> ## Content Index
> Fetch the complete content index at: https://nexi.fund/llms.txt
> Use this file to discover other available public pages before exploring further.

# The orbital infrastructure stack: where the margin actually sits
- URL: https://nexi.fund/orbital-infrastructure-stack-2026/
- Published: 2026-11-20T09:30:00.000Z
- Updated: 2026-11-20T09:30:00.000Z
- Description: Launch is no longer the product. Six layers make up the orbital infrastructure stack, and value is migrating to the operators that reduce coordination costs between them. What D-Orbit, OwlO, ESA's RISE mission and the project-finance shift say about where to underwrite orbit.
- Author: Nexi.fund Labs
- Tags: Space & Expansion, #mode-3, #hook-thesis, #track-D, #layout-asymmetric-long

Launch used to be the product. In 2026 it is closer to a toll road: necessary, increasingly price-competitive, and rarely the place where the durable margin sits. The companies that will matter over the next decade are the ones that make orbit usable after the rocket has done its job — the operators that move a satellite to its final slot, dock with it, refuel it, relay its data, dispose of it, and handle the licensing that nobody sees on a mission poster.

That is the orbital infrastructure stack. Six layers, one dependency chain. Launch gets a payload off the planet. Platforms keep it alive. Relay moves its data. Logistics moves the spacecraft itself. Manufacturing and servicing extend or rebuild it. Disposal and regulation decide whether the orbit stays usable at all. Each layer has its own cost curve, its own buyer, and its own bottleneck.

The thesis worth testing is simple. As the number of active satellites rises and launch prices converge, the scarce input stops being access to orbit and becomes coordination across these layers. Whoever reduces the cost of coordinating them — through standards, interfaces, ports, scheduling, or plain operational reliability — captures a position that is hard to displace.

Our read, before the evidence: underwrite the coordination layer first. A company with the best single spacecraft in a crowded layer is a vendor. A company that other spacecraft have to interface with is infrastructure. Those are different businesses, with different multiples and very different failure modes.

## The stack, in one page

The first layer is launch. It is capital-heavy, visible, and politically loaded. It is also the layer most exposed to price competition, because cadence and reusability have turned a bespoke service into something closer to scheduled freight.

The second layer is the platform: the bus, power, propulsion, avionics, and thermal system that keep a payload functioning. This is where most satellite manufacturers sit, and where margins compress as production volumes rise.

The third layer is relay. Optical inter-satellite links, ground stations, and the ground segment turn a spacecraft into a node. The interesting fights here are over spectrum, latency, and who owns the landing rights for data coming down.

The fourth layer is logistics: orbital transfer vehicles, tugs, servicing craft, refuelling ports, and cargo movement between orbits. It is small today. It is also the layer that turns a one-off satellite sale into a recurring service relationship.

The fifth layer is manufacturing and refurbishment in orbit — hosted payloads, in-space production, and the repair or upgrade of existing assets. The sixth is disposal and regulation: debris removal, end-of-life compliance, licensing, and insurance. Both layers are being built largely on government demand, which makes them defensible near-term and fragile long-term if public budgets change.

The dependency runs downward. Nothing above the platform scales without reliable, frequent launch. Nothing in logistics scales without docking standards that existing satellites can accept. Nothing in disposal scales without a regulator willing to buy outcomes instead of demonstrations.

## Launch is the toll road, not the destination

The numbers explain why launch stopped being the story. In 2025, 329 space launches carried roughly 4,900 objects into orbit, according to the OECD's *Space Economy at a Glance 2026*. By the end of that year, more than 14,000 operational satellites were circling Earth. By mid-2026, the count had reached nearly 15,000\. Private operators accounted for 88% of satellites launched in 2025, up from 23% in 2010, and telecommunications took 82% of the total.

Concentration is the other half of the picture. One operator — SpaceX, through Starlink — accounted for around 60% of all active satellites in 2025\. In launch, SpaceX, CASC, Rocket Lab, and Roscosmos together handled more than four-fifths of orbital launches in 2025, with SpaceX alone representing about half the global total. The OECD estimates the global launch market at USD 12–16 billion in 2025, while noting that most providers do not publish pricing or revenue, and that large institutional segments remain captive markets.

