Amazon’s $11.57 billion agreement to acquire Globalstar gives its satellite business a faster route into direct-to-device services, deeper access to mobile satellite spectrum, and continuity for Apple’s existing satellite features on iPhone and Apple Watch. What it does not give Amazon is the one advantage SpaceX has already converted into dominance: the ability to launch at industrial speed.
On April 14, Amazon agreed to acquire Globalstar, the Louisiana-based satellite operator whose network already underpins Apple’s satellite-enabled emergency and messaging features. Under the deal, Amazon gets Globalstar’s satellite operations, infrastructure, assets, and mobile satellite spectrum licenses, and says those assets will help Amazon Leo — the satellite network formerly known as Project Kuiper — add direct-to-device, or D2D, services beginning in 2028. In plain English, Amazon is not just buying more space hardware. It is buying a faster path into the next layer of the space economy: connectivity that follows the user, not the dish.
What Amazon is really buying
It is tempting to see this as a simple scale play. It is not. Globalstar brings roughly two dozen satellites to the table, while Globalstar’s own results showed just $273 million in 2025 revenue. On headline terms, that means Amazon is paying roughly 42 times trailing annual revenue, and offering shareholders a premium of more than 31% to the April 1 closing price. That is not the kind of multiple buyers pay for a steady, mature cash-flow business. It is the kind of price they pay for spectrum, regulatory position, operating history, and a shortcut into a market they believe is about to matter a lot more.
That matters because Globalstar is not just a small satellite company with a recognizable customer. It is already part of a live consumer service stack. Amazon and Apple said the new arrangement will keep satellite features running on supported iPhone and Apple Watch models, including emergency messaging and roadside assistance, while also giving Amazon access to Globalstar’s existing and planned upcoming constellations being manufactured by MDA Space. In other words, Amazon is buying not just current capability, but a foothold in a product category consumers already understand and use.
The problem money does not fix
And yet the hardest part of the race has not changed. Amazon’s problem has never been the absence of ambition. It has been the calendar. Amazon’s own mission update said that, as of its April 4 launch, it had deployed 241 spacecraft into orbit in support of its initial constellation. Reuters, counting the broader program, put the figure at 243 satellites against a planned first-generation network of 3,236. Earlier this year, Amazon asked the FCC to move its interim milestone deadline from July 30, 2026 to July 30, 2028. For a company of Amazon’s balance-sheet strength, that is a revealing request. It says the constraint is not financing. It is deployment velocity.
That distinction is critical. Amazon is not sitting around waiting to build satellites. Its Kirkland, Washington, production facility has capacity to build as many as 30 satellites a week, and Amazon says it already has hundreds of flight-ready satellites on standby, with more than 200 packed into six fully stacked payloads in Florida as of mid-March. The company’s bottleneck is getting those satellites into space on a dependable schedule.
The irony is that Amazon has not been timid about securing launches. In 2022, it announced up to 83 launches across Arianespace, Blue Origin, and United Launch Alliance, calling that the largest commercial procurement of launch vehicles in history. By January 2026, its FCC filing said the total had grown to 102 launches across four providers: 18 on Ariane 6, 24 on New Glenn, 38 on Vulcan, 9 on Atlas V, and 13 on Falcon 9. And still, the same filing says Amazon completed only 7 of the more than 20 launches originally scheduled for 2025, after delays tied to vehicle readiness, range access, weather, and other issues outside Amazon’s control. That is the deeper point of the Globalstar deal: Amazon can buy spectrum faster than it can buy certainty.
Why Starlink’s lead is different in kind, not just size
This is where comparisons with Starlink become uncomfortable for Amazon. SpaceX’s advantage is not simply that it got there first. It is that it controls the launch machine as well as the satellite network. Starlink’s subscriber base crossed 10 million in February, becoming the world’s largest satellite operator with just over 10,000 satellites in orbit. That is not merely a numerical lead. It is evidence of an operating model in which rockets, cadence, and network expansion all sit inside the same industrial system.
Starlink is also no longer just a broadband story. In the U.S., T-Mobile’s T-Satellite with Starlink now offers texting, photo sharing, location sharing, text-to-911, and access to select apps in areas beyond terrestrial coverage. That means the direct-to-cell market Amazon wants to enter is not hypothetical and not empty. A competitor is already teaching consumers what satellite fallback connectivity can look like, even if the experience is still constrained by bandwidth and coverage limitations.
Amazon’s strategy is not identical to SpaceX’s
Still, Amazon is not trying to copy Starlink line for line. Its strategy looks more enterprise-heavy, more operator-friendly, and more tightly connected to AWS. In his April 9 shareholder letter, CEO Andy Jassy said Amazon Leo is scheduled to launch commercially in mid-2026 and already has meaningful revenue commitments from enterprises and governments. He named Delta, JetBlue, AT&T, Vodafone, DIRECTV Latin America, Australia’s National Broadband Network, and NASA among the early customers. Delta separately said Amazon Leo will begin rolling out across 500 aircraft in 2028, while Vodafone said it would use Leo to backhaul 4G and 5G masts in remote parts of Europe and Africa. That is a different market-entry play from selling as many consumer terminals as possible as quickly as possible.
That distinction makes the Globalstar acquisition more logical than it first appears. If Amazon’s opportunity is not merely rural home broadband, but a broader connectivity fabric spanning aircraft, remote towers, enterprises, governments, and phones, then Globalstar’s spectrum and Apple-linked consumer relevance become far more valuable. Amazon says its D2D system will work with mobile network operators and support voice, text, and data beyond terrestrial coverage. Combined with Leo’s broadband network, this starts to look less like a Starlink imitation and more like an attempt to build a two-layer communications stack: terminals for high-throughput broadband on one side, device-native continuity on the other.
Why the acquisition still matters
The easiest bearish take is that Amazon has paid a giant sum for a nice strategic accessory. That may be too glib. In markets like satellite communications, time-to-position can be as important as time-to-market. Buying Globalstar gives Amazon licensed spectrum, regulatory footholds, operating infrastructure, and a working consumer use case connected to Apple. Those are not things a rival can replicate quickly, even with billions in capital. In that sense, Amazon may be trying to do something sensible: compress time in the part of the market where time is still compressible.
But the trade-off is clear. The acquisition improves Amazon’s future product map far more than it improves its immediate competitive position. The company still needs launch cadence. Its own materials say initial New Glenn missions can carry 48 satellites, initial Vulcan missions 40, and Ariane 64 missions 32. If those heavy-lift systems settle into a reliable rhythm, Amazon’s satellite backlog, enterprise demand, and new D2D assets could begin to reinforce one another quickly. If not, then the Globalstar deal risks becoming a very expensive hedge against a bottleneck that still sits elsewhere.
That is why this deal is more interesting than a standard “Amazon challenges Musk” headline suggests. Amazon has bought itself spectrum, Apple continuity, and a more credible mobile-satellite future. It has not bought parity with SpaceX. In today’s space economy, the scarce resource is no longer only capital, and not even demand. It is the ability to turn satellites on the ground into functioning coverage in orbit on schedule. Until Amazon proves it can do that at scale, Globalstar looks less like the answer to Starlink than a sharp, rational purchase made in recognition of what the real bottleneck has become.
