Welcome to the 122nd edition of Deep Tech Catalyst, the educational channel from The Scenarionist where science meets venture!
This week, I sat down with Alan Yu, Senior Associate at Space Capital, to unpack how satellites are becoming one of the invisible operating layers of the broader space economy, and what founders need to know about building, validating, and funding successful companies in the sector.
Key takeaways from the episode:
🛰️ Satellites Are Becoming Invisible Infrastructure
Most users do not experience satellites as space technology. They experience them as connectivity, navigation, timing, mapping, and intelligence embedded into daily systems.
🧠 The Next Generation of Satellites Will Do More in Orbit
Satellites are becoming more capable, more coordinated, and more useful as operational assets through onboard compute, propulsion, mobility, and system-level coordination.
🔄 The Hard Part Is Turning Data Into Decisions
Raw satellite data is not enough. The company-building challenge is to transform signals, imagery, and geospatial inputs into outputs customers can actually use.
🧱 Build Only What the Company Must Own
Founders should not confuse gaps in the stack with obligations to build. The real discipline is knowing what must be owned, what can be partnered for, and what has been validated by the market.
💸 Capital Efficiency Is About Proving the Right Thing
In space, some companies need significant capital. The question is whether that capital is tied to risk reduction, sharper milestones, stage-appropriate hiring, and evidence that the company is moving in the right direction.
BEYOND THE CONVERSATION — STRATEGIC INSIGHTS FROM THE EPISODE
The Space Economy Is Becoming an Operating Layer of Modern Life
Framing the satellite economy in 2026 requires moving beyond the image that still dominates much of the public imagination.
For many people, space still means rockets, astronauts, missions, or assets orbiting far above Earth. That view is not wrong, but it is increasingly incomplete.
A more useful way to understand the space economy today is to see it as an operating layer of modern life.
Satellites are no longer just instruments of exploration. They have become part of the infrastructure that supports transportation, communications, financial systems, mapping, climate intelligence, defense, logistics, and many other activities that now depend on space-enabled capabilities.
Most people do not experience this directly as “space.” They experience it as a phone that knows where it is, a plane that remains connected in the air, a ride-hailing app that works, a bank transaction that relies on precise timing, a map that updates, or a system that can observe conditions on Earth from above.
That is what makes the “satellite economy” strategically important.
Its value is not limited to what happens in orbit. Its value comes from the way orbital infrastructure becomes embedded into terrestrial markets.
The more these systems work, the more invisible they become. And the more invisible they become, the more dependent the economy becomes on them.
Satellites as invisible infrastructure
The satellite economy can be understood through three major capabilities: satellite communications, GPS and precision navigation and timing, and geospatial intelligence.
Each of these categories has a different history, a different technical architecture, and a different set of commercial applications. But together they reveal the same underlying pattern: space-based infrastructure has become a utility.
That utility-like nature matters.
Once society has access to these systems, it does not want less of them. It wants more coverage, more resilience, more precision, more bandwidth, and more intelligence.
No one expects GPS to disappear from daily life. No one wants connectivity to become less available. No institution wants timing systems to become less reliable.
The more digital and autonomous the economy becomes, the more important these satellite-enabled systems become.
GPS is the clearest example.
For consumers, GPS often appears as a convenience. It helps a phone identify a location, supports navigation, enables ride-sharing, and makes mapping applications feel effortless. But that consumer-facing experience represents only a small part of the underlying importance.
Precision navigation and timing reach much deeper into the economy. They support transportation networks, autonomous systems, logistics, and financial infrastructure.
Satellite communications have followed a similar path.
For a long time, satellite connectivity was associated with specialized use cases: aircraft, ships, remote locations, or government and defense environments. It was valuable, but not necessarily something most people thought of as a broad consumer or enterprise infrastructure layer.
That perception has changed.
With the rise of large satellite constellations and more ambitious connectivity models, satellite communications are becoming part of a much broader conversation about global access, redundancy, mobility, and resilience.
The expectation is no longer that connectivity should exist only where terrestrial infrastructure is convenient. Increasingly, the expectation is that connectivity should extend almost everywhere.
Geospatial intelligence adds another dimension.
Earth observation has long been one of the most important uses of satellites, but the strategic value is not in imagery alone. The value comes from the ability to understand what is happening on Earth and translate that understanding into decisions.
