SpaceTech: what are startups building now?
In our space economy deck, you will find everything you need to understand the market
SUMMARY
SpaceTech startups are building much more than rockets today: the market is filling in the factories, communications networks, sensing systems, transport, navigation, servicing, traffic management, power and computing infrastructure needed once thousands of spacecraft are already in orbit.
The biggest structural change is happening after launch. Global orbital launches increased about 25% in 2025, while spacecraft deployments jumped 54% to 4,544, creating demand for infrastructure that barely existed when satellites were large, scarce and mostly government-owned.
Launch remains one of the most capital-intensive parts of SpaceTech, but the investment case has changed. Stoke, Isar and Relativity are being financed around reuse, production rate, larger payloads and sovereign access to orbit rather than another wave of small rockets chasing occasional launches.
Satellite manufacturing is starting to look much more like an industrial business. Apex, Muon, K2 and Loft are standardizing buses, producing spacecraft ahead of individual missions and designing factories for hundreds of satellites a year.
Some of the clearest businesses are surprisingly specialized. ICEYE sells radar imagery that works through clouds and darkness, Pixxel focuses on hyperspectral information, and OroraTech is building thermal sensing around one urgent problem: detecting wildfires quickly.
Communications and navigation are becoming startup markets of their own. Kepler already operates an optical relay network, Hubble has connected ordinary Bluetooth chips directly to satellites, while Xona and TrustPoint are trying to build positioning systems that reduce dependence on traditional GPS infrastructure.
Defense spending now shapes a large part of SpaceTech. Governments are buying maneuverable spacecraft, responsive launch, sovereign sensing, navigation, communications, tracking and servicing, giving startups paying customers for capabilities that may take much longer to develop as purely civilian markets.
Orbital mobility is further along than routine repair or refueling. Impulse and Exotrail have already flown transport missions, while Starfish, Astroscale, ClearSpace and Orbit Fab are gradually pushing the market toward docking, disposal, life extension and eventually propellant delivery.
The more speculative categories are becoming easier to separate from the real ones. Varda has repeatedly returned capsules from orbit, while private stations are assembling flight hardware and commercial lunar landers have already reached the Moon. Hyperscale orbital data centers, shared power grids and lunar resource extraction still have much bigger economic assumptions left to prove.
The useful way to read SpaceTech now is by maturity rather than hype. Satellite manufacturing, Earth observation, communications, mobility and orbital tracking already have functioning products and customers; servicing, alternative navigation and in-space manufacturing are crossing into commercialization; orbital power, giant space data centers and lunar mining remain early bets.
Across almost every category, the same underlying market is emerging: thousands of machines in space need the same kinds of supporting industries that appeared around aviation, shipping and terrestrial computing. Rockets opened access to orbit. SpaceTech startups are now trying to build everything required to keep that orbital economy running.
This market map, featured in our space economy deck, highlights top companies and startups in the space economy
Why does SpaceTech suddenly mean much more than rockets?
SpaceTech today covers a much bigger business than launch because thousands of spacecraft now need factories, communications, transport, navigation, servicing, power and software after the rocket has done its job.
The scale of orbital activity explains the change. BryceTech counted 325 orbital launches and 4,544 spacecraft deployed worldwide in 2025. Launches rose about 25% from the previous year, while the number of spacecraft jumped 54%. Smaller spacecraft accounted for 98% of everything deployed.
Private companies increasingly own what is going up. The OECD’s latest Space Economy at a Glance report calculates that private operators accounted for 88% of satellites launched in 2025, compared with just 23% in 2010. Much of that growth comes from large communications constellations, especially Starlink, so every extra satellite does not automatically represent a new startup market.
Investment is moving with it. BryceTech recorded $10.9 billion invested across 208 space startups in 2025, up from $7.8 billion the previous year. Its data also shows that a growing share of the money is going into later-stage companies. The OECD independently puts 2025 private space capital at roughly $11 billion to $13 billion, the highest level since 2021.
