Amazon Files for 5,105 Direct-to-Device Satellites as SpaceX Approaches Its IPO
Amazon has filed an application with the Federal Communications Commission to deploy a constellation of up to 5,105 satellites beginning in 2028, intended to provide direct-to-device voice and data connectivity. The filing arrives as European operators Eutelsat and SES stand to receive roughly $6.1 billion in combined payments for helping clear 160 megahertz of satellite spectrum for United States wireless services, and as SpaceX moves toward a public listing.
Those three developments describe the same competition from three angles: who owns the spectrum, who owns the constellation, and what the resulting business is worth.
Direct-to-device is the high-value line, not the volume line
Satellite broadband to a fixed terminal is a known business with known economics. It serves customers beyond terrestrial reach, requires hardware at the premises, and competes against fiber and fixed wireless wherever those exist.
Direct-to-device is a different proposition. It connects an unmodified handset without any additional equipment, which makes the addressable market every phone rather than every underserved household. The service is technically constrained — link budgets to a handset antenna are severely limited compared to a dish, so capacity per user is low and the realistic product is messaging, basic voice, and emergency connectivity rather than broadband.
That constraint does not diminish the value. It concentrates it. A capability that extends coverage everywhere a subscriber travels is a feature carriers will pay to wholesale, and it attaches to existing subscriber relationships rather than requiring new ones. The revenue arrives through partnership rather than direct acquisition, at high margin, without terminal hardware costs.
The 2028 timeline is the operative detail
A filing is not a constellation. Amazon’s application covers deployment starting in 2028, which places first service well beyond the current competitive window and leaves considerable room between regulatory approval and operational capability.
For anyone modeling a SpaceX valuation, that timing cuts in a specific direction. It confirms that the most valuable satellite communications segment will be contested rather than monopolized, and it establishes the identity of the competitor as a company with effectively unlimited capital, its own launch relationships, and a consumer distribution surface no other entrant possesses. But it also confirms that competition is several years out, which means near-term direct-to-device revenue accrues to whoever is operating now.
The question for a valuation model is which of those facts the terminal value assumption depends on. A model that extrapolates current direct-to-device growth without a competitive entry assumption is extrapolating a monopoly with a filed expiry date.
The spectrum payments show where value is actually being priced
The Eutelsat and SES arrangement is the least discussed of the three items and arguably the most informative. Roughly $6.1 billion is changing hands to clear 160 megahertz for terrestrial wireless use in the United States.
That is the market pricing spectrum reallocation directly, and it establishes something useful: the incumbent satellite operators’ most monetizable asset in that band was not the service they were providing with it. It was the right to stop providing it. Any assessment of satellite communications equity should account for the fact that spectrum rights and orbital slots can be worth more in reallocation than in operation, which is a valuation floor that pure service-revenue models miss entirely.
What to watch between now and the listing
Three developments would materially change the competitive picture before Amazon’s constellation reaches orbit.
Carrier partnership announcements are the first, because direct-to-device economics run through wholesale agreements and exclusivity terms determine whether the market splits or consolidates. Regulatory treatment of the spectrum bands underlying handset-direct service is the second, since the technical approach depends on arrangements that remain contested internationally. Demonstrated capacity per satellite is the third, because the entire commercial case rests on how many concurrent users a given link budget supports, and published figures in this segment have consistently been optimistic relative to delivered performance.
A filing for 5,105 satellites is a statement of intent backed by a balance sheet. It is not yet a competitor. The interval between those two states is the window that a public market listing would be pricing.