Eutelsat Group has pinned its future on multi-orbit broadband services that it hopes will galvanize its expansion out of the declining satellite TV market.

The French company’s acquisition of UK-based OneWeb’s low-Earth orbit constellation last year made it the only global operator with wholly owned satellites in LEO and geostationary orbit (GEO).

In addition to low-latency capabilities, OneWeb’s more than 600 LEO satellites promise to give Eutelsat global broadband coverage, although ongoing ground segment issues have scuppered plans to begin offering full services in early 2024.

Despite the delays, Eutelsat has amassed more than $700 million in back orders for OneWeb capacity from government and enterprise customers.

Eutelsat Group CEO Eva Berneke. Credit: Eutelsat

One of the biggest customers is GEO operator Intelsat, which agreed in April to be sold to Eutelsat’s European rival SES, an operator of satellites in geostationary and medium Earth orbit.

Eutelsat Group CEO Eva Berneke tells SpaceNews where else demand for OneWeb’s services is coming from and what it means for plans to start building
a second-generation LEO constellation this summer.

The current generation was provided by Airbus OneWeb Satellites (AOS), a Florida-based joint venture OneWeb had jointly owned with Airbus until Eutelsat sold off its share early this year.

However, rather than going all in on a second-generation constellation, Eutelsat recently opted to deploy upgraded satellites progressively to reduce technological risk as operator consolidation — and the growing dominance of SpaceX’s Starlink LEO network — rock the industry.

The pivot to a stepwise strategy comes as Europe considers a proposal for a sovereign multi-orbit network from a consortium that includes Eutelsat.

How does Intelsat’s planned sale to SES affect the competitive landscape for multi-orbit services?

Berneke: Generally speaking, consolidation amongst legacy satellite operators has been ongoing for several years now, driven by motivations including synergies and critical mass. Because of the nature of the satellite business with long cycles in terms of the procurement and lifespan of the assets, mergers and acquisitions rarely create market repair in the immediate term.

Specifically, in terms of the landscape for multi-orbit services, we do not anticipate that it will be significantly altered by the combination of Intelsat and SES in-orbit assets. Intelsat already has access to non-GEO capacity through the deal recently inked for LEO capacity on OneWeb.

Our GEO-LEO solution already offers customers capacity, global coverage, resiliency and the unique low latency that enables real-time applications, which we think really sets us apart.

Where is Eutelsat OneWeb on its journey to full global services?

We’re providing services in quite a few places. North America, most of Latin America, Australia and Europe. We’re starting services in Africa, Asia and the Middle East over the next couple of months as our gateways go live. Of course, we have some areas we will never cover, like China and Russia.

There will still be a few gaps in some oceans by the end of the year, and potentially in parts of Africa because there are a couple of gateways there that have been delayed more than just a month or two. But we should be up to 90% of global coverage by the summer.

What is delaying ocean coverage?

We’ll probably have a gap or two in the Pacific because we’re missing fiber there. A sea cable is probably not going to make it to the island we had in mind in time, so we either need to find a different exotic island with fiber or some other way to connect the area.

But it’s not always a question of coverage. It can be a question of national regulatory approval, and that’s a country-by-country thing. For example, we are still waiting for regulatory approval in Japan, which has been ready to go for a while with customers lined up and backlog there. We’re currently in the last round of interference tests with mobile operators to get regulatory approval in Japan.

We are also in one of the final phases of the process to get regulatory approval in India.

OneWeb satellites enable global coverage by operating in near-polar orbit. Credit Eutelsat Group

Which upcoming market has the most potential for you?

Some that we have already brought online represent super important segments. The mobile backhaul deal with Telstra in Australia highlights how telecom operators can put up a mobile mast in an area that would never get fiber.

Right now we’re seeing a lot of capacity going toward maritime, especially in the cruise segment.

On the other end of the range, only really coming in 2025, is aero multi-orbit. We’re getting through tests with multi-orbit terminals on airplanes and I think that’s a huge market.

A lot of the take-or-pay contracts in our $700 million backlog sit in countries that have yet to go live. We have some in Saudi Arabia and India, which have great potential for coverage of white zones, giving connectivity to people who do not have it today.

What kind of customer makes up the bulk of that backlog and where?

There is a good chunk in mobility. That’s probably around a third split, between aero and maritime connectivity. And we need a decent chunk of global coverage to be able to serve these mobility customers — oceans, airplane routes and all of that.

Then we have a few areas where some of our partners have taken large geographical commitments across Asia — Thailand, Japan, Saudi Arabia, India — where they want fixed capacity. We have a lot of telcos that have signed up for fixed capacity in a geographical area, like Telstra, and that represents probably about another third of backlog.

The remaining third is for mixing in GEO capacity. Several GEO customers Eutelsat Group already had are part of this, and then we have 10-20% in government contracts. Not just in military, but also applications such as public safety.

Are any areas of the backlog accelerating faster than others?

The hunger for capacity in the mobility segments is growing really fast.

How will the multi-orbit terminals now becoming available drive that demand and where?

We’ve seen quite a lot of interest for the resiliency that multi-orbit gives. Resiliency is a buzzword for government and military users, and that makes a ton of sense. You always want to have the best of all worlds and potentially all worlds in one. You definitely don’t want to go without a backup solution.

We’re also starting to see how a multi-network is a really strong value proposition in mobile backhaul and in mobility. A ship can pretty much have as many terminals as they want, but a plane has limited space and needs a small form factor.

A single hybrid terminal is currently still far more expensive than two mono-network terminals, if you have space for them.