A broader tracker from Seraphim reaches an even larger number because it uses a wider definition of SpaceTech: $23 billion invested over the twelve months through Q2 2026, versus $9.7 billion over the comparable previous period. Those datasets do not measure exactly the same universe, but both point in the same direction.
What has changed is the number of credible things a startup can build after launch becomes available. Satellites need propulsion. Sensors need fast data links. Operators need collision warnings. Military customers want spacecraft that can maneuver. High-performance payloads need more electricity. Manufacturers want repeatable buses instead of bespoke projects.
If you want more recent data on this point, please see our latest space economy report.
Are SpaceTech startups still betting billions on rockets?
SpaceTech launch startups are still attracting enormous amounts of money, and the serious bets today are centered on reuse, production rate and sovereign access to orbit.
Stoke Space gives us the freshest example. The company has just raised $1 billion in a Series E, bringing total capital raised to $2.3 billion. Nova Pathfinder, its first orbital configuration, is targeting its first flight in the first part of 2027. Stoke has also revealed Nova Block 2, a larger vehicle designed to put 15 metric tons into low Earth orbit.
That amount of financing is striking because Stoke has yet to complete an orbital flight. Investors are effectively paying upfront for the possibility that another company can turn rapid full reusability into a production business.
Isar Aerospace has now cleared a threshold that most launch startups never reach. Spectrum reached orbit on its second flight, making Isar the first commercial company from Europe to deliver satellites successfully into orbit from continental Europe. The company subsequently confirmed that all customer payloads had been released and that most had established contact.
Isar raised €270 million in June 2026 and is ramping a 40,000-square-meter factory near Munich designed for as many as 40 Spectrum vehicles a year. Several additional rockets are already moving through production.
Relativity Space is making a similar bet at a larger scale. Its Terran R program completed first-stage structural qualification, and the company reported that more than 11,000 first-stage flight parts had been fully released by July. The U.S. Space Force has also added Relativity to the National Security Space Launch Phase 3 Lane 1 program, allowing Terran R to compete for future missions before its first flight.
The launch pitch has changed since the early small-rocket boom. Customers now care much more about flight rate, payload capacity, reuse and whether governments can rely on another launcher when they need one.
As this chart shows, and as featured in our space economy deck, search interest in the space economy has been rising steadily
Why are SpaceTech startups building satellites like factory products?
Satellite startups are currently turning spacecraft into repeatable products, and that may be one of the biggest industrial changes happening in SpaceTech.
Apex represents the clearest version. Its Aries, Nova and Comet satellite buses are sold as configurable platforms instead of being redesigned from scratch around every payload. Apex raised more than $200 million earlier this year at a $2.3 billion valuation. Its Factory One site is designed for peak output above 200 satellite buses per year.
Muon Space is pushing the same idea into full constellations. The company has deployed 11 satellites across six missions, has more than 50 additional spacecraft in development and recently raised $250 million in Series C financing. Its new manufacturing operation is designed to reach capacity of up to 500 satellites annually by 2027.
K2 Space is doing something slightly different. Cheap heavy launch gives K2 room to build much larger standardized satellites. Its Mega Class platform offers up to 30 kilowatts of peak payload power and roughly 3,000 kilograms of payload mass, while a first full-scale Gravitas spacecraft is already operating in orbit.
Loft Orbital shows where productization can lead commercially. Loft has procured more than 50 satellite buses from five suppliers and keeps spacecraft available before every final customer mission is known. It then integrates the customer's payload, arranges launch and operates the spacecraft.
Loft said in July 2026 that it had deployed more than 35 customer missions over five years and plans to grow its fleet to 30 satellites by the end of 2027. The French space agency CNES also awarded Loft and Magellium a multi-year Earth-observation contract worth up to tens of millions of euros.
