Eutelsat Communications S.A. (ETL) Earnings Call Transcript & Summary

Hello, and welcome to the Eutelsat Group Second Quarter and First Half 2024/'25 Results. My name is Caroline, and I'll be your coordinator for today's event. Please note, this call is being recorded. [Operator Instructions] For today's event, we have Eva Berneke, the CEO; and Christophe Caudrelier, the CFO. I will now hand over the call to your host, Eva Berneke, the CEO, to begin today's conference. Thank you.

Good morning. Welcome, and thank you for joining us today for Eutelsat's Second Quarter and First Half '25 Results Presentation. I'm Eva Berneke, the CEO, and I'm joined by Christophe, our CFO. Today's agenda, as usual, we'll run through a few of the key events in the first half as well as our operational and financial performance. And we'll take a few next steps for the remainder of our financial year, which ends this summer, and then wrap up around the financial outlook. Start out with the key events of the first half. The key event was, of course, the signature of the agreement that will set the SpaceRISE consortium designing, building, and operating the IRIS2 multi-orbit constellation. This presents a big step in Eutelsat's LEO build-out strategy. The signature was on December 16 in Brussels. Following this confirmation of the IRIS2 contract, Eutelsat procured the first batch of 100 LEO satellites that are required to ensure continuity, enhancement of services, and new technology compatible with the IRIS2 future assets. In December, we also exercised the put option of the sales and leaseback of passive ground infrastructure, where proceeds are still expected due in first half calendar '26. In terms of financial results, we reported a first half with operating vertical revenues of EUR 600 million, up just shy of 4% at 3.9%, and an adjusted EBITDA margin of 55.2%, which is stable year-on-year. This performance enabled us to confirm our full-year revenue and EBITDA margin objectives, albeit with less headroom than in the beginning of the year. Elsewhere, our gross CapEx is now expected to be lower than initially guided, expected to be between EUR 500 million and EUR 600 million, thanks to the timing of LEO investments and reinforced vigilance on GEO expenditures. Finally, and a less positive note, we took a goodwill impairment of EUR 535 million on various GEO assets, reflecting a lower expected future cash flows from these assets. I'll come back to this A quick look at the key numbers. Total revenues at EUR 606 million, up by almost 6% at 5.9% on a reported basis and a 4.4% on a like-for-like. As you remember, we only integrated OneWeb at the end of Q1 last year. When we look at our operating verticals, revenues for our 4 operating verticals stood at EUR 600 million, up by 3.9% or almost 4% on a like-for-like basis. Reported adjusted EBITDA stood at EUR 334.9 million at the end of the year '24 compared to EUR 365.6 million a year earlier. On a like-for-like basis, adjusted EBITDA was up 4.9%. The adjusted EBITDA margin stood at 55.1% at constant currency versus 63.8% reported and 54.8% on a like-for-like basis. The net debt to adjusted EBITDA ratio stood at 3.92 compared to 3.79 at the end of June '24 and 3.79 at December '23. I'll hand over to Christophe to look at a bit more details on the operational and financial performance before I come back to some of the strategic outlook for the year.