If you have the space just install two terminals. It’s a simple solution because terminal price has been driven down on a single terminal by single terminal basis. We now need to bring down the cost of multi-orbit terminals.

Why did you pivot to a progressive Next Gen deployment?

The original plan was to have our Gen 1 and then, boom, we switch over to Gen 2. We would have needed to decide all the features that Gen 2 would need to have and then bring it live, which is great if those features were technologically mature and had a certain roadmap.

Now, if you get engineers into a room, they will always want to have some of the new fancy stuff that is absolutely not mature and would give me a huge timeline risk.

But the question I get most from customers is whether we can provide them with assured capacity, and I need to be able to say, yes, and it’s not going to go away after a year or two because my Gen 2 is delayed. I need to have a more gradual introduction of some of those technologies so I can bet on them being there in 2028.

So, the first two batches will probably only have a few new technologies onboard. These are the batches needed to replace the oldest satellites in the network with an estimated design life ending around 2027-2028.

We have not decided what additional functionality we want to bring but the next four or five batches launched, let’s say, 12 to 18 months later, will probably bring some new elements to them.

The idea is to have a gradual move forward. Even though my investors like me saying that’s actually going to be less expensive in the next five years, it will be more expensive overall because it’s going to last longer. I’m probably going to keep adding and renewing technology when it becomes available.

It also has to do with the advantages of having all of these technologies mature over the next few years. We would have been paying quite a big risk premium and a lot of the nonrecurring development costs. I’m not sure I want to do that if something like Iris2, or another constellation, might be pushing on the same roadmap.

5G, for example, which is probably something most of the sector is going to be chasing. I might as well try to move with the sector, and have the entire sector share the development cost of 5G baseband terminal development, rather than insisting this needs to be ready to go by 2027.

Another example is optical links. When is the price curve going to come to a place where we like it?

If the first satellites are going to be close replacements, does that mean Airbus will help build them again?

That is not certain. They could. We’re running a couple of co-engineering teams right now and Airbus is in one of them. A first step up from a Gen 1 is something others can do as well and we don’t want to be locked in. That’s also why we sold the AOS factory, so we could have a clean slate for the next development.

Is there a date for when you want to select a manufacturer or manufacturers?

Yesterday. No, it’s going to be over the next couple of months. Ideally before the summer.

Does your strategy pivot require more rockets?

I don’t think it actually means more launches. We have quite a lot of options on launches and have made some reservations to spread them out, and potentially also get export financing, which we can get in India and Japan.

A OneWeb-compatible terminal from Kymeta mounted to the roof of a vehicle for testing. Credit: Eutelsat Group

Eutelsat Group previously said it had launches mostly covered for the 300 Next Gen satellites under the initial plan.

It’s also a question of whether our plans fit into the launch manifests at the time we want them launched, or whether they require reshuffle. We have an Ariane 6 somewhere in there. Are they going to be available in the next three to four years?

Given that Amazon has acquired Ariane 6 rockets, if we wanted to use it in, say, 2027, are we going to fit into their launch manifest or not? We might say we’ll take that rocket at a later time and then use one from Relativity Space or whoever gets there in time. It’s a bit of a puzzle. If the puzzle was perfect and I could decide everyone’s manifest then it would be more straightforward.

Has your GEO strategy changed since embarking on a progressive Next Gen LEO approach?

That’s not impacted by the progressive LEO approach, but we are pushing strongly on the GEO-LEO combination. Customers want resilience and GEO-LEO achieves more than 99.5% network availability when combined, so we see better services in combination. 

I sometimes compare it with terrestrial telcos. Your mobile phone is essentially in a network where you’re roaming the whole time, and sometimes the signal sometimes just drops, which doesn’t happen on a fixed line. 

That’s a little bit like what you have in GEO because the connection is always in the same spot, but in LEO you’re roaming to another satellite every couple of minutes and sometimes the connection can just fail.

The evolution of the GEO-LEO strategy is the combination of GEO-LEO. It doesn’t change the GEO approach, it makes convergence between the two the long-term strategy.

When do you see connectivity providing the majority of Eutelsat Group revenue instead of video?

We have a big internal competition around that. Today, it’s around 60/40-ish video to connectivity. Video is falling around 4-5% [annually] while connectivity is continuing to grow. So that will happen. Right now we’re seeing it happen in 2026 or 2027.

What sets your multi-orbit services apart from others chasing this capability, such as Telesat?

We’re there and operational, and that’s a good place to start. I’m not sure Telesat is as focused on the GEO-LEO play and continuing a lot of GEO development and launches. They have a video business in GEO but only a small connectivity business.

Compared to SES, which is trying to do it in MEO, our big differentiator is we’re global coverage with our LEO.

Is Eutelsat interested in adding MEO to the mix?

It’s not the plan right now.

How could Eutelsat Group help Europe meet its IRIS² ambitions and what’s the next step there?

We’re part of a consortium that is the only group that’s really active right now. We’ve handed in what’s called a best and final offer and are waiting for a response back.

I don’t think it responds to all the Commission’s wish list. It was very hard to answer all of them. But we’ll have to wait and see what their response is.

Does it include OneWeb Next Gen?

No, IRIS² is a sovereign constellation but it’s one of the areas we’re trying to find synergies between the technology roadmap that needs developing for my stepwise next-gen constellation.

So it’s about leveraging the same technology development but not using the exact same satellites in the constellations.

This article first appeared in the June 2024 issue of SpaceNews Magazine. The interview has been edited for clarity and length.

Jason Rainbow writes about satellite telecom, space finance and commercial markets for SpaceNews. He has spent more than a decade covering the global space industry as a business journalist. Previously, he was Group Editor-in-Chief for Finance Information...