Apex standardizes the bus. Muon standardizes much more of the constellation. K2 increases what a standard platform can carry. Loft buys buses ahead of demand and wraps integration and operations around them.
| Company | What is becoming standardized | Current scale |
|---|---|---|
| Apex | Configurable satellite buses | Factory capacity above 200 buses a year |
| Muon Space | Complete satellite constellations | 11 launched, 50+ in development, factory designed for up to 500 annually |
| K2 Space | Large high-power satellite platforms | Gravitas operating in orbit; Mega Class offers up to 3,000 kg payload mass |
| Loft Orbital | Satellite inventory, integration and operations | 50+ buses procured and 35+ customer missions deployed |
If you want more recent data on this point, please see our latest space economy report.
What are Earth-observation startups seeing that older satellites miss?
Earth-observation startups are currently winning by seeing one specific thing unusually well and getting that answer to customers faster.
Pixxel is doing it with hyperspectral imaging. Its operational Firefly satellites collect imagery across 135 spectral bands at about 5.4-meter ground resolution. That lets customers distinguish material characteristics that a normal color image can hide. The U.S. National Reconnaissance Office has already selected Pixxel under its Strategic Commercial Enhancements program to study how hyperspectral data can support intelligence work.
ICEYE attacks a different physical limitation. Synthetic-aperture radar works at night and through cloud cover, making it valuable when optical satellites cannot see the ground clearly. ICEYE has now launched 76 satellites, and its latest generation offers imagery down to 16-centimeter resolution.
Recent ICEYE launches included dedicated spacecraft for national customers, including Finland. Governments increasingly want their own guaranteed sensing capacity instead of depending entirely on imagery from a shared commercial fleet.
OroraTech has taken specialization even further. Greece now has four dedicated OroraTech thermal satellites built around wildfire detection. The system can send information to Greek emergency services within minutes and detect hotspots as small as roughly four by four meters. The original program represented about €20 million of investment.
Muon Space's FireSat network targets the same problem at global scale. FireSat is being built around a constellation that should eventually exceed 50 satellites, with sensors designed to identify very small fires before they become the massive events visible to conventional systems.
The startup opportunity today is often a very precise answer: Is there a new wildfire? What material is on this site? Can we see this location through heavy cloud? Has something moved since the last pass?
This chart, featured in our space economy deck, illustrates yearly venture capital funding for space economy startups
Are SpaceTech startups building new communications and navigation networks?
SpaceTech startups are currently building both new satellite communications networks and alternatives to traditional GPS, with some already operating commercially and others still deploying their first real constellations.
Kepler Communications provides the clearest current example on communications. Ten roughly 300-kilogram satellites launched at the beginning of 2026 have moved through commissioning and are now delivering commercial optical relay services. Instead of every customer satellite waiting to pass over its own ground station, data can move through optical links across Kepler's network.
Kepler also put distributed computing into the same architecture. Its first operational tranche carries 40 NVIDIA Jetson Orin modules across ten spacecraft. The company says its current network has contributed to roughly tenfold year-over-year revenue growth, while future satellites are being designed for communication rates reaching 100 gigabits per second.
Hubble Network is working at the other end of the bandwidth spectrum. Hubble has demonstrated direct communication between ordinary Bluetooth Low Energy chips and satellites. Seven operational spacecraft were in orbit after its 2025 deployments, and the company raised $70 million in Series B funding, bringing total financing to $100 million.
The individual Hubble messages can be tiny. That is enough for an asset location, a sensor reading or the status of industrial equipment without installing a conventional satellite terminal.
Astranis adds a third communications model. Its five small geostationary satellites provide dedicated networks from high orbit, and the company says it has more than $1 billion of contracted satellite services.
Navigation startups are building another type of network. Xona Space Systems uses satellites in low Earth orbit for its Pulsar positioning system instead of the much higher orbits used by GPS and other global navigation systems.