That combination — concentrated supply, opaque pricing, and rising volume — is exactly the setup in which a toll road becomes commoditized at the top and lucrative in the middle. Europe is testing the middle. The European Launcher Challenge has signed first contracts with Rocket Factory Augsburg at €186.9 million, PLD Space at €158.9 million, and Isar Aerospace at €197.8 million. Each provider must reach an orbital launch before 2028\. The point is not to beat SpaceX on price. It is to create a second supply option for institutional and commercial buyers who cannot afford single-source exposure.

For an investor, launch is a capacity bet with a political floor and a brutal cadence requirement. Fixed costs — factories, pads, engineering teams — are spread across flights, so utilization decides unit economics. A provider with a full manifest can look healthy at a price that would bankrupt a provider with empty slots. That is why launch equity is so volatile: the business is a utilization curve wearing a rocket costume.

It is also why launch is a poor proxy for the rest of the stack. A cheaper launch makes every layer above it more viable, but it does not automatically transfer value upward. The value transfer happens when a service becomes a recurring line item for the satellite operator. Cheaper rockets alone do not move it.

## The layer that is actually compounding

In-orbit services are still small in absolute terms. Novaspace projects roughly $3 billion in cumulative service revenues over the next decade, with satellite refuelling at $1.2 billion and life-extension services at $860 million. Those are not headline-grabbing numbers next to a launch market measured in the low tens of billions. They matter because of the shape of the revenue: recurring, contract-based, and tied to an asset that already exists.

D-Orbit's operating record is the clearest evidence that this is no longer a demonstration business. The Italian company had flown 26 commercial ION missions by October 2026, with 89 hosted payloads and 168 satellites deployed. On Transporter-18 it flew three ION vehicles on a single flight, carrying close to 1.5 tonnes. The company says the total mass of IONs it has launched since 2020 now exceeds that of all other orbital transfer vehicles worldwide combined. Treat the comparison as a vendor claim, but the mission count is checkable and the cadence is real.

The next step is life extension, which is where the recurring revenue becomes obvious. ESA is co-funding RISE, a D-Orbit mission to demonstrate rendezvous and docking with a geostationary client satellite. In September 2026, D-Orbit held a RISE customer workshop during World Space Business Week in Paris; ESA says it gathered senior executives from 20 world-leading satellite operators, primarily in GEO. ESA Director General Josef Aschbacher framed the moment plainly:

> This is no longer a distant ambition. A clear market for in-orbit servicing is emerging, driven by the need for more sustainable and efficient space activities.— Josef Aschbacher, Director General, ESA

GEO is the right beachhead. A geostationary satellite is expensive, its orbital slot is valuable, and its operational life is often limited by propellant, while the payload still works. If a servicing craft can dock and take over station-keeping, the operator gets years of additional revenue from an asset that was already fully depreciated. That is a straightforward return calculation, which is why 20 operators showed up to a workshop rather than waiting for a brochure.

ClearSpace is building the parallel European capability. ESA signed a contract with the Luxembourg company in September 2026 to advance the Phoenix servicing spacecraft, aimed at extending the life of GEO satellites and maturing the design toward a recurring commercial service. In October, the UK Space Agency selected ClearSpace as prime contractor for a £61 million active debris removal mission. Disposal is becoming a procurement category with a budget line.

The risk in this layer is operational, and it is not theoretical. Starfish Space's first full-scale Otter launched in October 2026 on NASA's $15 million SSPICY inspection mission. By early November, the company had lost contact with the spacecraft. One failure does not invalidate a market. It does show how thin the margin for error is when the product is a rendezvous in vacuum, and why operators with flight heritage will price at a premium for longer than the pitch decks suggest.

## Infrastructure capital meets orbital plumbing

The most interesting financing shift is happening one layer below the servicers. OwlO Cloud, a Luxembourg startup founded in 2026, emerged from stealth in November with a plan to operate at least 10 orbital cloud nodes by 2030\. Each node is expected to generate about 10 kW of power, an order of magnitude below the 100+ kW pitched by orbital data center developers. The company's stated use case is narrower: help satellites store and process their data in orbit, with no attempt to replace terrestrial data centers.