Thank you, Eva. Good morning, everyone. As mentioned above, total revenues for the first half '24/'25 stood at EUR 606 million, up by 5.2% on a reported basis and by 4.4% like-for-like. They reflected a perimeter effect of EUR 8 million due to the acquisition of OneWeb, a EUR 2 million negative currency effect, and a EUR 5 million positive swing in other revenues, mainly from hedging. Excluding other revenues, revenues of the 4 operating verticals were up 3.9% on a like-for-like basis. Let's have a look at the segmental reporting. Video, representing 51% of revenues, stood at EUR 309 million, a decline of 6.4%. Fixed Connectivity revenues, representing 20% of the group total, rose 22% to EUR 119 million. Government Services, 16% of revenues, stood at EUR 96 million, a rise of 22%. And finally, Mobile Connectivity revenues, representing 13% of the group total, stood at EUR 75 million, a rise of 7%. Let's go into more detail, starting with Video. First half revenues were down by 6.4% to EUR 309.2 million, in line with the broader secular market decline. Second quarter revenues stood at EUR 157.4 million, down by 5.6% year-on-year and up 3.8% on a sequential basis, reflecting the linearization of revenue recognition on certain contracts. This trend does not alter the underlying cadence in Video of a mid-single-digit decline and the second half is expected in line with the trend of the first half. Moving to Fixed Connectivity. First half revenues stood at EUR 118.9 million, up 22% year-on-year, mainly reflecting the continued growth of LEO-enabled connectivity solutions as well as a one-off impact from catch-up revenues from ADO customers. Second quarter revenues stood at EUR 62.2 million, up 16% year-on-year and by 9.9% on a sequential basis, mainly reflecting the above-mentioned one-off impact. Key contracts signed during the past quarter include a new multiyear agreement with Q-KON to expand LEO satellite services across Sub-Saharan Africa as well as a multiyear, multimillion-dollar partnership with NIGCOMSAT to deliver LEO satellite services in Nigeria. Second half revenues will reflect more challenging conditions for GEO-enabled consumer broadband in Europe and notably by the temporary stop of revenue recognition from a specific customer on a KONNECT VHTS satellite. Against this backdrop, Eutelsat is repurposing the capacity on KONNECT VHTS to address a broader range of applications, notably Mobile Connectivity. First half Government Services revenues stood at EUR 96.4 million, up by 21.9% year-on-year, reflecting the contribution from LEO services. Second quarter revenues stood at EUR 50 million, up by 23.3% year-on-year and by 8% quarter-on-quarter. This vertical is benefiting from improved U.S. DoD renewals in the latest campaigns as well as increased demand from non-U.S. governments. And finally, Mobile Connectivity revenues stood at EUR 75.3 million for the first half, up 7% year-on-year, mainly reflecting demand for LEO-based solutions, notably for maritime applications. Second quarter revenues stood at EUR 33.3 million, down 4.5% year-on-year and by 20.4% quarter-on-quarter. This decrease reflected lower GEO revenues as well as a one-off contract in Q1 of around EUR 3 million not repeated in Q2 and higher equipment sales in Q1. As a result, the backlog stood at EUR 3.7 billion at the end of December '24 versus EUR 3.9 billion a year earlier. This decrease reflects the natural erosion of the backlog, especially in the Video segment, partly offset by the growing LEO backlog. Backlog was equivalent to 3.1x of '23/'24 revenues, with Connectivity representing 56% of the total and LEO now accounting for 48% of this segment. Let's turn now to the financial performance. Reported adjusted EBITDA stood at EUR 335 million at the end of December '24 compared with EUR 365.6 million a year earlier, down by 8.4%. On a like-for-like basis, adjusted EBITDA was up 4.9%. The adjusted EBITDA margin stood at 55.1% at constant currency versus 63.8% reported and 54.8% on a like-for-like basis. Operating costs were EUR 64.3 million higher than last fiscal year, reflecting the impact of the consolidation of OneWeb for 6 months of the current fiscal year compared with only 3 months for fiscal year '23/'24. On a pro forma basis, costs were up 3.7%, reflecting, on one hand, the embarkation of OneWeb at full operational run rate and on the other, cost control measures implemented since the merger. Group share of net income was a loss of EUR 873.2 million versus a loss of EUR 191.3 million a year earlier. This reflected higher other operating expenses of EUR 690.8 million compared to EUR 183.9 million last year. They included a goodwill impairment of EUR 535 million in respect of GEO assets based on the test performed at the end of December 2024. It reflects the cash flow forecast adopted by the group in its latest 5-year plan, embarking the lower future cash flows the group expects to be able to generate from its existing GEO assets. Please take account of increased competition in the connectivity market and a greater-than-expected decline in demand for video services. This is consistent with the impact already experienced by the group in lower video customer renewal rates and more recently, the transfer of demand from GEO to LEO connectivity services. Then higher depreciation of EUR 433.7 million versus EUR 316.1 million a year earlier, reflecting the perimeter effect from OneWeb as well as higher in-orbit and on-ground depreciation. EUTELSAT 36D satellite and 20 LEO spares entered service during the first half. Net financial results of EUR 99.1 million negative versus minus EUR 60.7 million a year earlier, reflecting higher interest costs, partly offset by favorable evolution of foreign exchange gains and losses. Corporate tax expense of EUR 7.6 million versus a tax gain of EUR 28.5 million a year earlier, implying an effective tax rate of minus 0.9%. It reflects the nonrecognition of deferred tax assets related to losses in France and in the U.K., the net impact of the exemption mechanism for profits allocated to satellites operated outside France, the effect of the tax rates of foreign subsidiaries, and the impact of impairments on the group satellites, particularly those in the Satmex arc. Losses from associates of minus EUR 1 million versus minus EUR 23 million, reflecting the contribution of the stake in OneWeb in the first quarter of fiscal year '23/'24, now fully consolidated. Moving to CapEx. Gross CapEx amounted to EUR 174.8 million versus EUR 313.7 million last year. This decrease reflects the GEO satellite program delivery and launch last year as well as lower LEO on-ground CapEx versus last year. First half CapEx is not representative of expected '24/'25 outturn, which will embark the 100 LEO satellite batch order. Nevertheless, CapEx for the full year is now expected in the EUR 500 million to EUR 600 million range, lower than previous range of EUR 700 million to EUR 800 million, reflecting the timing of LEO investments as well as increased vigilance on GEO CapEx. At the end of December 2024, net debt stood at EUR 2,695.8 million, up EUR 151.6 million versus the end of June 2024. It was mainly due to CapEx-related movements and higher financial costs, partially offset by net cash flow generated by activities. The results, the net debt to adjusted EBITDA ratio stood at 3.92x compared to 4.13x at the end of December '24 and 3.79x at the end of June 2024. The average cost of debt after hedging stood at 4.84%, was 3.16% in H1 '23-'24. The weighted average maturity of the group debt stood at 3 years compared to 3 years at the end of December '23. Undrawn credit lines on cash stood at around EUR 1.24 billion. Now back to Eva to comment on the outlook and next steps.

Thank you, Christophe. The past 3 months have been an alignment of several strategic elements paving our way for our LEO build-out strategy. First, we exercised the put option with the EQT Infrastructure Fund regarding the sale of a majority stake in our passive ground infrastructure, leading to the signing of a binding share purchase agreement. As a reminder, the transaction consists of the carve-out of Eutelsat's passive ground infrastructure assets to form a stand-alone company in which EQT will acquire 80%, while Eutelsat will remain committed as a long-term shareholder, anchor tenant, and partner of the new company with a 20% holding. The transaction values in the new entity at an enterprise value of EUR 790 million with the closing of the deal expected in the first half of calendar year '26. This will deliver net proceeds of around EUR 500 million after tax to Eutelsat at the sale of 80% and will strengthen our financial resources and contribute to the LEO constellation extension. The second very important element is the SpaceRISE consortium and the signature of IRIS2. SpaceRISE consortium, where we are a leading member received the go-ahead from the European Union to design, build, and operate the IRIS2 constellation due to enter service in 2030 with an initial 12-year concession contract. As a reminder, the project is valued at EUR 10.6 billion with the public funding representing around 60% of the total project cost, supplemented by private financing from the 3 consortium members. Eutelsat will invest in the region of EUR 2 billion back-end loaded to the later stages of the project, and it will allow Eutelsat to have access to the additional sellable LEO capacity of around 1.5 terabits out of a total of 2 terabits of LEO capacity, access to the KaMil capacity, which is not consumed by sovereign needs of EU. Scale advantages and the shared fixed infrastructure and R&D investments in new technology funded by EU and ESA, commitments from EU and other member states for IRIS2 capacity worth several hundred millions of euros in take-or-pays and a clearly capped financial commitment with strict milestones providing for exit and compensation in the event of missed targets that will compromise returns to Eutelsat. So Eutelsat is expected to generate around EUR 6.5 billion of revenues over the 12-year concession period. Finally, given that IRIS2 is now confirmed and moving forward, this also allowed the confirmation of our involvement and represents a key step in our strategy of how to develop and expand our LEO constellation. Specifically, it gives a clear road map for the extension of the existing LEO constellation in terms of technology road map, which is compatible with the future IRIS2 asset. Following confirmation of the IRIS2 contract, Eutelsat, therefore, procured the first batch of 100 LEO satellites to be delivered by the end of calendar '26 that will ensure continuity and enhancement of services. We estimate the extension of LEO constellation up to the availability of IRIS2 will require a further 340 satellites on top of these initial 100, and this equates a total of the extension program in the order of EUR 2 billion to EUR 2.2 billion between this financial year '24/'25 and '28/'29. As mentioned, our contribution to IRIS2 will be back-end loaded during the period ahead of the availability of the constellation. The availability of the constellation is expected to end '30 and thereby, our back-end loaded contribution will be beginning in '28/'29. We're actively working on a financing plan in line with our strategic road map and also in line with our longer-term leverage targets. So with all of that, we confirm our first half performance was in line with our expectations, enabling us to confirm the financial year '24/'25 revenue and profitability objectives. As a reminder, the '25 financial year revenues of our 4 operating verticals are around the same level as financial year '24 and an adjusted EBITDA margin slightly below the level of '24. Nevertheless, these objectives will be more challenging in light of the GEO consumer broadband headwinds identified above and in Christophe's notes. Our gross capital expenditure for this year initially expected in the range of EUR 700 million to EUR 800 million is now expected around EUR 200 million lower in the range of EUR 500 million to EUR 600 million. That also means we continue to target a leverage of 3x in the medium term. So a few words to sum up before we go to Q&A. H1 revenues and EBITDA margin in line with expectations and our objectives for the financial year confirmed. Our CapEx is now expected EUR 200 million lower at EUR 500 million to EUR 600 million range. The put option for the sale and leaseback of our passive infrastructure and ground stations is expected to generate the EUR 500 million in H1 calendar '26 and the go-ahead of IRIS2 constellation representing a key milestone in our multi-orbit strategy also allows us to define the road map for interim LEO extension. With that, I wrap up here, and Christophe and I are ready to take your questions. Thank you for listening.