Xona has already flown a private navigation satellite, transmitted navigation signals from orbit and received commercial FCC authorization for those broadcasts. In March 2026, it raised $170 million in Series C financing to move Pulsar toward larger-scale deployment.
The lower orbit gives the signal a shorter distance to travel. Xona says that can produce stronger signals and better resilience for autonomous vehicles, robotics, construction, agriculture and defense. Topcon has secured early access to Pulsar, while companies including Murata and Furuno have worked with Xona around receivers and industrial applications.
TrustPoint is building a more explicitly independent system. Its planned C-band low-Earth-orbit constellation is designed to provide positioning, navigation and timing without relying on GPS, Galileo, BeiDou or GLONASS. SpaceWERX awarded TrustPoint $4 million earlier this year for an end-to-end demonstration using four satellites and four ground stations.
The maturity levels differ sharply. Kepler already has a commercial network. Hubble has operational satellites. Xona and TrustPoint still need much larger constellations before they can become widely used navigation infrastructure.
Why does so much new SpaceTech look like defense tech?
A large part of SpaceTech is becoming defense infrastructure because military customers are paying for faster spacecraft, persistent sensing, autonomous maneuver and sovereign control.
True Anomaly shows how quickly the capital has arrived. The company raised $17 million in 2023, $100 million later that year, $260 million in 2025 and another $650 million earlier this year. That adds up to more than $1 billion of disclosed financing in roughly three years.
True Anomaly's Jackal spacecraft is built for rendezvous and proximity operations, while its Mosaic software plans and controls missions. During the U.S. Space Force's VICTUS HAZE program, Jackal successfully detected and pursued Rocket Lab's Puma spacecraft after Puma maneuvered away.
Turion Space is following a related path with its DROID spacecraft. The company originally focused heavily on inspection, space-domain awareness and orbital logistics. It has since been selected to participate in U.S. Space Force work around space-based interceptors and has raised more than $85 million in Series B financing.
Impulse Space offers another example. Mira and Helios are transport vehicles, yet the company is increasingly tied to national-security missions. The Space Force added Helios to its National Security Space Launch contracting framework and later selected Impulse to support additional VICTUS SALO responsive-space missions.
Astranis is moving in the same direction from communications. Its newly announced Perceptor spacecraft applies its existing MicroGEO platform to tracking and responding to activity in strategically important high orbits.
The spending data explains why this keeps happening. According to the OECD's newest review, military programs represented about 46% of U.S. government space spending in 2025. Japan's defense-related space spending increased almost sevenfold between 2022 and 2025, while Germany has announced €35 billion of space-related defense investment through 2030.
Defense customers can fund capabilities years before a large civilian buyer exists. That influence now reaches launch, manufacturing, tracking, navigation, servicing and communications.
If you want more recent data on this point, please see our latest space economy report.
This chart, featured in our space economy deck, shows why SpaceX is leading in the space economy
Can startups move, refuel and service satellites after launch?
Orbital logistics and servicing are becoming real SpaceTech businesses, although moving and deorbiting spacecraft are currently much further along than routine refueling or repair.
Impulse Space has moved furthest toward building a second transportation layer after launch. Its Mira spacecraft performs orbital maneuvering and hosted missions, while Helios is designed to move payloads rapidly from low Earth orbit toward destinations such as medium Earth orbit and geostationary orbit.
Impulse raised $500 million in Series D financing in June 2026, bringing total capital raised above $1 billion. The company says it has flown three missions and accumulated hundreds of millions of dollars in customer contracts.
Exotrail is proving the smaller end of the same market. Its spacevan transfer vehicle has completed more than 1,000 electric-propulsion ignitions and more than 120 kilometers of cumulative altitude change. A subsequent mission sold capacity to six international customers for hosted payloads and satellite deployments.
Servicing is beginning to follow transportation. Early this year, the Space Development Agency awarded Starfish Space $52.5 million to provide end-of-life disposal for satellites in the Proliferated Warfighter Space Architecture. A few weeks later, Starfish received another $54.5 million U.S. Space Force contract for an Otter spacecraft that will dock with and maneuver satellites in geosynchronous orbit.
Together, those two awards are worth $107 million. Starfish has also completed three in-orbit demonstration missions and subsequently raised more than $100 million in Series B financing.
Astroscale's 520-kilogram ELSA-M spacecraft is intended to capture and remove an end-of-life Eutelsat OneWeb satellite. ClearSpace has PRELUDE targeting launch in 2027 for inspection, rendezvous and life-extension testing, followed by the ESA-backed €100 million ClearSpace-1 debris-removal mission planned for 2028.
Orbit Fab sits further out on the logistics curve. Its new architecture combines NEST orbital depots with RAVEN shuttles designed to carry about 150 to 200 kilograms of propellant to compatible spacecraft.
| Company | What it is trying to do | Where it stands now |
|---|---|---|
| Impulse Space | Move payloads after launch | Three missions flown; $1B+ raised |
| Exotrail | Last-mile deployment and hosted missions | Operational spacevan missions with paying customers |
| Starfish Space | Deorbit, dock and maneuver satellites | $107M across two major U.S. government awards; three orbital demos completed |
| Astroscale | Capture and remove end-of-life satellites | ELSA-M under development for a Eutelsat OneWeb removal mission |
| ClearSpace | Inspection, life extension and debris removal | PRELUDE targeted for 2027; €100M ClearSpace-1 targeted for 2028 |
| Orbit Fab | Refuel satellites | Interfaces developed; RAVEN shuttle and NEST depot network still being built |
Which SpaceTech startups keep crowded orbits from turning into a traffic mess?
Orbital-tracking startups are already selling the traffic and intelligence layer needed to understand where satellites are and what they are doing.
LeoLabs is the most mature example. Its network of ground-based radars tracks tens of thousands of objects in low Earth orbit and feeds that information into collision warning, maneuver detection and space-domain awareness services.
The company's commercial numbers give us a clearer view of demand than the size of its catalog. LeoLabs reported more than $60 million of contract awards during 2025. U.S. government contract value grew 186% year over year. It has since received another $20.68 million Space Force prototype award for a transportable radar system.
LeoLabs is also increasingly tracking behavior instead of simply position. Its software detects maneuvers and proximity activity, and the company says its network now observes more than 2,000 orbital maneuvers per day.
Digantara takes some of the sensing into space itself. The company is developing orbital electro-optical sensors under its Space-MAP architecture so objects can be detected and characterized from spacecraft rather than depending exclusively on ground radar.
Commercial satellite operators also contribute their own telemetry, creating a data problem that increasingly resembles air-traffic intelligence: radar observations, predicted paths, planned maneuvers and operator information all need to be combined quickly enough to make a decision.
Modern satellites can change course, servicing vehicles deliberately approach other spacecraft, and governments increasingly care about why a spacecraft moved as much as where it moved.
This chart, featured in our space economy deck, illustrates yearly funding for space economy startups
Can SpaceTech startups really manufacture valuable products in orbit?
In-space manufacturing has become technically credible, and startups are now testing whether pharmaceuticals and advanced materials can justify the cost of making them in microgravity.
Varda Space Industries has gone furthest in building the full logistics loop. Its W-series vehicles process materials in orbit and bring them back to Earth inside autonomous reentry capsules. The company has now completed multiple reentries instead of relying on one showcase flight.
Varda's W-4 mission spent almost a year in orbit and returned in May 2026 after carrying a pharmaceutical-processing experiment. W-5 reentered earlier in the year, and W-6 launched in March and returned in May carrying government experiments around hypersonic navigation and thermal-protection materials.
The U.S. Federal Aviation Administration has also granted Varda a vehicle operator license covering additional capsule reentries through 2029. Reentry is one of the hardest pieces of an orbital manufacturing business, so making it repeatable is a meaningful step.
Varda is still pushing pharmaceuticals as the commercial end market. In August 2026, it added former Pfizer R&D chief Mikael Dolsten to its board as the company works toward pharmaceuticals made in space for eventual use on Earth.
Space Forge is testing another high-value category: advanced semiconductor materials. Its ForgeStar-1 spacecraft has generated plasma more than 100 times in orbit using an onboard growth chamber. The company is trying to use microgravity to improve materials such as wide-bandgap semiconductor crystals.
These categories are attractive because value per kilogram can be extremely high. A small amount of pharmaceutical material or a high-performance semiconductor crystal could potentially justify expensive transportation.
The harder test comes next: repeat production quality, useful yields and customers buying the resulting materials at meaningful scale.
If you want more recent data on this point, please see our latest space economy report.
Are private space stations and Moon startups becoming real businesses?
Private space stations and lunar startups are now building real hardware and flying real missions, but both markets still rely heavily on government spending to create the first customers.
Vast currently provides the easiest station project to understand. Haven-1 is targeted for launch in 2027. The station is designed for four people, with roughly 45 cubic meters of habitable volume, 80 cubic meters of pressurized volume and a mass of about 14.6 tonnes.
Vast has moved into integration work on flight hardware. NASA has also selected the company to operate the sixth private astronaut mission to the International Space Station, targeted for no earlier than summer 2027.
Axiom Space is taking a different route. Its first station modules are being built so they can initially connect with the ISS architecture before eventually forming an independent Axiom Station. Thales Alenia Space has already been machining and welding primary structures for the first module.
Axiom closed more than $525 million of financing in June 2026, with the proceeds supporting human spaceflight, spacesuits and the station program.
Starlab has picked up one of the more useful commercial commitments in the sector. Mitsubishi Corporation increased its investment while also reserving and pre-purchasing research volume and laboratory capacity aboard Starlab.
The Moon is slightly further along operationally because commercial hardware has already landed there. Firefly Aerospace's Blue Ghost became the first commercial lunar lander to complete a fully successful soft landing on the Moon. NASA has continued buying missions since then, including another $144 million CLPS award announced in June 2026.
Firefly is also trying to reuse lunar infrastructure for something beyond delivery. Its Ocula service puts imaging telescopes on Elytra vehicles already being sent toward lunar orbit, allowing them to collect imagery for landing-site mapping, mineral analysis and cislunar monitoring after their original transport work.
Astrolab is building lunar mobility. Its smaller FLIP rover is scheduled to carry NASA instruments on an upcoming south-pole mission. NASA has also selected Astrolab as one of the providers moving forward on a crewed lunar rover for Artemis.
Interlune pushes the story into resources. The company received a $6.9 million NASA contract in May 2026 to develop technology that heats lunar soil and measures volatile gases such as helium-3 and hydrogen.
Interlune has also been lining up potential buyers before large-scale lunar extraction exists. Quantum-computing company Maybell Quantum has agreed to buy future helium-3, and cryogenic equipment maker Bluefors has announced a supply relationship.
Those commitments are early. Station operators still need enough paying users beyond NASA, while lunar resource companies still need to prove that extraction and return can work economically. Today, government-backed transport, research and infrastructure remain the foundation of both markets.
This chart, featured in our space economy deck, compares the main business model options for Earth observation satellite operators
Are orbital data centers and power grids actually happening?
Orbital computing already works at small scale, while giant space data centers and shared power grids are still trying to prove that their economics can work.
Kepler gives us the best baseline because its compute hardware is part of a commercial network operating today. The company's first ten optical-relay spacecraft carry 40 NVIDIA Jetson Orin modules between them. Customers can process information near the sensor, then move results through Kepler's optical network.
A satellite taking huge amounts of imagery may only need to send a small part of that imagery to Earth once onboard software has detected the interesting object, fire or event.
