to a certain level for being part of the defense efforts. In the coming years, we consider Exosens is very well positioned to capture an increasing demand in Europe, in the Middle East, and in Asia, but we aim at also gaining traction in the U.S. market. First, concerning the U.S., it is by far the largest single market in the world. It represents about 45% of the whole market. It's a very dynamic market where efforts of modernizing land forces are constant and where we understand there is a sort of a form of a constrained production capacity. This is why we have chosen to implement a new factory for several reasons.
First, to free up capacity in Europe so that Europe factories serve countries where ITAR free is required, and these countries could be served from these countries where ITAR free is required will be served from Europe, and countries where it is not required could be served from our U.S. factory. Second is on the U.S. market itself. U.S. is targeting, U.S. is launching and having multi-year contracts. We also know penetrating the U.S. market could take time, and we believe being on the ground there in the country will be a change for the longer term in this country. Second, concerning Europe, a few countries like Germany, like Poland, like Finland have plans to increase their armed forces. I mentioned that. Also their reserves. Other countries are already entering fully multi-year programs with large volumes that had not been seen in the past.
It is the case, for example, in Spain, in Italy, in the U.K., in Greece. We see these large volumes progressively being more frequent. Finally, in the Middle East and in Asia, there are a few large countries that will constitute the bulk of the upcoming volumes. Here, Exosens enjoys a very strong position, a key position as a supplier of high-quality exportable tubes. The group, for example, has a 30-year or close to 30-year-old partnership with Bell in India. Now, when we consider capacity, we are facing a growing demand. We have had, since 2020, a staggered capacity approach with stages of capacity increases that have been decided over time. Between 2020 and 2024, our capacity will have nearly doubled, but our output will have more than doubled. We did that in several stages.
The last stages already announced consist in increasing our capacity by 25% until end 2026, beginning 2027, in the U.S. and Europe. However, we are facing a growing demand because these announcements were made before the announcements of the GDP increase, before the European, let's say, consideration about increasing budgets following the Munich conference. We are evaluating our further increase in capacity, and we will make announcements in due time as we consider necessary. Obviously, in end 2024, the market was seen stronger at the time of the IPO, and today, we continue to see a growing trend in this market. We have demonstrated that we have the ability to increase our capacity. We have the cash and the financing necessary to do so. We have a staggered approach, and we will adapt constantly to the market and to the demand.
Furthermore, our ability to increase our capacity will be increased in the future because we will also be able to do that not only in Europe, in our two factories in Europe, but also in the U.S. where this new factory was setting up. Concerning our DNI segment, the DNI segment is also enjoying a strong tailwind coming from the defense. On that market, we are targeting four main applications. Let me start perhaps with the two first ones: drones and counter drones. Drones and anti-drone devices have become a pillar of modern warfare following the lessons learned from the return of identity conflicts, especially between Ukraine and Russia. We see this segment as a very fast-growing subsegment of our business in defense. Our technology is a very good fit for short and medium-range surveillance. Actually, the thermal and the cooled and the cameras are perfectly fit for long range.
Shorter range, more invisible in the visible world. That corresponds to medium-sized tactical drones, which units or programs count in the range of a few thousand units, a few tens of thousand units in case for larger programs, perhaps concerning certain countries. The use of infrared and cooled and cooled fits perfectly the purpose, and it requires differentiated, well-integrated, high-performance technology that Exosens is providing. We have made a significant design in sales to get undisclosed customers in Europe, and we see this program and these subsegments really continue to pick up. We expect, in fact, a significant part of the increasing budget to fuel this drone and, consequently, the anti-drone warfare in the NATO countries. Our other applications in defense, the two remaining ones, are really around platforms and missile warning systems.
These programs come either through modernization of existing programs with optronics technologies of a new generation or, for new programs, generalization of optronics on all types of platforms together with large equipment programs. The visible low-light cameras and the thermal cameras are becoming part of any new or recent platform, and the demand of these platforms is expanding rapidly, aligned with the budget increases, the reinforcement of the troops, the reinforcement of the land forces, and the put in capacity of the armed forces. Key customers design wins add to the expansion of our business, together with expansion within existing and current customers. Rheinmetall's driver's enhancement systems, for example, as well as handheld missile warning systems in Germany and other NATO countries, are fueling and will continue to fuel the growth in the medium to long term. In DNI, our markets are generally supported by strong underlying factors.