That distinction is essential.
A satellite image by itself is not always useful to the final customer. A raw signal does not automatically create economic value.
Between the capture of data and the moment when a user can act on it, there is a long chain of processing, interpretation, contextualization, and distribution.
That is why geospatial intelligence sits at the intersection of infrastructure and application. It begins with orbital assets, but its value is realized only when the information becomes usable for someone on Earth.
This is one of the central tensions in building space tech startups. The infrastructure is technically complex, but the customer often wants simplicity.
The system may involve satellites, sensors, signal processing, coordinate systems, imagery pipelines, AI models, and distribution platforms. But the buyer cares about the decision the system enables.
That is where much of the entrepreneurial opportunity now begins to appear.
From space assets to a broader space economy
The deeper shift is that the space economy can no longer be defined only by the assets that physically operate in space.
Those assets remain foundational. Rockets, satellites, propulsion systems, ground stations, sensors, and orbital logistics are still essential.
Without them, there is no infrastructure layer to build on. But the market is larger than the hardware in orbit.
The space economy now includes the companies that build and operate satellites, the firms that move and manage assets in space, the platforms that process data, the software systems that distribute intelligence, and the businesses that turn space-derived capabilities into products for Earth-based customers.
That broader view is important because it changes who can build in the sector.
Historically, space was accessible to a narrow group of actors.
It required enormous capital, specialized engineering talent, government relationships, and tolerance for long development timelines.
Those realities have not disappeared, especially for companies building core infrastructure. But the ecosystem has expanded, with direct and indirect implications for founders.
First of all, a founder does not always need to build the entire satellite system.
A company may focus on a specific sensor, a specific mission, a specific data layer, a specific distribution problem, or a specific customer workflow.
As manufacturing improves, launch costs fall, satellite buses become more available, and the surrounding ecosystem matures, the opportunity space becomes more modular.
That does not make space easy. It does make it more strategically open.
Venture design should now begin with a sharp question:
“Which part of the stack is essential for us to own, and which part can be accessed through the ecosystem?”
That question is central to company building in the current satellite market. It forces founders to distinguish between technical ambition and strategic necessity.
There is a difference between building something because it is impressive and building it because the company cannot create durable value without owning it.
The more capital-intensive the market, the more important it becomes to understand where true differentiation lives.
The satellite economy is therefore not just expanding in size. It is changing in structure. One useful frameworks that emerged from the conversation is the three-layer view of the satellite economy:
At the base is foundational infrastructure: launch, satellite manufacturing, propulsion, orbital logistics, ground systems, communications, and the physical systems that make space operations possible.
In the middle, there is distribution: the processing, filtering, interpretation, packaging, and delivery of space-derived data and capabilities into forms that customers can actually use.
At the application layer, there are businesses that build products on top of those capabilities for agriculture, insurance, defense, mobility, logistics, autonomous systems, financial services, urban planning, and many other markets.
This layered view matters because each layer behaves differently. At a high level:
Infrastructure companies often require deeper technical teams, more capital, longer development cycles, and stronger execution against physical systems.
Distribution and application companies may be more software-driven, but they still depend on the maturity, availability, reliability, and cost structure of the underlying space infrastructure.
The pitfall is assuming that infrastructure availability automatically means market readiness.
A satellite may be able to capture the data. A provider may be able to make that data available. A founder may be able to build a product concept around it.
But the company still has to prove that the cost structure works, that the customer understands the value, that the insight can be delivered frictionlessly, and that the willingness to pay is real.
That is why satellites should not be viewed only as a technology category. They should be viewed as the foundation of a developing business architecture.
The current moment is compelling because several things are happening at once.
Satellites are becoming more capable.
Launch and manufacturing are becoming more efficient.
Connectivity is expanding.
Sensors are improving.
Compute is moving closer to orbit.
The number of companies contributing to the ecosystem is growing.
And customers on Earth are becoming more aware that space-derived intelligence can support real operational decisions.
But the market has not reached full maturity.
The infrastructure is advancing faster than many applications can absorb. The capabilities are becoming stronger, but the translation into customer value remains uneven. In other words, the potential is broad, but it is not automatic.
That is the right starting point for understanding the satellite economy in 2026.
Space is no longer a separate frontier sitting outside the rest of the economy. It is becoming a layer beneath it. The companies that matter most will not simply be the ones that put assets in orbit. They will be the ones that understand how orbital infrastructure becomes usable, valuable, and investable on Earth.
Satellites Are Becoming Smarter
As discussed, satellites are becoming more capable, more coordinated, and more useful as operational assets.
They are no longer just passive machines placed in orbit to capture images, transmit signals, or provide connectivity.
Increasingly, they are becoming intelligent infrastructure: machines that can compute, filter, coordinate, move, and eventually operate as part of larger systems.
That distinction matters because it changes the strategic role of satellites in the space economy.
A satellite used to be thought of as a highly specialized node. It had a mission, a payload, a communications function, and an orbital position. It was expensive to build, expensive to launch, difficult to move, and often limited in flexibility once deployed.
That model is giving way to something more dynamic.
The next generation of satellites is not simply cheaper. It is becoming more operationally relevant.
The cost of building and launching systems with comparable capability has fallen, but at the same time the capability of those systems has increased dramatically.
Better manufacturing, stronger sensors, more powerful onboard compute, and more flexible propulsion are beginning to reshape what satellites can do and how companies can build around them.
From static assets to autonomous machines
The most important change in satellite infrastructure is the move from static orbital assets toward more autonomous systems.
This does not mean every satellite suddenly becomes fully independent or intelligent in the way people sometimes imagine when talking about AI or robotics.
The point is more practical.
Satellites are gaining the ability to do more of the work closer to where the data is generated.
Onboard compute is a simple but powerful example. If a satellite captures imagery, not every image has the same value. Some data may be redundant, cloudy, low quality, or irrelevant to the customer’s need.
In older architectures, much of that data would need to be transmitted down and processed later, creating cost, latency, and inefficiency.
As satellites gain more compute capability, they can begin to filter and prioritize information in orbit. They can help decide what is worth sending down and what is not. That changes the economics of data movement and the usefulness of the system.
It also changes the meaning of satellite infrastructure itself.
The satellite becomes less like a remote instrument and more like part of a a computing-enabled asset inside a broader space infrastructure layer.
It is still a physical asset in orbit, but it increasingly participates in the intelligence layer of the system. It does not merely collect information. It helps make information more usable.
That matters as the volume of data increases.
More satellites mean more coverage, more sensors, and more raw information. But more information does not automatically create more value. Without filtering, processing, and distribution, the system can produce complexity faster than it produces insight.
The smarter the satellite becomes, the more it can help reduce that friction.
This will matter beyond Earth observation. It will matter for communications, defense, orbital coordination, robotics, lunar operations, and any environment where space assets need to work together in real time.
The broader direction is clear: satellites will need to coordinate with other satellites, ground stations, logistics providers, defense networks, data platforms, and end-user applications.
The more congested and strategically important the orbital environment becomes, the more valuable that coordination becomes.
This is why the satellite of the future should not be understood only as hardware. It should be understood as a node in an increasingly complex operating system.
The New Satellite Stack Changes What Founders Can Build
Historically, building a space company often meant confronting the full burden of space infrastructure.
A founder who wanted a specific type of data or sensing capability had limited options. If the required satellite data did not exist, or if existing providers could not meet the need, the company faced an enormous problem.
Building the full system meant designing spacecraft, sourcing components, managing payloads, dealing with launch, handling operations, and raising significant capital before proving much about customer demand.
That is no longer always the case. Here are a few reasons, drawn from our conversation:
The ecosystem has become more modular.
Manufacturing capacity is improving.
Sensors and power systems are becoming more accessible in certain categories.
Launch remains difficult, but lower launch and manufacturing costs have changed what early-stage teams can realistically attempt.
Specialized companies can now take on parts of the stack that every founder previously might have been forced to confront directly.
That changes the early-stage roadmap.
A company may have a differentiated sensing concept, a specific customer insight, or a mission-driven application without needing to build every piece of the space system from scratch.
It can focus on the part of the architecture that matters most to its strategy and rely on partners or suppliers for the rest.
This is a profound venture-building shift.
It also means founders need to be much more precise about what they choose to own.
Owning more infrastructure can create control, defensibility, and performance advantages. But it also increases cost, complexity, hiring burden, development time, and capital requirements.