The capital structure is the signal. OwlO's CEO said the company will use venture capital for IP and R&D, and infrastructure financing to deploy the constellation over time. That is how terrestrial infrastructure gets built — development risk on equity, deployment risk on project finance. If orbital infrastructure can actually be financed that way, the addressable capital pool changes completely. It also changes the underwriting: project finance cares about contracted cash flows and counterparty quality; technology elegance is secondary.

Government demand is providing the anchor contracts that make this financeable. GAO reported in November 2026 that the US Space Force awarded a $2.3 billion firm-fixed-price other transaction agreement in May 2026 for a fully operational prototype of the Space Development Network Backbone. The Proliferated Warfighter Space Architecture it supports is planned at 300+ satellites with $15.6 billion in planned cost over the next five fiscal years. GAO also notes that 94% of US satellites in orbit are privately owned, a useful reminder that the commercial stack and the defense stack are now the same supply chain.

Europe is funding the same transition from the other side. ESA's Investor Forum materials put roughly €300 million of ISAM spending on the 2025–2028 window, spanning life extension, refuelling, disposal, and robotic servicing. The money is small relative to US defense budgets. Its function is to de-risk first-of-a-kind missions so that private capital can price the second and third units.

Layer economics as of November 2026; revenue-quality assessment is editorial, figures from OECD, ESA, Novaspace, and GAO reporting cited in Sources.

| Layer                     | Who pays today             | Revenue quality                  | Binding constraint                    |
| ------------------------- | -------------------------- | -------------------------------- | ------------------------------------- |
| **Launch**                | Commercial + institutional | Cyclical, utilization-driven     | Cadence and fixed-cost absorption     |
| **Platforms**             | Prime contractors          | Project-based, margin-compressed | Production volume                     |
| **Relay**                 | Operators + governments    | Subscription-like, sticky        | Spectrum and ground segment           |
| **Logistics / servicing** | Operators + agencies       | Recurring, contract-backed       | Docking standards and flight heritage |
| **Disposal / regulation** | Governments first          | Procurement outcomes             | Budget cycles and liability rules     |

The table points to a conclusion that is easy to state and hard to execute. Launch and platforms reward scale. Relay, logistics, and disposal reward standardization and reliability. The first group is a manufacturing business. The second group is an infrastructure business, and infrastructure businesses are valued on the durability of the cash flow. Hardware novelty is a secondary consideration.

## What to underwrite

Four criteria separate a coordination-layer business from a vendor with good marketing.

**First, an interface that others must adopt.** A docking port, a refuelling standard, a data relay protocol, or a scheduling layer becomes valuable when a second and third operator build to it. RAFTI and GRIP-style refuelling interfaces are the clearest current example; the company that owns the standard does not need to own every spacecraft. Watch for adoption by operators who are not also investors in the standard-setter.

**Second, flight heritage that compounds.** D-Orbit's 26 ION missions are worth more than a single spectacular demonstration because each mission reduces the operational uncertainty on the next contract. In servicing, the same logic applies to docking attempts. This is a business where the learning curve is physical, and physical learning curves are hard to buy with capital alone.

**Third, a revenue line that survives the demo phase.** Government anchor contracts are a legitimate way to cross the valley, but they can flatter a business model. The test is whether a commercial operator renews a service after the first subsidized mission. Life extension in GEO is attractive because the buyer already has a revenue-generating asset and a clear propellant constraint. Disposal is weaker on that test until regulators make it mandatory.

**Fourth, a capital structure matched to the asset life.** OwlO's split between venture equity and infrastructure debt is the template to watch. A constellation financed entirely on venture timelines will be forced to sell or refinance at the worst moment. A constellation financed entirely on project debt will struggle to fund the technology risk. The winners will blend the two deliberately, as a stated policy.

What we would avoid: launch providers with no cadence visibility, single-mission servicing concepts priced as platforms, and any business whose margin depends on the customer never comparing a second quote. Concentration cuts both ways. A supplier with one dominant customer has revenue. It does not have a market position.

## Where the thesis breaks

The strongest counterargument is that coordination layers are being built inside vertically integrated players. SpaceX can internalize relay, logistics, and disposal for its own constellation. A government can mandate a standard and leave the startup outside. If that happens, the independent coordination layer becomes a feature of someone else's platform, and the margin goes to the integrator.

