Alright, kids, gather ‘round the whiteboard. Today’s kata is “How to Pivot a Satellite Company While Pretending Nothing’s on Fire.”
We open the sprint with Eutelsat rolling out LEO revenue like a proud junior dev showing their first green test. Revenue up 70%? Nice. Of course, they don’t mention it was up from a pretty small base. Context is a feature, not a bug.
Now here’s the twist: while LEO’s flying, their GEO segment (the old reliable monolith) just dropped another 10%. That’s what you get when you architect your business around broadcast TV in the streaming era. But don’t worry, they buried that decline under a pile of acronyms and sovereign deals. Solid obfuscation pattern. Seen it before. Still compiles.
Connectivity is the new main thread, kids. They’re calling it 53% of revenue now. If you squint hard enough, that’s a pivot. LEO’s doing the heavy lifting. GEO? That one’s throwing exceptions and quietly deprecating itself. Video (still 47% of the revenue pie) is the legacy system that throws warnings every time you run the app. They say BHS renewed. Great, but “renewed” isn’t “scaled.” You don’t brag about keeping the lights on unless you’re trying to distract from the fact that you’re rewiring the entire grid.
They’re booking wins in weird, wonderful places. Greenland? India? Ukraine? That’s a bug tracker where only the edge cases pay their invoices on time. Sovereigns love LEO because fiber doesn’t love mountains, war zones, or subzero tundra. When your best customer is a government, you trade margin for mission assurance. That’s subsidy dressed in a parka.
Let’s talk cash. IRIS² is bailing them out this quarter. Five million in “other revenues” includes EU program funding. That’s milestone money, not recurring revenue. You get it once for checking a box. Feels more like a grant than product-market fit.
And TIM, the Italian telco that used to pay for broadband from Eutelsat’s KONNECT-VHTS? Gone. They pulled the plug in January. No funeral. Just a line item. If GEO broadband in Europe was a viable product, someone would be using it. Instead, it’s like a class with no methods. It exists, but it does nothing.
Their backlog? €3.5 billion. Half of that is LEO now. That’s the good part. It means the new stuff is finally doing more than demoing well on stage. But don’t miss the fine print: backlog didn’t grow this quarter. You burned what you booked. Momentum is velocity over time. And right now, that velocity looks steady, not accelerating.
Then there’s the capital raise. €1.5 billion of state-sponsored, government-endorsed, industrial-policy-flavored runway extension. France, the UK, Bharti, CMA CGM, every strategic partner with a checkbook got the call. If that sounds like a business plan, it’s not. It’s a survivability patch. Eutelsat didn’t go to the market. They went to their board, then to their patrons, then to their press release.
They’ll tell you this is sovereignty. What it is: a bailout with a marketing department.
Execution? Still in pre-prod. Satellites need launching. Ground segments need scaling. Margins are taking a hit because onboarding a constellation is like onboarding a legacy enterprise client: painful, expensive, and absolutely non-negotiable. They say EBITDA margin will dip “slightly.” In CFO-speak, that’s “hope the rocket lands on time.”
Let’s not ignore the terminal problem (literally). LEO constellations live or die on terminal pricing and availability. It’s one thing to promise gigabit-from-space. It’s another to deliver it in remote Norway on a Tuesday with snow on the dish and a customs delay in Helsinki.
So what do we learn?
- GEO is your slow, stable dependency that now throws more warnings than logs.
- LEO is the hot new module that works great in dev, still shaky in prod.
- Governments are the test users who’ll pay to click around before GA.
- Europe is betting the house on not outsourcing space infrastructure to Elon.
- The build is stable, but the deployment pipeline is held together with duct tape.
The strategic roadmap makes sense. Pivot from broadcast TV to secure LEO connectivity. Sell to institutions, not consumers. Anchor the company in sovereign contracts and let someone else worry about churn. It’s smart. It’s survivable. It’s state-sponsored agility.
But don’t mistake a running backlog for running code. This quarter worked because IRIS² paid up, not because customers beat down the door. When that dries up, LEO’s got to stand on its own legs. Preferably before GEO fully craters.
Final score?
LEO is the future. Good architecture, terrible unit tests.
GEO is the past. Still compiles, but throws compiler warnings.
Revenue mix is finally changing. That’s the signal.
Video isn’t dead yet. It’s just in hospice with recurring invoices.
Government money is the Band-Aid. Let’s hope it lasts through launch.
Now push the code, kids.
And pray the next satellite passes QA.