[Operator Instructions] We will take the first question from line Roshan Ranjit from Deutsche Bank.

I've got 3, please. Eva, you mentioned and highlighted in the release the increased vigilance on GEO and Christophe mentioned the headwinds. Is this driven by kind of cannibalization of the GEO business from LEO? And when -- I guess, if we think about the LEO business, that ramp-up, is that enough to kind of offset the new kind of GEO decline that you are highlighting today? Secondly, and tied to that around the normalized GEO CapEx, you've provided the envelope for the LEO extension. EUR 2.2 billion over 4 years, so roughly just over EUR 500 million a year. How much GEO CapEx should we layer on top of that? Are we still thinking over the midterm, EUR 700 million kind of EUR 800 million annual CapEx prior to IRIS2 started or IRIS2 CapEx started? And lastly, just a quick question. We've seen a lot of noise around the C-band development, potential new auction. How much C-band do you guys actually have that you could monetize?

Thank you for your questions. Those are important. The vigilance on GEO is, of course, a key element. In terms of the headwinds, I think, especially in what we would call B2C connectivity, it is clear that that is where especially Starlink is picking up a lot. And by the way, also driving a bigger market. So we see in our market estimates, actually market estimates or addressable market for B2C connectivity from space actually being increased continually. And in that segment for B2C, it's very clear that we see headwinds. And as you know, we do have KONNECT VHTS satellite, which was specifically targeted for B2C connectivity over Europe and Africa, and that's where we've seen those headwinds specifically there. And I'd also say we are looking at alternative losses, especially within the Mobility segment for the KONNECT VHTS satellite. So we hopefully will be able to -- or we will be able to use it also for other things than pure B2C connectivity. So yes, we do see, especially in the B2C segment, a move towards LEO on the connectivity. Now you'll say B2C is not our key focus, and that is absolutely right. We are a B2B player. But we do also see Starlink starting to come into some of the B2B segments as expected. We'd expected that Starlink would also start addressing the large B2B market. So that's seen in mobility, especially in maritime mobility. And you've also seen some of the announcements on Aero, where we also have a very strong backlog on Aero and Starlink is starting to come in there as well. So it is clear that we do see some of that also being a switch between GEO and LEO in the connectivity market. We still expect to see the connectivity market being a very strong growth market in total. The LEO extension and the GEO CapEx is, of course, and just given your first question, you'll be happy to hear that, of course, we're being very vigilant on our GEO CapEx going forward because if it's a switch we're seeing and we're seeing the market trend slightly more negative on GEO compared to LEO, that is a key one. So that is something we're looking at in the next -- over the next 5-year period. And we have also in our outlooks there, pinpointed a few places where we'll be very vigilant on the GEO CapEx in terms of how we renew it. That goes, of course, for Video, which has been a trend for a while with mid-single-digit declines. So of course, there, everything we can do in terms of being smart about how we renew long-term strategic assets, and that will go also for GEO connectivity. On the C-band, we do note the recent comments by the FCC. We don't know exactly when or if at all, it will be repurposed. We do have the capacity on 3 satellites, 172, 115, 170, and the later 2 are over the KONNECT that is relevant. So we will have a little bit less than we had previously because we no longer have the 113, which counted in the previous C-band round, but we will have our -- if it moves forward, we will, of course, have our relatively small part of the C-band. The majority, as you know, is with SES and Intelsat should this move forward. But right now, we simply don't know whether this has been just a discussion. By the way, you know the discussion has been going up for at least a couple of years now. So still no confirmed time line on it.

We will take the next question from line Mark Watts from Citi.

Just a few points here. So on the ground terminal proceeds, can you just clarify what the lease cost will be going forward on a cash basis? And also of the EUR 500 million cash proceeds that you mentioned, what the intended use might be there? The second one is just on the bond side on the refinancing and capital allocation. So how are you thinking about the kind of the '27 wall? It's fairly sizable out to '29. So keen to understand how you think about the interest cost side of the business if I could.

Okay. So let me maybe start with the second one and come back. I'm not quite sure I got your first question. But the second question is, of course, we have a '27 refinancing. And as we put out, we are looking at various financing options. Now we have both the IRIS2 CapEx and also our bridge CapEx clear. It is looking at the various financing options is one of them that can go through partnerships. Of course, we're exploring quite a lot of export finance as well. And also how we time the -- especially the IRIS2 CapEx will be towards the end of the period. So with that, we are looking at various financing options actively in this moment. Christophe, do you want to take the one that I didn't get, maybe you got it?

