Blue Origin’s New Glenn rocket did something genuinely impressive on its third flight — and then immediately did something genuinely frustrating. The reusable booster stuck its landing, marking the first time Blue Origin successfully recovered and reflew a New Glenn first stage. The upper stage, meanwhile, missed its target orbit and released a cellular broadband communications satellite for AST into the wrong place. One half of the rocket worked beautifully. The other half did not. That tension is worth sitting with for a moment.
As someone who spends most of my time reviewing AI toolkits — things that either work or don’t, ship features or break them — I find this story oddly familiar. It’s the classic “partial success” problem. You build something with two moving parts, one of them performs exactly as promised, and the other quietly undermines the whole point of the exercise. The demo looks great until you try to actually use the output.
The Part That Worked
Let’s give credit where it’s due. Reusing a booster is not trivial. New Glenn stands 321 feet tall, and getting that first stage back in reusable condition — ready to fly again — is a real engineering achievement. Blue Origin has stated they expect the New Glenn booster to be reused every 30 days to support their 2026 launch cadence. That’s an aggressive target, and the fact that the booster performed on its second flight is a meaningful data point toward hitting it.
In the toolkit world, this is like a platform finally nailing its API reliability after months of flaky calls. You notice. You appreciate it. You just don’t pop champagne yet, because the API is only one layer of the stack.
The Part That Didn’t
The upper stage failed to place its payload — a satellite built for AST’s cellular broadband communications network — into the correct orbit. Blue Origin confirmed the upper stage missed its aim and released the satellite anyway. That satellite is now in the wrong place, which for a communications satellite is roughly equivalent to shipping a product with the core feature broken.
This is the part that stings. The whole reason you build a reusable rocket is to reduce the cost of getting things to space reliably. If the booster comes back but the payload doesn’t reach its destination, the economics of reuse don’t matter much to the customer whose satellite is now drifting in a non-operational orbit.
What This Looks Like From a Product Angle
I review tools for a living. When a tool has one component that’s solid and another that fails silently or catastrophically, the review doesn’t average out to “pretty good.” The failure mode defines the user experience. Nobody remembers that the authentication flow was smooth when the data export corrupted their files.
Blue Origin is in a similar position here. The narrative around New Glenn’s third flight will be shaped by the satellite loss, not the booster recovery — even though the booster recovery is the harder, longer-term technical win. That’s just how product perception works. The thing that broke is the thing people remember.
Where Blue Origin Goes From Here
The honest read is that Blue Origin is still building toward something real. A 30-day booster reuse cadence, if they can actually hit it, would put meaningful pressure on competitors and bring launch costs down in a way that matters for the broader commercial space market. That’s a legitimate goal backed by a legitimate technical milestone.
But the upper stage failure is a problem that needs a clear post-mortem and a visible fix. Customers — especially satellite operators — need to see that Blue Origin understands what went wrong and has a credible path to preventing it from happening again. Confidence in a launch provider isn’t built on one good booster landing. It’s built on consistent, end-to-end delivery.
- The booster reuse worked — first time for New Glenn, and a real milestone
- The upper stage failed to reach the correct orbit, and the payload was lost
- Blue Origin is targeting 30-day booster reuse cycles through 2026
- The mission’s outcome will be defined by the failure, not the success
Two things can be true at once. Blue Origin made genuine progress on one of the hardest problems in rocketry. They also failed to deliver for their customer. Holding both of those facts together, without letting one cancel out the other, is the only honest way to assess where they actually stand.
🕒 Published: