The LEO Economy in 2026: Who's Making Money and How
Low Earth Orbit is no longer a research domain. It's a functioning economy with public companies, private equity, and trillion-dollar valuations. Here's who's making money in 2026 and how the next phase is shaping up.
The market, in numbers
The global space economy in 2026 is approximately $600 billion annually (Bank of America estimate), with LEO-related activity accounting for roughly 60% of that. The breakdown:
| Segment | 2026 estimate | YoY growth |
|---|---|---|
| Satellite broadband (LEO constellations) | $30B | +50% |
| Earth observation (LEO) | $8B | +18% |
| Launch services (mostly LEO delivery) | $15B | +12% |
| Space stations & in-orbit services (LEO) | $2B | +60% |
| Government LEO spend (NASA, DoD, others) | $80B | +8% |
The LEO economy is dominated by satellite broadband (Starlink, soon Kuiper) and launch services (SpaceX, Rocket Lab, others). The fastest-growing segment is in-orbit services.
Starlink: the dominant player
SpaceX's Starlink is, by any measure, the most successful commercial space business in history. As of mid-2026:
- ~7,500 satellites in operation
- ~5 million subscribers globally
- $12B+ projected annual revenue for 2026
- First profitable year: 2024
The unit economics are remarkable. A Starlink satellite costs SpaceX roughly $500k to build and launch (using a Falcon 9 ride-share). Each satellite generates ~$2,500/year in subscriber revenue. Payback period for a single satellite: under 3 years.
The competitive moat is real. SpaceX builds and launches its own satellites. No one else has that combination. Kuiper will be able to launch on multiple providers, but at higher cost per kg to orbit.
Kuiper: the challenger
Amazon's Project Kuiper is the only realistic near-term competitor to Starlink. As of mid-2026:
- First production satellites launched April 2026 on ULA Atlas V
- ~200 satellites in orbit as of June
- Commercial service launch targeted Q1 2027
- Total planned constellation: 3,236 satellites
Amazon's bet is that consumer broadband is not the only market. Kuiper is positioning for enterprise and government customers with custom ground hardware and SLAs. Whether that wedge is big enough to challenge Starlink's consumer dominance is the open question for 2027.
The Earth observation business
Earth observation is a more mature market with multiple players:
- Planet Labs — daily global imagery, ~$250M ARR, public company (PL)
- Maxar — high-resolution imagery, ~$1.7B ARR, private
- BlackSky — real-time monitoring, ~$100M ARR, public (BKSY)
- ICEYE — SAR (radar) imagery, ~$200M ARR, private
- Capella Space — SAR imagery, ~$80M ARR, private
The customers are split roughly evenly between government/defense (60%) and commercial (40%). The commercial side is driven by agriculture, forestry, insurance, and financial services (commodity trading, supply chain monitoring).
Launch services
The launch market in 2026:
| Provider | Vehicle | 2026 launches | Cost/kg to LEO |
|---|---|---|---|
| SpaceX | Falcon 9 / Falcon Heavy | ~150 | $1,500-3,000 |
| Rocket Lab | Electron / Neutron | ~25 | $5,000-15,000 |
| ULA | Atlas V / Vulcan | ~10 | $8,000-15,000 |
| Blue Origin | New Glenn | ~5 | $4,000-8,000 |
| China (various) | Long March family | ~70 | $4,000-10,000 |
The two big stories for 2026 are:
- Starship reaching orbit — SpaceX aims for regular Starship orbital flights in late 2026, with a target cost of <$200/kg to LEO. If achieved, this changes every commercial math in the industry.
- Neutron's first flight — Rocket Lab's mid-lift vehicle, expected late 2026. Could be a strong competitor in the 1-15 ton payload class.
In-orbit services: the new frontier
The fastest-growing segment is in-orbit services. Companies building this market:
- Northrop Grumman — Mission Extension Vehicle (already demonstrated in GEO)
- Astroscale — Debris removal (multiple contracts with JAXA and UK Space Agency)
- Momentus — Orbital transfer vehicles (Vigoride)
- Atomos Space — Satellite orbital transfer
- Orbit Fab — In-orbit refueling
The pitch to investors: as more satellites launch, the addressable market for servicing them grows. The pitch to operators: extending a $100M satellite's life by 3 years via a $30M refueling mission is good economics.
What to watch in 2026-2027
Three things to keep an eye on:
- SpaceX IPO rumors. Persistent chatter about a Starlink IPO in 2026 or 2027. A public Starlink would be the largest financial event in space history.
- Commercial space stations. Axiom, Vast, Voyager — all targeting 2028-2030 for first commercial station modules. ISS is currently scheduled for retirement in 2030.
- Lunar LEO services. NASA's planned Lunar Gateway at NRHO (near-rectilinear halo orbit) will be a separate market from LEO proper, but the suppliers will largely be the same.
The LEO economy is no longer a forecast. It's a measured, growing, increasingly competitive market.
Where the LEO economy sits in mid-2026
As of mid-2026, the LEO economy has three distinct segments. The first is broadband mega-constellations (Starlink, Kuiper, Guowang, Qianfan) — the largest segment by both deployed mass and revenue, but also the most capital-intensive and the slowest to return capital. The second is Earth observation (Planet Labs, BlackSky, Satellogic, Capella, ICEYE) — a smaller segment by revenue but with higher gross margins and clearer paths to profitability through commercial and government contracts. The third is in-space services (orbital transfer vehicles, life extension, debris removal, manufacturing) — the smallest segment by deployed mass but the segment with the most optionality and the longest lead times to revenue.
Why the segments behave differently
Broadband constellations require continuous capex — satellite refresh cycles, gateway deployment, ground station buildout — and the revenue ramps slowly because each new satellite adds incremental capacity rather than a discrete new customer. Earth observation is closer to a SaaS business model: a satellite is built and launched once, then generates recurring revenue from data subscriptions and image sales with minimal marginal cost per customer. In-space services is the highest-margin segment because the customer is typically a government or a satellite operator with a specific high-value problem (a tumbling spacecraft, a defunct satellite that needs to be deorbited, a payload that needs to be moved to a different orbit).
The investment implications are different across segments. Broadband is a play on Starlink (which is closed to outside investors) and Kuiper (which is hidden inside Amazon). Earth observation has publicly traded pure-plays (Planet Labs, BlackSky) and is the easiest segment to gain exposure to. In-space services is private (K2, Atomos, Starfish, Astroscale, etc.) and is accessible only through venture or growth-equity structures. The risk profile is highest in broadband (capex overruns, regulatory pressure from ITU on spectrum), intermediate in in-space services (technology risk, slow customer ramp), and lowest in Earth observation (mature technology, established customers).
What the next 12-24 months will resolve
Three things will become clearer in the next 12-24 months. First, whether Kuiper can reach commercial-scale deployment. The first production satellites reached orbit in May 2026; the cadence through the rest of 2026 will determine whether Kuiper becomes a real competitor to Starlink or remains a follower. Second, whether the Earth observation companies can convert their data backlog into recurring revenue at the pace their valuations imply. Third, whether the in-space services segment produces its first commercial success at scale — meaning a mission that solves a multi-million-dollar problem for a non-government customer, paid for by the customer's own budget rather than a government contract.
If all three resolve positively, the LEO economy becomes investable through a much broader set of public companies than it is today. If one or more resolves negatively, expect consolidation in that segment. We are tracking each of these threads and will update this analysis as conditions evolve.