In a market where timing and capital efficiency matter, unnecessary ownership can become a liability. The best founders understand this tradeoff early.
Some companies should absolutely be infrastructure companies.
They need deep engineering teams, flight heritage, physical systems expertise, and the ability to build hard technology that works in space. These companies may require more capital because the problem itself demands it.
But many other companies in the space economy do not need to own the whole stack. They may be building intelligence products, data platforms, customer workflows, analytics layers, or application-specific systems that depend on space infrastructure without being pure infrastructure companies themselves.
For them, the advantage may come from customer understanding, distribution, workflow integration, or the ability to turn complex signals into actionable insight.
That difference matters to investors.
Capital intensity is not automatically bad in Deep Tech. Some problems require capital because the technical challenge is real and the reward is large enough to justify the effort. But capital intensity must be matched by the right kind of value creation.
In the end, the changing satellite ecosystem gives founders more room to build. But it also forces them to decide early what kind of company they are actually building.
Are they building a core infrastructure company?
Are they building a distribution layer?
Are they building an application business that depends on space-derived intelligence?
Are they creating a new capability that requires owning hardware, or are they using existing infrastructure to solve a customer problem more effectively?
Those questions shape the team, the fundraising strategy, the roadmap, the customer discovery process, and the investor narrative.
Founders Should Build Only What Truly Matters
One of the most important disciplines in Deep Tech company building is knowing what not to build. That sounds simple, but in technically ambitious markets it is one of the hardest judgments to make.
The best founders are not the ones who try to own everything. They are the ones who understand which part of the stack defines their advantage.
Core space infrastructure companies
For companies building core space infrastructure, the technology simply has to work.
There is no way around that. A satellite company, propulsion company, orbital logistics company, sensing company, or hardware-driven space startup cannot compensate for weak technical execution with storytelling alone.
The team has to understand the engineering reality.
The company has to attract people who know how to build hard systems, ideally with experience that reflects the practical difficulty of getting technology to operate in space.
Therefore, a founder does not necessarily need to be the most technical person in the company, but the founder does need to understand the technology deeply enough to lead the organization, recruit the right people, ask the right questions, and avoid being disconnected from the technical truth of the business.
That is different from simply having a technical background.
It requires enough fluency to make strategic decisions around tradeoffs, timelines, architecture, risk, and customer fit.
It requires the ability to work closely with engineers without reducing the company to an engineering project.
This is where many Deep Tech founders encounter a subtle trap. Because the technology is difficult, it can become the center of gravity of the entire company.
Every milestone becomes a technical milestone. Every discussion returns to the next engineering challenge. The company keeps moving toward better technology, but not necessarily toward a better business.
That is not enough.
In venture-backed Deep Tech, the goal is to build a system that matches what the customer needs at the moment when the customer is ready to adopt it.
If this direction is not managed carefully, the path can consume years, force unnecessary fundraising, dilute the company, and make the team believe it is making progress while the market remains unproven.
Distribution and applications
For founders building distribution systems or applications on top of satellite infrastructure, the risk is different.
These companies may not need to build satellites, launch systems, or orbital vehicles themselves.
They may be using existing data, connectivity, geospatial intelligence, or distribution systems to create products for specific markets.
In principle, that should make them less capital intensive and more flexible. But the risk is assuming that the underlying inputs are ready simply because they exist.
A founder may see an opportunity in agriculture, climate, insurance, logistics, defense, autonomous systems, or infrastructure monitoring.
The use case may be compelling. The customer problem may be real. The value of space-derived intelligence may be easy to imagine.
But the company still has to prove that the inputs can support the business.
Is the data available at the right frequency?
Is it accurate enough?
Is it affordable enough?
Can it be processed reliably?
Can it be delivered into the customer’s workflow without too much friction?
These questions determine whether an application business can actually exist.
In geospatial intelligence, this issue appears repeatedly. Satellite imagery can be powerful, but the cost of acquiring, processing, and translating that imagery into something useful can still be significant.
In some cases, price is a constraint. In others, the deeper constraint is that customers do not yet understand how to use the intelligence in a way that changes behavior.
That distinction matters.
For application founders, the work is therefore not just technical development. It is market development.
They must prove that the infrastructure they depend on is mature enough to support their use case, and that the customer is ready enough to turn that capability into revenue.
