Top LEO Satellite Stocks 2026: Orbital Networks Get Real

LEO used to be a fantastic place to burn money and make PowerPoints.

In 2026, that changed.

SpaceX went public. Amazon agreed to buy Globalstar. Rocket Lab agreed to buy Iridium. Telesat landed a giant Canadian military satcom contract and expanded Lightspeed. AST SpaceMobile kept putting much bigger BlueBirds into orbit. Earth-imaging companies started selling less “pretty satellite picture” and more “give the analyst an answer now.”

One simple way to see how far the market has moved: Starlink finished 2023 with 2.3 million subscribers, 2024 with 4.4 million, 2025 with 8.9 million, and reported 12.0 million at June 30, 2026.

Starlink economics won’t automatically carry over to every LEO constellation. The category has still graduated from “science project” to “real infrastructure.”^[1]

We’re slicing LEO satellite stocks four ways: Orbital Networks, Earth Intelligence, Picks & Shovels, and Special Situations.

These are the companies trying to own the network itself: the satellites, spectrum, terminals, customers, and recurring service revenue. That gives them the biggest upside if the network works, and the ugliest capital bill if it doesn’t.

SpaceXNASDAQ: SPCX

SpaceX is now the benchmark every other LEO operator gets compared against, which is both flattering and slightly unfair. Starlink had 12.0 million subscriber service lines at June 30, 2026, and SpaceX reported $4.29 billion of Q2 Connectivity revenue, up 65.8% year over year. Its earlier IPO filing said the network had more than 9,600 Starlink broadband and mobile satellites in orbit at March 31. (Q2 2026 filing)

The interesting part now is capacity, not proof of demand. SpaceX has said its next-generation V3 Starlink satellite is designed for roughly 1 Tbps of downlink capacity per satellite, with Starship intended to deploy up to 60 at a time. If that works, SpaceX gets the nasty combo competitors hate: the network owner also owns the rocket and can push its own marginal deployment cost down. The catch for investors is obvious: post-IPO SpaceX is no longer a clean Starlink stock. It also contains launch and an AI business, so you’re buying a much broader platform.

AST SpaceMobileNASDAQ: ASTS

AST SpaceMobile is the cleanest public bet on one specific idea: your normal phone connects to a satellite without a special Starlink-style dish.

By August, AST said it had agreements with 60+ mobile network operators covering more than 3 billion subscribers, about $1.30 billion of contracted revenue backlog, and BlueBirds 17 through 46 in various stages of production and assembly. BlueBirds 11, 12, and 13 launched on August 5. (Q2 2026 update)

That’s why ASTS stays on the list: it wants to become the roaming layer above Verizon, AT&T, Vodafone, and other carriers. If the carrier-partner model converts into paid commercial usage at scale, the distribution advantage is enormous. If launch cadence slips, satellites underperform, or carrier economics disappoint, there is nowhere to hide. This is still an execution stock wearing a telecom-sized opportunity.

Telesat (NASDAQ/TSX: TSAT)

Telesat is the “boring old satellite company suddenly building one of the most important new networks” entry.

Its Lightspeed plan got materially bigger in August after Telesat signed a C$2.3 billion Canadian military satcom services contract, with options taking total potential value to C$2.7 billion. The company expanded the fully funded Lightspeed constellation to 225 satellites, and said pro forma LEO backlog would be about C$5.6 billion. The first Lightspeed satellite launch is targeted for December 2026. (contract announcement)

The attraction is that Telesat has real operator experience, spectrum, government relationships, and now a much more tangible anchor customer. The ugly bit: the legacy GEO business is declining while the company funds a huge LEO transition. Q2 2026 revenue fell to C$79.5 million from C$106.1 million a year earlier. Lightspeed increasingly is the thesis.

Eutelsat (Euronext Paris/LSE: ETL)

Eutelsat is the listed home of OneWeb, giving investors exposure to a network that is already global. The company operates more than 600 OneWeb satellites in LEO alongside its GEO fleet, making it the first fully integrated GEO-LEO satellite operator.

For FY 2025-26, LEO revenue reached €297 million, up 69.5%, and accounted for 25% of group revenue. (FY 2025-26 results) The next leg is as much political as commercial: Eutelsat has a central role in Europe’s IRIS² sovereign-connectivity build while also funding OneWeb replenishment and expansion. The tradeoff is that ETL carries a capital-intensive GEO business too, so the stock is broader than the LEO thesis.

Broadband gets the headlines. Earth intelligence can get the cleaner business model.

Here, the product is timely, differentiated data for governments and enterprises that care what changed on Earth today, sometimes this hour.

Planet LabsNYSE: PL

Planet Labs operates roughly 200 Earth-imaging satellites and built its reputation on imaging the world’s landmass every day. The more interesting 2026 story is the move upmarket: higher-resolution Pelican satellites, sovereign reconnaissance programs, and more defense-oriented contracts.

In fiscal Q2 2027, Planet posted record $116.1 million of revenue, up 58% year over year, with cash, cash equivalents, and short-term investments of about $865 million. The quarter included the Pelican Tech Demo launch and shipment of Tanager-2 and SuperDove satellites for launch. (Q2 2027 results)

Planet is what happens when LEO becomes a data business instead of a hardware business. The satellite fleet matters, but the real product is the recurring dataset and the workflow built around it. Watch whether the backlog turns into durable subscription growth without requiring endless hardware spending to keep the constellation fresh.

BlackSkyNYSE: BKSY

BlackSky is smaller, more defense-heavy, and more focused on low-latency intelligence than Planet. Its Gen-3 satellites are designed around very-high-resolution imagery and rapid revisit, while the Spectra platform pushes analytics into the same workflow.

Q2 2026 revenue reached $33.3 million, up 50% year over year, including a record $25 million of space-based intelligence and AI services revenue. The company also said its next two Gen-3 satellites were expected to launch in Q3. (Q2 2026 results)

The bull case is simple: governments increasingly want commercial satellite intelligence that is fast, persistent, and available without waiting years for a bespoke national system. The bear case is equally simple: defense sales can be lumpy, procurement cycles are weird, and BlackSky still has to keep spending on the constellation. This one belongs on the watchlist because Gen-3 adoption is finally giving the story measurable operating traction.

Spire GlobalNYSE: SPIR

Spire Global is the “listening” company in the Earth-intelligence group. Its LEMUR nanosatellites collect radio-frequency data for weather, aircraft and ship tracking, spoofing and jamming detection, and other near-real-time intelligence. Spire also sells Space Services, letting customers put payloads or missions on its satellite platform without building the whole stack themselves.

Q2 2026 revenue was $18.0 million, down 6% year over year on a reported basis but up 16% excluding the maritime business it sold in 2025. Spire launched 10 satellites in July, bringing 2026 launches to 29, while continuing to add government work across NOAA, EUMETSAT, and defense programs. (Q2 2026 results)

This one is earlier-stage and messier than Planet or BlackSky, but the angle is distinct: RF sensing plus “space as a service,” instead of optical imagery. Watch whether contract momentum turns into sustained revenue growth and improving cash burn.

If picking the winning constellation feels like choosing the winning railroad in 1870, buy the people selling rails.

These companies make money when multiple LEO networks get funded, launched, replenished, and upgraded.

MDA Space (NYSE/TSX: MDA)

MDA Space may be the cleanest picks-and-shovels name in the group. It builds satellite systems and components for other operators, which means it can win even when investors argue about which constellation ultimately dominates.

Q2 2026 revenue was C$498.6 million, up 33.6% year over year, and backlog finished at about C$4.0 billion. Satellite Systems revenue grew 44.5%, driven primarily by higher work volume on Telesat Lightspeed. Then in August, MDA received another C$474 million order tied to the Lightspeed expansion. It also built the first eight Globalstar replacement satellites that launched on August 15. (Q2 results, Lightspeed expansion)

Think less lottery ticket, more “someone has to manufacture all this stuff.” The thing to watch is working capital: big satellite programs can make reported growth look gorgeous while cash flow gets temporarily ugly. Q2 free cash flow was negative C$150 million, so execution still matters.

Rocket LabNASDAQ: RKLB

Rocket Lab used to be easy to describe: Electron launches plus a growing spacecraft-components business.

Not anymore.

