The Capability-Business Gap

The commercial space sector has built more capability in the last decade than at any comparable period in its history. Reusable launch is operational. Smallsat manufacturing has industrialized. Constellation deployment is routine. Earth observation produces data volumes that would have been unthinkable in 2015. By any technical measure, this is a triumphant decade.

By any commercial measure, it is a more complicated decade. Most of the space companies that listed publicly between 2020 and 2022 fell 70 to 90 percent or more from their peak in the years that followed. The cohort has since split in two. A handful with genuine commercial traction have recovered sharply, some past their old highs, while many others restructured, were acquired below their raise price, went private at a fraction of peak, or stopped operating altogether. That divergence, between the companies that pulled through and the ones that did not, is the subject of this issue. Even SpaceX, the most successful company in the sector, took the better part of two decades, by most external estimates, to demonstrate consistent positive unit economics at the launch-vehicle level. That is the benchmark case.

The gap between these two pictures, extraordinary technical capability set against structurally challenged commercial performance, is the most important analytical question in commercial space today. It is also the question the industry's surface narrative is least equipped to answer, because the surface narrative is built by people whose incentives align with the technical achievement framing.

This issue is about the gap. What it is. Why it persists. And what the small number of companies that closed it did differently.

The industry's standard framing treats technical capability and commercial viability as continuous, as if the same execution that produces a working spacecraft will, with sufficient time and customer development, produce a working business.

This framing is structurally wrong. Technical capability and commercial viability are not points on the same line. They are separate axes that can vary independently.

A company can have extraordinary technical capability and almost no commercial viability: a working spacecraft, a launched constellation, demonstrated data products, and customers who will not pay enough or buy reliably enough to fund the next iteration. A company can have moderate technical capability and significant commercial viability: a less impressive product that solves a specific problem for a willing buyer at a price that supports the business.

The conflation produces specific errors. It causes investors to underwrite technical milestones as if they were commercial inflection points. It causes founders to defer commercial development on the assumption that customer traction follows engineering proof. It causes the press to celebrate launches as if successful insertions were commercial events. None of these errors are stupid in isolation. They are rational responses to a narrative that conflates two genuinely different things.

The cost of the conflation appears at predictable moments in a company's lifecycle: the post-launch period when revenue does not arrive on the projected curve, the Series B raise when investors begin asking commercial questions the company is not prepared to answer, the moment when the anchor customer's priorities shift and there is no diversified pipeline behind them. Each of these is described in post-mortems as a commercial failure. They are usually described as surprising. They are almost never actually surprising in retrospect.

Three structural conditions keep the gap open across the industry.

The first is that capital and capability are easier to organize than customers. Funding a space company requires convincing a small number of investors. Building technical capability requires hiring engineers, who are available, and managing programs, which is hard but bounded. Acquiring customers in space requires translating capability into a problem a specific buyer has, in a language that buyer uses, on a timeline that matches their procurement cycle, which is eighteen to thirty-six months for most enterprise customers and longer for government. Each of these activities draws on different organizational competencies. The first two are inside the company. The third is outside it, slower, and significantly less responsive to capital.

The second is that the industry's incentive structure rewards capability signaling over commercial discipline. A successful launch generates press. A failed sales cycle does not. Investors at early stages cannot easily evaluate commercial readiness in a sector with eighteen-month sales cycles, so they evaluate technical readiness as a proxy. Founders raise on the proxy. The company optimizes for the proxy. The proxy is real (technical capability matters), but optimizing for it does not produce the underlying commercial reality the proxy was supposed to represent.

The third is that the supply chain rewards platform language. Founders pitch platforms because platforms get larger multiples than products. Investors fund platforms because platforms imply optionality. Both sides know that most platforms in commercial space (broadband from LEO, EO data across verticals, ground networks serving multiple operators) require sustained capital deployment over years before they generate platform-level returns. Both sides also know that the company in front of them needs point-solution economics to survive the next eighteen months. The mismatch is rarely examined explicitly, because examining it would surface that the pitch and the business are operating on different assumptions about what is being built.