Cost structure shapes the business model
The cost of data, infrastructure, talent, manufacturing, deployment, customer support, integration, and technical iteration determines how much flexibility a company has.
It affects pricing. It affects margins. It affects the sales motion. It affects whether the company can experiment with different customer segments or whether every customer has to be large enough to justify the effort.
A business model is not a spreadsheet exercise performed after the product has been defined. It is shaped by the architecture of what the company chooses to own, what it chooses to avoid, and how directly its costs map to customer value.
In space-enabled markets, this is particularly important because the cost of inputs can be substantial.
Satellite imagery, data processing, infrastructure access, customer integration, and domain-specific interpretation can all shape the economics of the company.
Simple, but not obvious:
If those costs are too high, the company loses flexibility. It may need to charge more, sell only to larger customers, or narrow its market to the few buyers who can justify the price.
If the costs are lower, the company has more strategic room. It can experiment with pricing. It can serve more customer types. It can adjust the sales motion. It can protect margins while learning. It can avoid forcing the market into a pricing structure it is not ready to accept.
That is why cost drives the business model more than founders sometimes want to admit. The product may define what the company does, but the cost structure defines what the company can afford to become.
One of the most effective ways to approach this problem is to work backward from the customer.
Not backward from the technology. Not backward from the product vision. Not backward from the total addressable market slide. Backward from what the customer values enough to pay for.
That requires a founder to ask uncomfortable questions early.
How much would this customer actually pay?
What would make them pay more?
What would make them stop paying?
What part of the product creates the strongest pull?
What existing tool or process would this replace?
Capital Efficiency Begins With the Right Founder Mindset
Some space companies genuinely need significant capital.
A company building orbital logistics, propulsion, energy transfer, satellite infrastructure, or another deeply technical system may require specialized talent, expensive testing, and enough runway to reach a meaningful technical milestone.
In these cases, every dollar should be tied especially clearly to risk reduction and company value.
Capital efficiency is not simply about spending less. It is about knowing what the next dollar is supposed to prove. It is about separating essential progress from motion, and understanding which risks deserve capital now versus which can be tested through sharper prioritization, better sequencing, or a more focused roadmap.
Founders may be tempted to raise more than they need.
That can feel rational. More capital seems to mean more safety, more hiring power, more technical ambition, and more room to maneuver. But over-raising can create its own risk.
When a company raises too much too early, it can start behaving as if the next stage has already been earned. It hires ahead of clarity. It expands the roadmap before the company knows which milestones really matter. It becomes easier to hide weak signal behind a larger budget.
Early-stage companies do not need to prove everything.
They need to prove the right thing.
The founder’s task is to understand what the next financing round, customer commitment, technical milestone, or strategic inflection point will depend on.
The amount of capital raised should reflect that path.
It should be enough to reach the next proof point with credibility, but not so much that the company loses urgency, accumulates unnecessary complexity, or creates expectations it cannot support.
Capital is not the strategy. Capital is the fuel for a strategy that has to be clear before the money arrives.
Early teams should stay flexible
Capital efficiency also shows up in how a founder builds the team.
Early-stage companies often feel pressure to look complete. They want defined executive functions because that structure resembles a mature company. But early startups are not mature companies. They are learning systems.
At the earliest stages, rigid roles can become premature.
The company may not yet know what kind of operations it truly needs, what kind of sales motion will work, how technical development will evolve, or which customer segment will define the first real market.
That is part of what makes venture building in Deep Tech difficult.
The problems are often undefined. The markets may still be forming. The technical architecture may change as the company discovers what is feasible, valuable, and affordable.
Because of that, every company needs to identify its own catalyst.
For some companies, the key metric may be revenue.
For others, revenue may matter less than a specific customer commitment, a technical demonstration, a signed partnership, a successful test, a reduction in cost, or evidence that a particular buyer is engaging seriously.
The important thing is to know what evidence will change the probability of success.
That mindset should show up in investor updates as well.
A company can share product developments, customer conversations, technical progress, hiring updates, partnerships, experiments, and operational work.
All of that can be useful. But the best updates reveal focus. They show what the company is learning, where the team is concentrating effort, and how the business is extracting more value from the resources it already has.
That input-output discipline is a powerful signal.
It shows that experimentation is not random. It is tied to a strategic hypothesis about what will make the company work.