As a reminder, we estimate that the medium-term growth for these markets is in the high single-digit range between 7%- 10%. Specifically in nuclear, we have started a sort of, I don't know, a supercycle. The word is a little bit of a buzzword, but this is what it is. Supported by the renewal of safety standards into the existing platforms, the existing plans, the increase in electricity demand that is rising everywhere due to the electrification of the industry, of our societies, of transports, and the emergence of new business models of SMRs, all of that has made nuclear inevitable and totally part of any modern energy mix, amongst other things, but nuclear is renewed because of these few but strong factors.
In nuclear, we enjoy a strong position on all types of reactors, especially large reactors, but also SMRs we announced, and we had some win in 2024, but also research reactors. We have a unique radiation technology, and we continue to develop and to invest into new developments, new products, especially on high-temperature detectors for SMRs with new technologies of SMRs emerging. These, what is now today's studies, feasibility, qualification studies, will be followed in a few years by production, and we see that starting by the end of the decade for some projects. Industrial control enjoys a certain market recovery, but prospects remain strong until the medium term. As we speak, the rise of artificial intelligence applied to production is fueling growth, requiring more data, more information.
We see we're starting the end to the destocking in some industrial markets, like sorting, as well as a restart to a certain level in CapEx investments that should translate into a renewed interest for optronics controlling control systems. This market is tempered for the time being, still with the current reshaping of the U.S. scientific market. Overall, we see the trend in the industrial market as being more positive moving forward, and we see that remaining a high single-digit growth for the next few years. Last, on our life science vertical, we see still some softness with sustainable drivers for the medium term. When the uncertainties around the U.S. research will have resolved, it's more about uncertainty rather than absolute levels. Exosens maintains a huge strategic differentiation, ensuring that we remain at the leading edge of the technology, the most adapted provider of detectors to our customers' instruments.
In the short term, the Chinese market remains soft to medium-oriented. The U.S. Inflation Reduction Act slightly impacts the pharmaceuticals. Some of our customers are actually seizing the opportunity of this market status to invest into new generations of instruments that will probably come to the market in a few years. We see them kicking in, let's say, in two years, and the interest for Exosens progress is massive. Thanks to our unique technological edge and our strategic choices, we see this market as continuing to fuel the growth of the medium term. Overall, the perspective for growth for the group on the business remains strong. Perhaps it's time to go a little bit more into details on how we did during H1. I think, Quynh-Boi, you will lead us through that.
Thanks, Jérôme. Now let's look at how these market trends have translated in our financial performance for the first half of full year 2025. As I mentioned in the beginning, we continue to deliver strong growth, but we also improve our margin. Revenue is up 20%, adding EUR 38 million of revenue overall, EUR 24 million from amplification, and it's driven by very solid defense investments. Our backlog is strong at the moment and gives us a very good visibility for the months ahead. We also combine this with excellent execution, with our factories running at full capacity and very high yield. EUR 20 million of additional revenue is coming from detection and imaging, mainly thanks to our acquisition: Sandtronic, that we closed on July 31st, 2024, EliaTech, that we closed on September 1st, 2024, and more recently, Noxant, that we closed in March 2025.
Our adjusted gross margin has also grown by 22%, reaching 49.6% of sales. That's up 89 basis points compared to last year. This adds about EUR 20 million in gross margin, of which EUR 16 million from amplification and EUR 4.5 million from DNI. Now, let's dive into the details by segment in the next slide. First, amplification. The revenue grew by almost 18%, and it's driven by three main factors. The first is sustained market demand, as Jérôme mentioned. Second is increased capacity, thanks to additional CapEx that we consistently invested since 2022. Third, by our sustained high yields. On the demand side, our growth is largely due to increased volumes of night vision goggles for land forces. These night vision goggles are powered by our image intensifier tubes.
The war in Ukraine and the high-density combat operations have made all armies realize that night missions are critical and give a real competitive advantage to the armies that are properly equipped. Thanks to the industrial investment that we have made since 2022, we've been able to gradually increase our production capacity. At the same time, our industrial yields have remained very high, with limited scrap and rework, as our interim workforce is now fully trained and fully effective. It shows how our industrial excellence is delivered at Exosens, while we are able to grow capacity while maintaining a very high quality of our products with low return rates from our customers and very low scraps at the same time. We are also seeing steady demand for higher performance products, and this translates into a better product mix for us.