Yes. Maybe before answering precisely on the question, we also have to take into consideration that the impact of the operation will be a decrease also of the CapEx, and this is very important to notice. We estimate the annual cost around EUR 70 million to EUR 80 million, most of which actually should not be lease costs, should be OpEx. Just a small part of the cost would be treated or should be treated as leases.

Yes. So the total is, of course, that we will have these additional annual costs, a mix of OpEx and CapEx, and an overall lower CapEx envelope because we no longer have to invest in our ground infrastructure. That will be picked up by the company, and then around EUR 500 million cash in. So those are kind of the key elements.

Got you. And then, sorry, can I just clarify that at the end of the year, there were obviously some headlines around system outages? Do you mind just clarifying what the issue was?

Yes, we had a software glitch with a supplier, which meant that the ground infrastructure became out of sync with the satellite. Ground infrastructure was operating perfectly and satellite infrastructure was operating perfectly, but the match between the 2 were off in a date. This was simply a software supplier who had forgotten we had a leap year last year, not a very brilliant thing to do, but we did experience an outage of around at least 24 hours while we got that reset and got the 2 dates back in sync. We estimate penalties and costs being around EUR 1 million of this. And of course, we have a bit of a heated discussion with our software supplier on this point. It doesn't seem to be rocket science to know that last year was a leap year, but it seems to be that way.

We will take the next question from line Ben Rickett from New Street Research.

Firstly, just a follow-up question. You mentioned you were hoping to use ECA financing. Could you talk a bit about how much of the EUR 2 billion to EUR 2.2 billion of Gen 1 continuity CapEx you would hope to finance with ECA? I think this time last year, you were talking about two-thirds to four-fifths. So is that still the right sort of range?

I think pretty much, yes.

Okay. That's helpful. Second, another sort of hopefully quick CapEx question. On Slide 24, you're showing the IRIS2 CapEx. And from the diagram, it almost looks like it's out beyond 2030. I was just wondering if you could clarify the sort of exact phasing of when you're expecting to pay the IRIS2 CapEx.

I think we expect the IRIS2 CapEx, which is on the order of magnitude of EUR 2 billion, starting a bit in '28. '29 will have a chunk, but then '31 and '32 will also have it. So one of the key advantages of IRIS2 for us is that we'll be able to push the CapEx much closer towards the go-live date. So we have a little bit cut in '28, and then '29 will start stepping it up. '29 and '30 are the 2 bigger years.

That was great. And if I could have sort of one final question. It was just on the OneWeb trajectory. So it looks like you did OneWeb revenue of about EUR 40 million in H1. You're saying the OneWeb backlog is still at around EUR 1 billion. I'm just wondering, what gives you confidence that you will see OneWeb revenue ramping up towards the EUR 6.5 billion that you mentioned?

No, the EUR 6.5 billion is IRIS2. I think you're talking about the IRIS2 billion, right? The EUR 6.5 billion revenues are over the concession period out of the 1.5 terabits of capacity that's in IRIS2. As you know, that will be a subpart of our LEO constellation. That will be additional capacity to Eutelsat. So that's where the EUR 6.5 billion comes in. LEO revenues are ramping up well. We don't split our connectivity revenues into LEO and GEO, but fully recognize numbers where the LEO revenues are ramping up nicely. That is part of how we can also confirm our full-year guidance.

We will take the next question from [indiscernible].

I just have a few questions that are probably revealing that I'm a relative beginner on satellite. But could you tell -- remind me again what led to the relatively sudden drop at the moment in CapEx in the last quarter? And where your current rollout is of that Gen 1 LEO generation and it seems to be delayed. Where are you there in resolving that delay? Have you resolved it? Or is perhaps the drop due to ongoing delays? I'm not entirely sure there. Please correct me if anything I'm saying is wrong. And another question would then be one with revenue. Is that all equipment still? Or is it already capacity? And then my last question would be obviously, that OneWeb constellation as you're outlining it, and with that second-generation or additional satellites of, I think, Gen 1 that you're looking at here of 340 additional Gen 1 satellites. That is still going to be a lot smaller than, say, your big competition out of the U.S. How are you intending to compete there? Can you remind us of your strategy there, and how that's going to work?

