Everyone in the space sector knows that space is expensive. But fewer stop to ask why. The real answer is actually simple: in space, it’s the mistakes that are expensive. Often prohibitively so. That single fact shapes everything about the business of space, from how spacecraft are designed to how companies are funded. And while the “New Space” movement promised to rewrite the rules, we haven’t yet solved that central problem that makes space so hard to begin with.
That’s because you can’t call a tow truck in orbit (well not yet, anyway). You can’t patch a leak. You don’t get a second try. That brutal truth is still that every mistake, whether in design, integration, or operations, carries a disproportionate cost because every mistake prevents you from learning from any others. So Fail Early in space usually just means the waste of an opportunity.
So, space companies, especially those building hardware, are left with a terrible choice:
- Spend more up front — on design, analysis, testing, and margin — to reduce the chance of mistakes.
- Accept that mistakes will happen and budget for failure, including the cost of schedule delays, redesigns, and in some cases, full mission loss.
Either way, it’s expensive. And that, more than the cost of launch or materials, is why space products cost what they do.
Complexity and the front-loaded nature of space design
What makes this even harder is that space systems are inherently complex. Not just complicated — complex in the project management sense. The development of space mission involves dozens of interdependent subsystems, operating in hostile environments, often with long lead times and strict interface constraints. But, despite that complexity, the design process requires certain key decisions to be made early. Decisions about architecture, power budgets, interfaces, mechanical structure, and launch configuration must often be locked in before downstream design and testing can begin.
In a complex environment, this is a problem. Because, if you follow these columns, you know that complexity naturally breeds contingency. So, as the design matures and real-world conditions start to assert themselves, new constraints emerge. Thermal behavior is different than expected. A component fails qualification. The mass budget slips. A late-arriving payload partner needs more volume or power. These are not just possible — they are inevitable. And when they occur, they often undermine those early decisions.
Then you’re left with two bad options:
- Change the expected outcome — accept degraded performance or a narrower mission.
- Go back and redesign — delay delivery, consume cash, and possibly break contractual or launch commitments.
Both are expensive. Both are disruptive. And both are routine in the development of space hardware.
New Space didn’t eliminate the cost of mistakes
Much of the energy behind the New Space movement was about speed, iteration, and lower cost — principles borrowed from software and consumer tech. The idea was that smaller teams, faster cycles, and tolerance for failure would drive down the cost of innovation.
And it did work in certain limited ways. Launch got cheaper. Hardware got smaller. Teams got scrappier.
But what didn’t change was the cost of mistakes. We just made more of them, more quickly. The economics shifted, but the physics didn’t. The harshness of the environment, the integration risks, and the cost of in-space failure — none of that went away.
The core problem remained: in order to be reliable and repeatable, the space design process must be front-loaded, rigid, and vulnerable to unexpected change. Which makes the cost of being wrong dangerously high. A lower cost of launch and miniaturized, more capable electronics has not reduced the price of failure far enough that there is money to do it twice if you don’t do it right.
The real revolution: reducing the cost of being wrong
So, I genuinely believe that the next leap forward in space innovation won’t come from moving faster. It will come from getting better at being wrong. That means developing tools and methodologies that let teams:
- Delay irreversible decisions until later in the process.
- Simulate and test subsystems earlier and in isolation.
- Use data and feedback to refine assumptions, not just verify them.
- Build flexibility into the architecture, so that late-breaking changes don’t require starting over.
In short, the real opportunity is to make the design process itself more adaptive — not in theory, but in practice.
There are companies out there doing this now. I know some of them. They are building design processes that tolerate learning. They are pioneering things like hardware-in-the-loop simulation to discover issues earlier. Or innovations that allow teams to swap subsystems or reallocate resources late in the build cycle and that allow size, weight and power margins to be tracked dynamically, rather than assuming they are fixed early on. Or they are developing ways to make launch more responsive rather than just cheaper so that the launch itself does not become a cost driver of the design.
This is not just smart engineering. It is good business. Because it makes learning cheaper. And in space, cheaper learning is the only way to reduce the cost of mistakes, and the only real way to make use of the reduced cost of access to space itself.
The moat that investors overlook
Classically, investors look for moats in things like patents, market share, or customer contracts. But in space, one of the most durable moats is actually a process: the ability to iterate without starting over. That process, once built, is incredibly hard to replicate. It is shaped by flight heritage, failure analysis, tooling, and institutional knowledge.
It becomes the reason customers trust you. The reason you can quote faster. The reason your cost per iteration is lower than your competitor’s. And the reason you survive when others don’t.
The companies that develop this ability for themselves and who enable it in others are the companies that will emerge from this creative disruption cycle as the leaders of the industry.
Because if the first phase of New Space was about launching cheaper and faster, the next phase will be about designing smarter. The companies that learn to adapt without compromising mission outcomes will be the ones that define this era in the long run.
Because in the end, space will always be hard. But we can choose whether it has to be inflexible and expensive. And in that choice lies the future of innovation — and the real returns for those willing to invest in it.