All of this leads to an improvement in our gross margin in amplification of 276 basis points, improvement year over year. The gross margin stands now at almost 50%, more precisely 49.8%, which is a record high for us. Second, under detection and imaging, we also grew by 23%, 23.6%, and driven by three factors: two positive and one more adverse one. The first is, as I mentioned earlier, the external acquisitions that fuel our growth. Second is the negative impact from reduced U.S. scientific research spending, as Jérôme mentioned earlier. Third is the increased market demand for defense imaging and detection applications. As I mentioned earlier, a big part of our DNI growth is coming from our acquisitions closed in H2 of 2024 or H1 of this year. On a like-for-like basis, H1 growth is actually down by 2.5%, mainly due to the decline in the U.S.
scientific research investment. This caused a drop in Q1, which still affects Q2, by the way, but to a lesser extent. If we exclude the U.S. scientific research cameras, our underlying growth would have been in the low single digit in H1 2024, which is more or less what we had in H1 of 2024. In Q2, we have seen a recovery in DNI with a 7% like-for-like growth. We have experienced a growing demand for imaging and protection systems in defense applications. For example, camera on drones, long-range surveillance cameras that detect massive drone attacks, and also missile warning systems on aircraft or on tanks. We have also benefited from positive pricing and product mix, which has helped us partially offset the negative volume impact. All in all, our DNI gross margin came in at 48.7%, which is roughly flat compared to full year 2024, which was at 48.6%.
However, this is almost a three-point decrease compared to half year 2024, which was at 51%, which was very high. This drop is mainly due to the dilutive effect of our acquisitions, which are on average about 10 points below our group's margin. Their EBITDA margin is around 20%. Next page, EBITDA and EBIT. We deliver best-in-class EBITDA margin of 30.9% and EBIT margin of 26.1%. That reflects our strong position in a market that demands high technology and very good industrial expertise. While we drive both growth and margin expansion, we also kept control over our fixed costs and as we benefit from scale effects. This has led us to improve our EBITDA margin by 92 points and our EBIT margin by 142 basis points. Both EBITDA and EBIT margins have reached record high levels for us.
Net income, we deliver a net profit of $27.9 million, which is a significant improvement versus last year, which was at $2.9 million. Three points that I would like to highlight here. First half of 2024 included one-off impact due to our IPO. We had close to $4 million of operational costs, operational IPO consulting fees that were booked in other incoming expenses, as you can see in the note one. We also booked $13 million of one-off costs related to the refinancing of our debt. This is close to $17 million of one-off costs related to the IPO. We also restructured our debt following the IPO and have significantly reduced our financial costs. The third impact is our income tax, which has significantly increased compared to last year as we increased our profitability. Now, on R&D, next page.
We are an industrial tech company, and at the core of the model is our strong commitment to R&D, which is a key lever for innovation and to sustain our future growth. Over the period, we have continued to prioritize our R&D investment, and we maintain a very solid 7.1% of sales for R&D, which is consistent with our guidance of 7% and 8%. This ratio is also very comparable to last year's 7.9%. In absolute terms, this is a gross R&D spend of $17 million, which is up $2.3 million versus last year. It's an increase of R&D spend by $1.6 million, of which half a million is due to our scope effects primarily linked to acquisitions. As you know, technology is critical to our M&A strategy, and the integration of this M&A has also a direct impact on R&D spending.
The remaining $1.1 million increase reflects our continued investment in R&D strategic initiatives. Amongst others, there is the 5G night vision project that is almost complete and for which we will commercially launch in September of this year. We also continue to leverage external funding mechanisms for our R&D spend. Over the period, it represented $3.5 million of subsidies and grants received either from tax credits or from customers. This is an increase of 0.7% versus last year. This is fully in line with our strategy to co-develop solutions with our customers, as it helps us de-risk innovation and also secure repeat business with our customers and strengthen long-term partnerships. Our second pillar of our industrial tech model lies in industrial excellence.