Those are pretty big questions. But let me start out with the drop in CapEx. The drop in CapEx is as much in this financial year, which, as you know, by the end of June. So there's some of that, that, of course, flows into next year. So, there's an element of it, which is the exact payment schedules of the orders of the next satellites, which influences this, so I think that is the good chunk of the drop in CapEx. There's also some CapEx linked to the finalization of the LEO rollout. As you know, we've been talking into gateways. We still have around 5 gateways to go in some of the complicated places like Tanzania, Senegal, and Martinique, and those are flowing into the second half of the year, which also means that that impacts some of our coverage that is driven by that. That's linked to delays in the partners that are helping us building those 5 hard gateways. We have 39 now passing commercial traffic. So that has started to go live, but the last 5 are still dragging their feet a little bit into the second half of the year. So that is the same story on the delay in the rollout, but that also moves CapEx into a slightly later period. In terms of OneWeb revenues, no, it is certainly mostly capacity, service revenues are what we call it. There is a bit of UTs in there, but it is a smaller part of the OneWeb revenues. The major part is certainly service revenues in the OneWeb revenue numbers. And then finally, in terms of the additional satellites, as you know, we fly higher than Starlink and Amazon. So they are around 400 to 500 kilometers, which means that just to cover the earth, they need many more satellites, just physics, if you want to point to any point on Earth. So we only need around 450 to 500 satellites to actually cover and have global coverage. And that also means that we don't need the thousands of satellites simply to have global coverage. We have a B2B-focused strategy as well, whereas Starlink has started out with a very standardized B2C product, so attacking kind of consumer broadband, driving volume in terminals, whereas we have opted for a B2B strategy where we work with distributors in the market. So those are 2 very different approaches. Now Starlink is also starting to address the B2B market just as we do address B2C market in a much smaller scale. So we are currently the only alternative to Starlink as Kuiper and Lightspeed is not expected before '28 earliest. And I think we see most customers not necessarily want to be in a monopoly situation. Lots of customers are putting a lot of importance also to have non-U.S., non-Chinese alternatives. So this is not a question of only resilience. A lot of customers want multiple networks for resilient purposes. And some customers simply want to have alternatives. One example is our large contracts recently signed with the Taiwanese government for coverage on Taiwan. They wanted to have an alternative to the Starlink. We see that in quite a lot of places. And of course, we also see especially European and other MODs being very interested in having all the alternatives available, not only for kind of geopolitical reasons but of course, also for resilience. So I think the main point is you want to have access to multiple networks for resilience purposes.

I just have 2 tiny follow-up questions there. How high do your satellites fly? And how many do you need to guarantee service levels?

So that's simple. It's around 1,200 kilometers. And I think I said that, but we need around the 400 mark to have full global service and a bit extra. So that will give a good global service everywhere, poles, equator, everything else. Now we have around 600 flying right now, which enables you to have slightly better elevation angles. The big question is what kind of elevation angle -- or you need a decent number of satellites if you want to have flat panels. Parabolic antennas are a little bit less -- a little bit easier to operate with fewer satellites. So that's one of the key questions when we have but with 400, we have good elevation angles on flat panels, which is what customers want today.

We will take the next question from line Sami Kassab from BNP Paribas.

I have a few questions, please. The first one, you talked about the several hundred millions of take-or-pay commitments from the EU on IRIS2. Is that on the Eutelsat part or on the whole program? And can you be more specific on the compensation mechanisms that should ensure minimum returns on the project and possibly share what the minimum guaranteed rate of return is on your EUR 2 billion investment? Secondly, you have suggested that geo connectivity trends are perhaps a bit weaker than expected. Yet in the press release, you suggest that the cessation of revenue recognition on KONNECT VHTS is said to be temporary. When would you resume recognizing revenues? And why is it temporary?

So let me start with the commitments of IRIS2. There are some take-or-pay plans for especially HardGov and SoftGov capacity for EU. That is not yet distributed, partly because we are talking about capacity that's going to be needed beyond, let's say, from '31 and on. So it's a bit early, and it will be a mix of kind of commercial and of course, the KaMil HardGov competence. So that's a total for the entire consortium. And of course, it can be -- if it's LEO, it's more likely it will -- that's where SES has its main interest. If it's LEO, it's where we have the main interest but it is not finalized yet. That is part of the discussion. The way the -- I think the other one was how the compensation was going to run or how the protection was going to be for the internal rate of return. And I think you know the mechanism of this first on the viewpoint, which is in about a year's time, which is a key element of having a confirmation of both time line, cost, and quality. And we have an exit option if we don't, at that time, see a supply chain that can meet those obligations. The same thing, by the way, of IRIS2. So right now, we're in the process of confirming the supplier setup, which is probably what you'll hear from a lot of the Airbuses, OHBs, and others that they are in competition to enable us to ensure that we can build this constellation on time and within the EUR 10.6 billion mark. So on the KONNECT VHTS, we had initiated a discussion with TIM. TIM is one of our customers on Eutelsat KONNECT and Eutelsat KONNECT VHTS. TIM has not started migrating customers from KONNECT to KONNECT VHTS as was the original agreement and is debating whether they have an interest in that. And that's why we have stopped recognizing revenue with TIM. We have a multiyear take-or-pay with them. So we are right now in discussions with them on how to bring that into the future.