Starcloud is testing a much larger computing class. Starcloud-1 launched with an NVIDIA H100 GPU, the type of chip normally associated with terrestrial AI infrastructure. The spacecraft later ran a version of Google's Gemini technology in orbit and trained a small language model.
Lonestar is approaching orbital computing through storage and resilience. The company announced a commercial space-based storage service called StarVault earlier this year and has signed a NASA Space Act Agreement covering lunar and orbital data storage, computing and communications.
Power is starting to become a standalone market as heavier payloads ask more from satellite energy systems.
Star Catcher has raised $65 million in Series A funding, taking its total financing to $88 million. Its planned satellites would collect solar energy and beam concentrated optical power onto the solar arrays of other spacecraft, allowing those spacecraft to draw extra electricity when they need it.
Star Catcher says it now has ten major commercial customers under definitive power-purchase agreements and more than $60 million of backlog. One of the newest customers is Aethero, which is developing high-performance orbital computing hardware.
That pairing is useful: one startup wants heavier computing loads in orbit while another is being paid to supply the extra energy those loads may need.
Kepler's edge-computing network is operating commercially today. Starcloud has demonstrated a powerful AI GPU in orbit. Lonestar has started selling storage services. Star Catcher has customers, while the full orbital power-beaming network still needs its core end-to-end demonstration.
For now, edge computing is real. Hyperscale orbital cloud infrastructure still has to prove power, cooling, radiation tolerance, maintenance and cost.
So what are SpaceTech startups actually building now?
SpaceTech startups are currently building the infrastructure that makes a crowded orbital economy usable, with satellite manufacturing, sensing, communications, defense and mobility much further along than lunar mining or giant orbital data centers.
Launch is still enormous. Stoke has just raised $1 billion, Isar has now reached orbit and Relativity is assembling Terran R. Yet launch no longer explains the full startup landscape.
Some of today's strongest businesses begin after a payload reaches space. Apex and Muon are turning satellite production into something closer to manufacturing. Loft keeps spacecraft on a shelf. ICEYE sells radar intelligence from a fleet of 76 satellites. Kepler has started commercial optical relay service. Impulse has flown three orbital-mobility missions and raised more than $1 billion. LeoLabs recorded more than $60 million of contract awards in a year. Starfish has $107 million across two servicing awards.
Another group is crossing that line now. Varda has made capsule reentry repeatable enough to fly several missions. Xona is deploying an alternative navigation constellation. Private-station companies are assembling hardware. Firefly has landed on the Moon and continues to win lunar missions.
Then we reach the much earlier bets. Routine spacecraft refueling, lunar resource extraction, orbital power grids and hyperscale data centers have customers, financing or demonstrations in various combinations, yet each still has a major technical or economic assumption left to prove.
Government spending runs through nearly every level. Defense agencies are buying rockets, satellite buses, radar data, maneuverable spacecraft, navigation systems, tracking and servicing. NASA and ESA are doing the same around lunar transport, stations and debris removal.
The broader pattern is hard to miss. As more spacecraft reach orbit, the industry needs manufacturers, transport companies, communications networks, navigation, maintenance, traffic monitoring, electricity and computing.
Rockets opened the market. Startups are now filling in what is needed to keep thousands of machines working once they get there.