As a process manufacturer, we transform raw materials into high-tech components in order to maintain best-in-class industrial assets, and this is critical for us to keep good yields and operational efficiency. On CapEx, two points that I would like to highlight. First, we made strong investment in 2022 and 2023 to meet rising demand, and we ramped up production throughout 2024. While we expected a normalization of our CapEx, the sustained strength in the defense environment has led us to maintain a high level of CapEx. As you can see, our gross CapEx has increased from EUR 7.2 million to EUR 9.7 million. This reflects our commitment to scale. Let me give you a few examples. At our Brive facility, construction is underway for a new building that is designed to house additional equipment. This is a key enabler for future capacity expansion.
In parallel, we have also made targeted investment in critical tooling and machines that will directly increase our output and operational efficiency. Second point I would like to highlight: our strategic investment plan of EUR 20 million that we announced in January. This plan is progressing fully on track. Technical engineering studies are underway, and assets are currently being built. We expect the initial commissioning to begin during 2026, with the full capacity coming by mid-2027. This timeline will allow us a smooth and scalable capacity increase. Free cash, our performance was once again a strong performance. We generated EUR 24 million of free cash, which is in line with the level that we achieved in the first half of 2024. As we increase our EBITDA, this positive impact was offset by higher working capital needs, which was expected given the 20% growth in our activity.
A large part of this working capital is due to inventory buildup for H2 deliveries in response to our sustained demand. Importantly, we control our CapEx carefully, even as we scale operations. As a result, we achieve a cash conversion ratio of 76%, improving from the 74% of last year and exceeding our guidance of between 70% and 75%. Finally, leverage ratio. At the end of 2024, our leverage ratio stood at 1.2 x our EBITDA. Even after the closing of the Noxant acquisition in March 2025, we've managed to maintain it at a conservative level of 1.3 x our EBITDA. With solid cash generation from our operations, the refinancing at the time of the IPO with undrawn RCF of EUR 100 million, we are in a very strong position to fund our organic growth ambitions, but also our bolt-on M&A strategy.
This concludes my first review of first half results, and I will now hand over to Jérôme, who will briefly discuss our outlook for 2025, sorry, and the 2024-2026 period.
Thank you, Quynh-Boi. Let me perhaps just remind you what we expect for 2025 and 2026. For 2025, we expect a continuing strong performance with revenue growth in the high teens and adjusted EBITDA growth in the low 20s compared to 2024. This implies a mild improvement in our adjusted EBITDA margin year over year on the full-year basis. Until 2026, we expect a 2024-2026 adjusted EBITDA CAGR to be in the high teens. On investments, with additional investment capacity, additional investment of EUR 20 million announced, we expect the cash conversion to be between 70% and 75% over the 2024-2026 period. On the M&A, we will continue our selective M&A strategy while maintaining a leverage ratio of about 2x. As a conclusion, our H1 results show a strong performance. We are very pleased with this performance that showed sustained revenue, sustained profit growth driven by different sale wins.
Looking ahead, we expect these positive trends to continue throughout the remainder of 2025. We consider today we remain fully on track to deliver our financial guidance for the year with this performance and the upcoming perspectives. This concludes our presentation. I think we will now be happy to take any questions you might have concerning this publication.
Thank you, sir. As a reminder, to ask a question, please signal by pressing star one. If you find that your question has already been answered, you may remove yourself from the queue by pressing star two. Please make sure the mute function on your phone is switched off to allow your signal to reach our equipment. Again, it is star one to ask a question. Our first question is from Alexander Peterc from Bernstein. Please go ahead.
Yes, good morning, and thank you for taking my question. The first one would be on amplification. I would like to understand to what extent you're now becoming very capacity-constrained in this business. If you could give us an idea as to how far out you are currently booked, I suppose it's well into 2026. If you could give us an idea of what is the capacity by end 2025, is it midway between 190,000 and 240,000 that you have for end 2024 and 2026? The second question would be on your intention to grow DNI through M&A. Would you not consider to put this on pause a little bit and direct all of your efforts into increasing capacity in amplification where you have the biggest constraints at the moment?
Finally, just a housekeeping one for Quynh-Boi, on tax for the year, what will be the P&L tax rate and the cash tax rate? Thank you very much.