And on IRIS2, is there a minimum IRR that you can share on your investment?

Yes, it's in line with our typical IRRs of 10% to 12%.

And that's the minimum that's in line with 10% or that's the expected one?

No, that's a minimum one. That's the one we expect to at least have.

[Operator Instructions] We will take the next question from the line Aleksander Pet from Bernstein.

I just have a little follow-up on CapEx first. So the EUR 200 million shift of CapEx out of this year, is that going into later years? Or did you just reduce the entire approach and reduce the amount to spend here? And then when you talk about increased vigilance on GEO CapEx, is this a warning that you may have to drastically reduce GEO spending going forward given the disruption in the market from LEO? Is that what's going on? And then I have a follow-up.

I think on the CapEx, we see a lower CapEx this year. We don't necessarily guide multiyear CapEx on it. I think there is some switch, which is simply the payment schedules and some of the ground infrastructure rollout that we will have in the second half of the year as well. But there is also a real reduction in it. And I think on GEO CapEx, what you'll see is that we will be looking at many types of partnerships. You saw the partnership we have with Thaicom as an example in a way of launching connectivity capacity in partnership, in this case with Thaicom, where we share a satellite rather than launch 1 each. And I think those are the type of partnerships, which both enable us to lower our own Geocapital or simply share the risk with other players. So those could be types of ways of being vigilant on future GEO CapEx spend.

And then the follow-up will be on your comments regarding the B2C disruption of Starlink. And you seem to be implying that prospects for KONNECT VHTS on the B2C side are materially weaker now. Does that mean that at least part of the capacity there is going to be a write-off or very little occupied? What exactly is going on there?

No. So KONNECT VHTS was originally meant for connectivity over Europe. And we actually are providing a lot of connectivity with KONNECT VHTS both in Spain, Switzerland, and France and a lot of the capacity that you, as an example, see, Orange has been doing a very strong job on migrating their satellite customers to KONNECT VHTS and allowing the much higher bandwidth with actually quite some solid success because it's a very strong value proposition for the consumer segment. So it's only TIM that has opted not to do an active migration into the KONNECT VHTS. However, we will probably not sell all of the capacity. It's a 500-gigabit satellite KONNECT VHTS. So what we are doing right now, we are developing terminals for also mobility use. As you know, we've seen quite a lot of traction in the mobility market, and it's a market that's actually very hungry for capacity, whether it's for land mobility, whether it's for maritime or for aero, that is the segment that is really increasing capacity requirements because it's a segment that simply gets other options. So for both -- for the mobility segment, we need slightly different terminals than the B2C terminals that we had for KONNECT VHTS, and adapting that satellite to be able to serve the mobility market is happening as we speak. So we're in the first test of that. So we will be able to leverage that capacity into the mobility market, which is actually not bad news in terms of pricing because it's typically a market where prices have been better. However, we do need to get the capacity out there as it's a segment that is asking for it.

And then my last question would be on the opportunity for LEO connectivity in mobility aero markets in IFC. I've seen Starlink taking sizable deals there, not least Air France-KLM, Qatar, and United. So my question is, do you have a road map for OneWeb serving these markets? Do you have any wins in this area? What's the time line for developing this vertical?

Yes. Thank you for the question because we actually do. It's been one of our core segments for a long time. We have a backlog of around 1,000 aircraft that are scheduled for installation. You've probably seen, I think we put that out on LinkedIn, some of the very positive early installations with Air Canada, but we also have American Airlines, Alaska, and Japan. It is not us direct. We, as you know, are working with distributors. We have a B2B approach. So you'll see that through Intelsat, Panasonic, even Viasat are using our LEO capacity for their aero services. So we are working with all of them and we have now Gogo business jets for smaller jets is also one of them who started to fly with our capacity. So we have a very strong backlog actually in Aero for our LEO services. And then for Europe, we'll be able to put some KONNECT VHTS on top of that and that will be even stronger. So aviation has been a very strong backlog element for us. Why it's taking a bit longer for it to start to actually generate revenue has to do with the different approvals of the FFA for terminals and of course, the test flights and then, of course, just installation on the fleet. But as I say, we have more than 1,000 aircraft in the backlog and Starlink is more recent into the market. Again, not a surprise that Starlink will start being interested in the B2C market, which has typically been 3/4 of the satellite market. And also in Aero and you saw especially the United deal where they, I think, are also going to install themselves, and then with Air France, which I believe is more that they are bringing Starlink into the mix, but not a full replacement. But I would expect to see many aircraft carriers and distributors would want to take in the different types of capacity, just like we are seeing in the maritime market that a lot of our distributors have both OneWeb and Starlink capacity and mixing the 2.

