Ten minutes. That’s how long the first of Amazon’s 38 booked Vulcan flights lasted before the whole thing came back down. The Vulcan rocket failure on Sunday took 40 Amazon Leo broadband satellites with it, and it’s the first total mission loss in United Launch Alliance’s twenty-year history.
ULA’s statement on X is about as warm as a server log: the vehicle “suffered an anomaly 10 minutes into the flight and failed to deliver the Amazon mission to orbit”. Cause unknown for now, investigation ongoing, transparency promised. You’ve heard this song before.
Vulcan rocket failure: a bad day for everyone but SpaceX
Amazon Leo (the thing that used to be called Project Kuiper) is supposed to be the Starlink competitor. So far it’s been the Starlink competitor the way my cousin’s five-a-side team competes with PSG. Only a small slice of the constellation is up, and Sunday was meant to be the start of the real ramp: 38 flights bought from ULA, each one dropping dozens of satellites into orbit.
Now that rocket is grounded for an undefined period. Vulcan already had a rough track record, with its strap-on solid boosters misbehaving on two of its four previous flights; it survived those, but each one triggered months of reviews. As The Verge reported, Amazon kept a brave face: “There is risk with any launch, and that’s especially true for inaugural missions like LV-01. ULA has been a trusted partner for Amazon, and we believe they’ll learn from this mission.”
Sure. But Amazon’s problem isn’t trust, it’s the calendar. Amazon’s FCC licence comes with deployment milestones for the constellation, and every grounded rocket pushes those numbers the wrong way. Meanwhile, the only company with a rocket flying every few days is the one Leo is trying to beat. Amazon has even bought Falcon 9 launches from SpaceX before. Expect more of that, and expect Elon to enjoy it.
Why you should care about Amazon Leo
Because a satellite internet market with one serious player isn’t a market. Starlink’s prices, its terms, the way it can be switched on or off over a conflict zone: none of that gets challenged without a credible second network. I’m no fan of Amazon’s habit of closing accounts over a few GPU orders, but on this one I’d like them to succeed.
There’s also the military angle. Vulcan is on contract for roughly half of the US national security launches in the coming years. A months-long grounding doesn’t just delay Amazon’s broadband dreams; it stacks up Pentagon payloads too, and they all end up knocking on the same door in Hawthorne.
Forty satellites, one lesson
Forty satellites sounds like a lot until you remember Starlink has thousands up there and treats a batch of 40 as a Tuesday. The real cost is time. Hardware you can rebuild; a launch slot on a rocket that works is the scarce thing, and we’ve seen this year how fast infrastructure assumptions crumble, whether it’s a cloud region losing data to drone strikes or a global outage with no explanation for a day.
Betting a constellation on a rocket with four flights and two booster scares was always going to be nerve-racking. Now Amazon gets to find out how patient the FCC really is, and how much a Falcon 9 ride costs when your rival knows you have no choice.
Space is hard; having a single working rocket to call is harder.
Discover more from TechPassion
Subscribe to get the latest posts sent to your email.