Okay, let me perhaps first answer on capacity. Our capacity is growing as we speak under several factors. First, we work continuously on the improvements of our processes, our yields. As we speak, the capacity is increasing. On top of that, yes, we add investments and we add machines and parts that we have to train. Let me perhaps illustrate the fact that the $20 million additional investment for a capacity increase of 25% is shared between the U.S. and Europe. Europe's capacity increase will come a little bit earlier because, let's say, the infrastructure is more advanced. It's already in place, in fact. We consider that the capacity for the European part will be hitting earlier than the U.S. part as soon as 2026. Meanwhile, until that, yes, we have these continuous improvements on capacity that should also be translated into our production output.
Add constantly yield, which we always work on, but we have to be careful about yields that can evolve over time. Reaching the beginning of 2027, the U.S. capacity will fully kick in. Meanwhile, we have the ability, as I mentioned, to decide further capacity increases that would hit simultaneously Europe and U.S. on their own. I hope that this translates your question, but we do not expect, let's say, a one-off capacity increase by 2027, but something more regular, especially starting. It's already starting, in fact, but it will add on and accumulate over the course of 2025 and mainly 2026. Concerning CapEx, capacity increase is mainly driven by operations. We have the financial means to both finance our capacity increase, which is our R&D, which is our priority. We self-refinance in priority our organic growth and also, at the same time, continue our bolt-on strategy.
In terms of internal resources, the CapEx investments are driven by operations in an established organization and are mainly in amplification, while we still consider detection imaging is a priority for any potential M&A. We consider that these topics can be run in parallel without impacting each other. The group has the capacity to run these two strategic axes simultaneously.
I would add, even if we wanted to, increasing capacity, if it's not possible, it doesn't mean that if you invest $20 million in organic growth for industrial CapEx, you would accelerate investment because there are also physical limits to it. Now, on your last question on the tax rate, P&L tax rate and cash tax rate, cash tax rate is around 20%, depending on the jurisdiction where we pay our taxes. As you know, in France, we don't pay taxes and we still have deferred tax losses. Our industrial footprint has also changed with increasing DNI that is mostly outside of France. In terms of P&L tax rate, we are normalizing around 25% more or less, depending on the jurisdiction where we pay our taxes as well.
That's very clear. Thank you very much.
Thank you. As a reminder, to ask a question, please signal by pressing star one. Now, our next question is from Aurélien Sévignon from ODDO BHF. Please go ahead.
Hi, good morning. Thanks for taking my question. Just to follow up on production capacity increase, could we start seeing an impact from Q4? I mean, will it be possible to exceed quarterly revenue of $81 million, $82 million in amplification as early as Q4, or is that still a bit too early? On the initial results, it was sharply down year-on-year in H1, but it increased versus H2 at 2025, I believe. Could you maybe provide some context or color around the evolution? Thank you.
Okay. We expect H2 to be not smaller than H1, despite the intra-half seasonality that we always see during the summer period. That means that the answer to your question is yes, we expect, in fact, the threshold you mentioned is not a threshold, in fact, and we expect it to be, I don't know how to say it, but we expect ourselves to be higher than this value over the last part of H2.
Thank you.
There was another question?
Is there currently no further questions at this time, sir? With this, I'd like to hand the call back over to you for any additional or closing remarks. Oh, pardon, we have a pop-up question from Paul Valentin from Stifel. Please go ahead. Your line is open.
Hello, everyone. Do you hear me well?
Yes.
Yes.
Thank you for taking my questions. Congratulations on these good results. My first question is about detection and imaging, specifically in the life science, industrial control, and nuclear segments. I was wondering how much visibility do you have with your customers in this sector? I mean, how long does it take to clear your order book with them on average? Maybe your answer first, and then I will ask my second question.
DNI is a short-cycle business. The typical lead time between the time we book an order and we deliver is two to three months, very short. At the same time, with our design-in approach, we know that when we are design-in, the customer will buy from us as long as they will sell their own products because we're design-in. We share forecasts with our customers, but without necessarily PO or frame agreement or firm contract, if you wish.
Yeah. As a consequence, when we sell our products to our customers based on the fact that themselves they sell it to the end customers, and because the cycle is short, we follow this trend. However, the repeatability is ensured on the very long term for the life of the application, 10 years, 15 years, only based on the design-ins that we manage in the first phases. This is why I mentioned earlier in the market section that when we see some customers seizing the opportunity of the current market status to actually invest into new generations of machines, that means that it's a time when we are getting design-in, and then we know we will have repeated revenues for the long period of time.