In June, Rocket Lab agreed to acquire Iridium for an implied enterprise value of about $8.0 billion, with the transaction expected to close in mid-2027 subject to approvals. That deal would give Rocket Lab an existing global LEO communications network, L-band spectrum, and recurring service revenue, basically turning the picks-and-shovels company into a vertically integrated network owner. (Iridium deal)

Meanwhile, Q2 2026 revenue hit a record $234 million, up 62% year over year, with backlog at $2.36 billion. (Q2 results)

The upside is obvious: build satellites, launch satellites, own the network, collect service revenue. The risk is also obvious: this is a much bigger company to integrate while Rocket Lab is still executing its Neutron roadmap and scaling multiple acquired businesses. RKLB is now one of the most strategically interesting names in space, and one of the easiest to overcomplicate.

This is where the technology thesis gets hijacked by lawyers, regulators, and merger consideration formulas.

Iridium CommunicationsNASDAQ: IRDM

Iridium Communications is what most aspiring LEO networks want to grow up into: global coverage, millions of billable subscribers, and a business where service revenue does most of the work. It ended Q2 2026 with 2.627 million billable subscribers and $225.2 million of revenue; service revenue was 72% of the total. (Q2 2026 results)

But IRDM is now a special situation. Rocket Lab agreed to acquire Iridium in June, with closing targeted for mid-2027 subject to shareholder and regulatory approvals. Until then, the stock’s upside and downside are tied partly to the operating business and partly to deal math. Strategically, the logic is the same one running through this whole watchlist: launch, satellites, spectrum, and recurring service revenue are starting to collapse into fewer, vertically integrated platforms.

GlobalstarNASDAQ: GSAT

Globalstar is still a real LEO operator with valuable spectrum, IoT service, and a major relationship with Apple. But as a stock, the thesis changed on April 14 when Amazon agreed to acquire it.

Globalstar shareholders can elect $90 in cash or 0.3210 Amazon shares per Globalstar share, with stock consideration capped at $90, cash elections capped at 40% of shares, and possible downward adjustment tied to satellite milestones. As of the August information statement, the maximum milestone-related adjustment had fallen from $110 million to roughly $97 million. The deal is expected to close in 2027, subject to remaining approvals and operational conditions. (merger filing)

Operationally, the company is still moving: Q2 revenue was $64.8 million, and the first eight replacement satellites launched August 15. (Q2 results)

But don’t confuse GSAT with a clean LEO growth stock anymore. The near-term trade is mostly the spread between the market price and the eventual merger consideration, adjusted for regulatory risk, Amazon’s share price, and satellite milestones. The strategic takeaway is bigger than the stock: Amazon is buying Globalstar specifically to add spectrum, satellites, and direct-to-device capability to Amazon Leo.

The first failure mode is economics.

LEO capacity can grow very fast. If several networks deploy thousands of satellites while launch costs fall, bandwidth can become a commodity. SpaceX already showed the tension in Q2: Starlink subscribers doubled year over year, but subscriber ARPU fell more than 20% as the company expanded internationally and added lower-priced plans. More users is great. More users at structurally lower unit economics is less great.

The second failure mode is execution.

These networks require launches, spectrum rights, ground infrastructure, terminals, replenishment satellites, regulatory approvals, and a ridiculous amount of capital. One delayed rocket, one production bottleneck, one spectrum fight, or one government procurement pause can move a catalyst by quarters.

The next 12–18 months will separate “has satellites” from “has a business.”

SpaceX’s V3/Starship deployment plan is the biggest capacity catalyst. AST SpaceMobile needs to turn its production pipeline and carrier agreements into broad commercial service. Telesat’s first Lightspeed launch in December 2026 is a major de-risking event. Eutelsat has to keep converting OneWeb’s scale and Europe’s IRIS² ambitions into durable LEO growth. Planet, BlackSky, and Spire need to prove that sovereign, defense, and intelligence demand can keep expanding recurring revenue.

Then there’s consolidation. Rocket Lab’s Iridium deal is targeting a mid-2027 close. Amazon’s Globalstar acquisition is also expected in 2027. If both close, the industry will have given us a pretty clear message: spectrum, launch, satellites, and distribution are more valuable together than apart.