The second risk is demand concentration. Roughly 60% of active satellites belong to one operator, and 94% of US satellites are privately owned. A services market that depends on a handful of constellation owners is a procurement relationship with a thin customer base. It is a procurement relationship with extra steps, and procurement relationships reprice when the buyer builds in-house.

The third risk is schedule realism. Orbital infrastructure timelines have a habit of slipping by years, and each slip burns the equity that was meant to fund deployment. ESA's requirement that European Launcher Challenge providers reach orbit before 2028 is a useful discipline. It is also a reminder that the 2028 date is a gate, not a forecast.

⚠️

**The failure mode to watch**  
A servicing or logistics operator that wins government anchor contracts, misses two docking milestones, and then needs a bridge round at a lower valuation. Flight heritage is the moat; a public anomaly is the moat draining. 

The fourth risk is regulatory and physical at the same time. Orbital congestion is already the reason disposal is becoming a budget line. But debris removal economics depend on liability rules that are still unsettled, and insurers price uncertainty before operators do. A disposal business can be technically excellent and still wait years for the demand curve to arrive.

## What we are watching next

The next twelve months should settle three questions. Does a second commercial operator renew a paid life-extension service after the first mission? Does a docking or refuelling interface get adopted by an operator that is not an investor? Does an orbital infrastructure project close project finance without a government guarantee?

Two yeses would support the coordination-layer thesis. One yes would move it to a watchlist; underwriting needs two. Zero yeses would suggest the stack is real but the margin stays with the integrators, which is a different investment conversation.

📊

**Key signals to track**  
  
D-Orbit RISE docking milestone and the first paid GEO life-extension renewal  
  
Adoption of a refuelling or docking interface by a non-investor operator  
  
Novaspace's $3B decade forecast versus actual contracted service revenue  
  
Project-finance close for an orbital infrastructure constellation without a sovereign guarantee 

Space infrastructure has always been sold as a vision. The investable version is duller and more specific: docking schedules, renewal rates, interface adoption, and the cost of capital for a constellation that has to survive its own deployment. That is where the next durable margin in orbit will be decided.

[ The Space Economy at a Glance 2026 Satellite counts, launch volumes, market concentration, and launch-market sizing used throughout the piece. OECD ](https://www.oecd.org/content/dam/oecd/en/publications/reports/2026/09/the-space-economy-at-a-glance-2026%5F92b228b7/cbf9b240-en.pdf?ref=nexi.fund) 

The cleanest public dataset for orbital occupancy; use it to check any vendor's market-share claim.

[ Unlocking the commercial potential of in-orbit servicing ESA's account of the RISE mission, the 20-operator customer workshop, and the Investor Forum financing debate. European Space Agency ](https://www.esa.int/articles/about-us/directorates/operations/space-safety/clean-space/unlocking-the-commercial-potential-of-in-orbit-servicing?ref=nexi.fund) 

Primary evidence that GEO operators themselves are evaluating paid life extension.

[ OwlO Cloud emerges from stealth to build European orbital cloud network Details on the 10-node plan, the 10 kW vs 100+ kW power comparison, and the venture-plus-infrastructure financing split. Payload ](https://payloadspace.com/owlo-cloud-emerges-from-stealth-to-build-european-orbital-cloud-network?ref=nexi.fund) 

The clearest current example of orbital infrastructure being financed like infrastructure.

[ In-orbit services market to generate $3 billion over the next decade Novaspace's market sizing for refuelling and life extension, with the government-demand caveat. SpaceNews / Novaspace ](https://spacenews.com/in-orbit-services-market-to-generate-3-billion-in-revenue-over-the-next-decade?ref=nexi.fund) 

Use it as a ceiling: the number is small enough that a few contract wins move it.

[ Into the Quiet Dark: D-Orbit's 26th commercial ION mission Mission record for the Transporter-18 flight, including deployments and hosted payload demonstrations. D-Orbit ](https://dorbit.space/into-the-quiet-dark-october-2026/?ref=nexi.fund) 

Operator primary source. The mission cadence is the part that matters.

[ Space supply chains: additional DOD actions could strengthen its enterprise approach Space Force contract values, PWSA constellation scale, and the private-ownership share of US satellites. US Government Accountability Office ](https://www.gao.gov/assets/gao-27-108683.pdf?ref=nexi.fund) 

The demand-side anchor: defense budgets are currently the most reliable revenue in the stack.