This will apply, by the way, also for defense application in DNI, the same model.
Yes, thank you. Very clear. My second question is about the amplification segment. Organic growth was only 8% in Q2, which marks a slight slowdown, most likely temporary. Is this due to a waiting certitude from customers preferring to wait for the next generation of tubes to be released in September, or is it related to temporary capacity constraints and bottlenecks on which we are working to keep growth at a higher rate?
This is more the base effect of the amplification business. As you know, 2024, we increased our output throughout the quarter. Q1 went lower, Q2 we increased, and so on and so forth. That's the reason why we have the base effect that has impacted us in Q2 this year. At the same time, Q3 last year was also very low. We saw close to high teens decline versus Q3 2024 versus Q2 2024. This year, we have a plan also to manage the summer closure as well.
Okay, thank you. My next question is about new capacity increases being considered for 2027 and beyond. I imagine you are keen to increase capacity to meet demand while being at the same time careful to not increase it too sharply in the short term to ensure that in the long term, capacity is always used at a good utilization rate. My question is, under this decision-making process, do you prioritize growth, even if it means taking probably possibly the risk of creating overcapacity in the long term, or do you prefer to guarantee a high level of margins like today in the long term by increasing capacity very gradually, even if it could mean taking the risk of losing a small amount of market shares, not totally following the demand increase in the short term? Do you see my point?
Yes. Actually, the capacity increase is a complex phenomenon because when we manufacture tubes, in fact, we have over 460 different steps. When we say capacity increase, it's more adding here and there certain machines, but with the different stages, we do not add exactly the same ones. Having said that, there is one constraint in the capacity increase that is coming from the ability of our suppliers to provide the specific machines that are proprietary. This is what is facing the capacity, not a choice between two different strategy directions. We have the capacity to add these suppliers to run the capacity as we wish, but this is the main constraint. It takes 18 months, basically, to install, 18 months- 24 months to install the capacity.
This is why, anyway, we have to take our decisions early in order to see the results within the next 18 months- 24 months.
Okay.
Thank you. We have a follow-up question from Alexander Peterc from Bernstein. Please go ahead.
Yes, thanks. I just have two follow-ups. One is on 5G. Do you have any update to provide on that? Do you have anything to share in terms of technical specifications, figure of merit increasements 40+? The second question is, you say in the slides that you're now eligible for U.S. Department of Defense contracts thanks to your new U.S.-based plant. Do you intend to start shipping the U.S. Department of Defense as soon as that plant is on stream, or will you first address ITAR compliant markets outside of the U.S.? I'd just like to understand to what extent you need to jump through any hoops to be able to start delivering to the Department of Defense, get the products validated, or is that already done? Thanks.
Okay. So, concerning the 5G, the development is ongoing. We have now reached the qualification phase. It's going well, and we still plan to launch commercially this product after the summer. We are on track and fully on track to deliver that. I don't think we have officially disclosed any performance data sheets, but we are, let's say, on track to our plan, and we will seize the opportunity of an international gathering of customers to launch it commercially beginning of September, mid-September, sorry. It's mid-September. We are on track on that. Concerning the DOD capacity, there are two things in making business with the U.S. Department of Defense. First, you have to be compliant and allowed, and this is our case because since years, we have been delivering devices to the U.S. Department of Defense, not night vision devices, but electronic amplification devices to the U.S. Department of Defense.
That is done through a special status for our affiliate, which is an SSA. We have this compliance is proven. The other part is we need to gain business so that we can ship to the U.S. Department of Defense. It is fully understood that the first shipments to the U.S. Department of Defense, should they occur earlier than when the factory will be ready and be able to ship, would occur from Europe. It's fully accepted and fully understood. We are starting, we started, and we are starting to continue to start the development, the business development in the U.S., which in a way can be served from Europe for the time being until we are fully installed and in capacity in the U.S., which we still plan for the beginning of 2027.
That's great. Thank you very much.
Thank you. If there are currently no further questions, I'd like to hand the call back over to Laurent Sfaxi for closing remarks.
Thank you. Thank you again for joining this call today. We remain at your disposal should you have any further questions. We wish you a very good day, as well as a great summer holiday. Thank you.
Thank you. This concludes today's conference call. Thank you for your participation, ladies and gentlemen. You may now disconnect.