None of these conditions are easily fixed at the industry level. Each is rational from the position of the individual actor producing it. The result is a sector where the gap is reproduced structurally, faster than any individual company can close it.

A small number of commercial space companies have closed the capability-business gap. Their commonalities are more instructive than their differences.

They began customer development before they had a product. Not after the technology demonstrated, not in parallel with the engineering program, but before. Eighteen months before launch, they had identified specific customers with specific problems. They had pilot conversations and letters of intent. The satellite was not the moment commercial work began. It was the moment a commercial commitment already in place was executed against.

They built ground and operations as core products, not as deployment infrastructure. Planet built and ran its own distributed ground-station network as a first-class part of the system rather than an afterthought, and Spire did the same, owning its global station network and treating mission operations as core infrastructure from early on. The companies that struggled discovered the cost and complexity of ground at integration, when both sides had become structurally expensive to change. The companies that succeeded priced ground into unit economics from the architecture phase. This single decision separates more operational survivors from technically successful failures than any other.

They priced for the customer they served, not for the cost they incurred. Defense-heritage companies that tried to sell commercial Earth observation products using cost-plus pricing logic consistently lost to commercial-native competitors who priced what the data was worth to a specific buyer. The discipline is harder than it sounds. It requires the organization to internalize that commercial buyers do not care what the satellite cost to build; they care what the answer to their question is worth. This is not a sales adjustment. It is an organizational architecture decision that affects hiring, pricing strategy, and customer relationship management.

They treated government revenue as a floor, not a business. Government contracts in commercial space are genuinely valuable. They provide predictable income that buys time while commercial markets develop. They become dangerous when founders treat them as commercial validation rather than runway. The companies that built durable businesses made one critical transition: from government-primary to government-anchored. Government revenue stayed in the model. Commercial revenue had to be built in parallel, on different organizational competencies, at different cadences, with different customer requirements. Most founders defer this work until the government relationship shows signs of stress. By then the runway is short and the commercial pipeline has no history.

What unifies these patterns is that they all treat the business as a separate engineering problem from the technology. Not a subordinate problem. Not a downstream problem to be solved after the technology demonstrates. A parallel problem that requires its own design discipline, its own organizational capability, its own timeline of milestones and reviews.

The companies that succeeded did not have better technology than the ones that failed. They had a second engineering discipline that the industry's narrative does not fully recognize as engineering.

Two specific developments in 2026 will test this argument.

The first is whether the next wave of space investment, which is visibly forming, recognizes the capability-business gap or repeats the SPAC era's confusion. The early signals are mixed. Investors who completed one full cycle are asking different questions in their first meetings, questions about contract backlog and mission success rate rather than TAM and announced launches. The new-entrant generalist VCs returning to the sector are not. Which cohort drives the next wave of valuations will determine whether the gap narrows at the industry level or persists for another cycle.

The second is whether the Indian commercial space sector, now well into IN-SPACe's operational framework, internalizes the lesson at the cohort level before its first listed companies face the test. The Indian sector has a structural advantage here: it has the example of the SPAC era to study. It also has a structural risk: the same investor enthusiasm that drove the US wave is forming for Indian space, and the same narrative-over-execution dynamics are visible in some of the recent fundraising language. The next twelve months will reveal which way the sector's cohort settles.

I will return to both questions in future issues as the evidence accumulates.

The capability-business gap is the analytical question I find myself returning to most often, because it is the one the industry's surface narrative is least equipped to surface and the one where careful analysis has the most leverage. A company that internalizes the distinction between capability and commercial viability is structurally more likely to survive. A company that does not is structurally more likely to become a post-mortem.

This is the work I want this newsletter to do. Take one question the industry's surface narrative answers wrongly or incompletely, and work through it carefully enough to change how a serious reader thinks about it. Not every issue will succeed. Some will land. The ones that do compound. The ones that do not teach me what to do differently next time.

Thank you for being among the first readers. The next issue, on September 6, is about what government revenue actually buys for a space startup, and what most founders mistake it for.

Chandrim

APOGEE · Issue 1 · August 16, 2026