So just to understand, so the strong backlog you have in mobility in LEO, is that going to transpire in your revenue trajectory in what time frame? Is this next year's event or longer? How long does it take?

I'm not sure I understand your question. Can you just maybe reformulate or--

Yes. Just to formula then. So looking at mobility, it fell well short of expectations in the reported quarter. So my question is, at what point will your strong backlog in LEO connectivity for Aero start to transpire in the mobility top line?

So Aero we'll start seeing some of it in the next financial year, start to ramp up in the next financial year, some from the summer. I mean it's going to be very small over the summer and then it's going to start ramping up with the installation on the fleet over the next financial year.

We will take the next question from Stefan Diehl from ODDO.

I've got 2 if that's possible. The first one is related to CapEx. Would you be able to give us an idea of how big is going to be your GEO fleet, let's say, I don't know, but let's say, at the end of this decade versus today? I'm just trying to understand to what extent you are redimensioning the GEO fleet. My second question is regarding IRIS2 again and just a follow-up. Could you give us more color on how the revenue calculations have been made? I guess what I'm trying to understand is how much extra EU state spending you're expecting versus what the states are actually spending today.

Those are big questions. What I can tell you by 2030, we're probably still going to have around 35 satellites flying because we are talking much longer investments. I think your question, and I'll pass on the actual CapEx to Christophe is that what we are and that's just the rhythm of GEO satellites is that within the next year, we're going to have to start thinking about the satellites that will go into service by 2030, 2031. So that's just the rhythm of how long it takes to develop, launch, bring into orbit, test into orbit, and then actually start migrating customers on it. So we're on some very long things. But of course, and I think that's what we're looking at closely with Christophe on is that let's not spend on something that's going to go into service in early '29 but more likely '30 and '31, let's make sure that we don't overspend because they are going to be in service for 15 to 20 years after that. So, do you want to put a word or 2 on GEO CapEx and how we think about it?

Yes. What we are looking is, obviously, we have orbital positions that we need to keep and use, exploit. So we are also thinking of different ways of different types of either satellites, either smaller satellites or partnerships. But clearly, if your question is related to how much CapEx we would spend in the coming years to replace the current fleet, it will be obviously much lower than in the past for the reasons that we have just mentioned. And just to give an idea of the range, it would be below EUR 200 million.

We will take a follow-up question from Sami from BNP Paribas.

I have a question on India. Given that one of your largest shareholders is a large Indian telco, given that India is a large, perhaps attractive market, can you please elaborate and tell us where you stand with market access rights into India and perhaps the business plan for India and how much revenues might India contribute to by the end of this decade?

I don't know where India is going to be by the end of this decade but it is clearly one of our core markets. We right now have some testing ongoing for specific use cases and they've opened up for actually allowing for kind of governmental and remote connectivity use cases in India. So we're the only one testing there. As you know, neither Starlink nor us have actually full market access rights in India yet. That's part of a geopolitical trend we see in quite a lot of places where governments are spending a lot of time studying what LEO connectivity will and which conditions they want to give market access. And so India, we expect to open up, especially for remote and government uses, and we are already testing. We've had a couple of quite successful tests with the Indian military. And of course, we are helped by one of our large shareholders and collaboration with Airtel on this. So partly Telecom and Airtel are close collaboration with them. And we have actually a significant take-or-pay on India already, which will kick in at the time where we actually gain market access to India. So we have a backlog sitting in India, which is ready to go live as soon as we get market access. We have the gateways in India up and running. So it's simply a question of regulatory approval of that market. So we'll likely get it at the same time as Starlink is my best guess, but it's been quite a long administrative process with the Indian regulator.

And I imagine you would not disclose the size of the take-or-pay contract in the backlog, would you?

No, but it's quite significant because India is a big market with quite a lot of interest as well.

It appears no further questions at this time. I'll hand it back over to your host for closing remarks.

Well, thank you. Thank you for showing up this morning, numerous and with some great questions, some very strategic and some more operational. Just want to take the opportunity to sum up the year. Strong revenue and EBITDA margin in line with expectations and objectives confirmed. CapEx lower with about EUR 200 million less than previously guided. And 3 very strategic steps forward. First of all, IRIS2, of course, which allows us to pave the wave of continuity and increased functionality on our LEO constellation towards an IRIS2 and the signature of the put option with Stargate, which is also on track for '26. So all in all, a year that was busy both on the operational, but also on the strategic front. With that, I just want to thank you for showing up this morning and have a wonderful Friday. Take care.

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