| SpaceTech area | What startups are building now | How real is it today? | Examples |
|---|---|---|---|
| Launch | Reusable and sovereign launch systems | Operational for some companies; new entrants still scaling | Isar Aerospace, Stoke Space, Relativity |
| Satellite manufacturing | Standard buses and constellation factories | Commercial and scaling quickly | Apex, Muon Space, K2 Space, Loft Orbital |
| Earth observation | Radar, hyperspectral and thermal intelligence | Mature startup market | ICEYE, Pixxel, OroraTech |
| Space communications | Optical relay, dedicated GEO links, direct-device connectivity | Commercial networks already operating | Kepler, Astranis, Hubble |
| Defense space | Maneuverable spacecraft, autonomy and domain awareness | Large contracts and heavy funding | True Anomaly, Turion, Astranis |
| Orbital mobility | Transfer vehicles and last-mile delivery | Already flying and selling missions | Impulse Space, Exotrail |
| Orbital tracking | Traffic monitoring and behavior detection | Commercial and growing | LeoLabs, Digantara |
| Satellite servicing | Deorbiting, docking and life extension | Major contracts; first full services still arriving | Starfish, Astroscale, ClearSpace |
| Alternative navigation | LEO positioning and GPS resilience | Early deployment | Xona, TrustPoint |
| In-space manufacturing | Pharmaceuticals and advanced materials | Repeated flight demonstrations; early commercial stage | Varda, Space Forge |
| Private stations | Commercial crewed habitats and laboratories | Flight hardware under development | Vast, Axiom Space, Starlab |
| Lunar infrastructure | Landers, rovers, imaging and resource systems | Real missions with heavy government backing | Firefly, Astrolab, Interlune |
| Orbital computing | Edge AI, storage and future data centers | Edge computing works; hyperscale remains experimental | Kepler, Starcloud, Lonestar |
| Orbital power | Shared power delivery between spacecraft | Customers signed; core network still pre-operational | Star Catcher |
If you want more recent data on this point, please see our latest space economy report.
This chart, featured in our space economy deck, shows revenue breakdown by customer segment in the space economy
OUR METHODOLOGY
This analysis asks what SpaceTech startups are actually building now and how far each part of the market has progressed. We separated SpaceTech into launch, satellite manufacturing, Earth observation, communications, navigation, defense, orbital mobility, servicing, tracking, in-space manufacturing, private stations, lunar infrastructure, orbital computing and orbital power, then examined each area independently.
For every category, we looked for the freshest concrete evidence available: hardware already flown, spacecraft deployed, missions completed, commercial services operating, production capacity, signed customers, contract awards, backlog, regulatory approvals and capital raised. Evidence that a product already works, is being manufactured or is being bought carried more weight than funding or announced capacity alone.
Different kinds of evidence were kept separate. BryceTech, the OECD and Seraphim Space use different definitions and time periods for private SpaceTech investment, so their totals were used to test the direction of the market rather than combined into one artificial figure. Planned factories, future constellations and announced missions were also separated from hardware and services already demonstrated in orbit.
Government procurement was treated as direct evidence of demand where governments are currently the first serious customers, particularly in defense, satellite servicing, lunar infrastructure and human spaceflight. That does not mean those categories already have broad commercial markets, so the article distinguishes between signed government demand and wider private-sector adoption.
The companies included were selected because they provide particularly clear evidence for individual parts of the market, rather than as a ranking of the “best” SpaceTech startups. The maturity assessment in the article comes from the combined evidence across deployments, customers, contracts, manufacturing, technical demonstrations and financing rather than from one company or one metric.
Key sources include BryceTech’s Global Orbital Activity 2025 report for launch and spacecraft deployment data, the OECD’s Space Economy at a Glance 2026 for satellite ownership, government spending and private-capital context, and Seraphim Space’s investment tracker for its broader SpaceTech funding measure. Company and mission evidence came from direct disclosures including Stoke Space, Isar Aerospace, Relativity Space, Apex, Muon Space, Pixxel, ICEYE, Kepler, Xona, TrustPoint, True Anomaly, Impulse Space, Starfish Space, Varda, the FAA, Vast, Firefly Aerospace and Star Catcher.
This chart, featured in our space economy deck, shows how satellite internet platform technology has evolved over time
Related blog posts
- SpaceTech: what are the biggest challenges now?
- SpaceTech: what’s changing now?
- SpaceTech: what are the biggest unsolved problems?
- The most recent funding news in the space economy
