Low Earth Orbit (LEO) constellations are not positioned as a global substitute, but rather as a set of solutions tailored to a mosaic of markets. The purpose of this article is to broadly outline their specific features and competitive positioning.
Introduction
The proliferation of low Earth orbit (LEO) satellite constellations is revolutionizing the space telecommunications economy, particularly by providing data transmission capabilities at a significantly lower cost than geostationary satellites (GEO). These new means are now being integrated into a wide variety of use cases.
The Legacy Market of TV/Video Broadcasting
The TV/video content broadcasting market is the satellite’s historical market. Here, the capabilities of geostationary telecom satellites are optimally exploited: one-way (one-to-many) signal distribution, wide coverage with just a few satellites, and no sensitivity to latency. This market is gradually eroding, both due to the migration of usage towards non-linear video content (YouTube, Instagram, etc.) and the development of terrestrial broadband access equipped with IPTV. LEO constellations do not offer any decisive advantage over GEOs here and are not expected to further disrupt this market, which will remain the most valuable until 2033, with players like SES or Eutelsat.
The Revitalized Fixed Telecommunications Market
For fixed telecommunications needs, LEOs are more of a complement than a substitute for fiber (FTTH/FTTO). The natural playground for satellites will be isolated areas: mountains, small islands, very low-density zones. This creates a dilemma for operators when extending fiber to the last subscribers, as the alternative based on Fixed Wireless Access (FWA) technology can sometimes be marginally cheaper.
Satellites can be seen as the "blind spot" of fiber: wherever fiber shows its limits—production issues, diversity of supply (backup), price (in some African countries)—there will be an opportunity for satellites. LEO technology offers similar performance to FWA and can substitute for it. Paradoxically, as fiber becomes more widespread, connecting the last subscribers will become increasingly critical, boosting demand for satellites.
By the end of 2025, Starlink had gained 9 million subscribers (70% of the market), compared to 1.6 billion broadband subscribers.
The Dynamic Market of Mobile Telecommunications
The mobile telecommunications market covers the equipment of vehicles in the broad sense (planes, trains, ships, and possibly trucks). Here, LEOs have their clearest competitive advantage, especially for non-terrestrial components (aerial or maritime). This market is rapidly developing with the digitalization of ships and the development of onboard Wi-Fi in airplanes. It is a high-value domain, representing up to a quarter of the space broadband connectivity market. In response to this threat, traditional players like Viasat/Inmarsat or SES are developing multi-orbit strategies to counter Starlink.
Specialized Markets: Defense and Emergency
For highly specialized markets such as defense or emergency intervention services, there are also few alternatives. Defense needs cover an extremely varied combination of communications: images/video, voice, drone piloting, sensor data collection, etc. These communications are potentially not fixed, as they are linked to the theater of operations, and some are completely mobile in the case of remotely piloted vehicles.
Space communications are also more robust, as part of the infrastructure is in space and therefore harder to destroy. The low latency of LEOs becomes a major tactical asset, which can be vital in combat situations. In the war in Ukraine, both belligerents already use commercial networks like Starlink alongside purely military means. The surge in military spending and the increasing robotization of armies offer promising prospects.
The Internet of Things (IoT)
Regarding Internet of Things (IoT) communications, satellites complement cellular solutions. The arrival of low Earth orbit solutions will reduce sensor costs and, above all, enable low-power use cases. Pioneers like Kinéis or Astrocast are already positioned in this niche, but with proprietary chips; they will be joined by players like Sateliot or Starlink, who aim for direct interoperability with standard cellular chips (5G NTN standard).
In any case, cellular solutions (95% coverage) will remain dominant, except for certain niches such as agriculture, energy, or maritime IoT.
The Hot Topic: Direct-to-Device Smartphone Connection (D2D)
The direct-to-device (D2D) market for standard consumer smartphones aims to provide connectivity everywhere and eliminate dead zones. There are already solutions with special phones, but these are premium and reserved for a niche professional clientele.
With D2D, two main value propositions emerge:
The first concerns extending coverage for already-served subscribers. Behind population coverage figures of around 99.5% in developed countries, territory coverage can be lower (from 95%); D2D promises to eliminate this gap.
The second value proposition concerns populations not covered by cellular networks (notably in sub-Saharan Africa) due to economic viability and low density. This represents a potential of 80 million people (out of 400 million uncovered), according to the ITU. This market could be partially subsidized by international donors.
In this market, mobile operators remain sovereign, as they have acquired exclusive usage rights by purchasing licenses. Satellite operators will rather act as connectivity providers behind incumbent operators (wholesale), especially as satellites are gradually being integrated into the 5G standard. The value that D2D can generate will depend on the usage value attributed by consumers in sometimes ultra-competitive markets. We're talking about fractions of euros per subscriber per month.
Market Projections
In summary, here is the market as projected by Novaspace for 2033:
| Segment | Value | LEO Share |
| Satellite — Broadband & Mobility | $1.5B | - |
| IoT | $1,5B | - |
| D2D | $10B | 90% |
| Video | $63B | 5% |
In terms of volume or bandwidth, the vision is very different, as 90% of the capacity should be provided by LEOs. For reference, the global telecom market should be around $2,500B–$3,000B in 2033.
Space Data Centers: The Final Frontier
The final frontier is that of data centers in space. These would have access to abundant (solar) energy, benefit from passive cooling in the vacuum of space (but would require the installation of thousands of square meters of exchange surfaces), extraterritoriality of orbits, and offer onboard computing functions to reduce application latency (edge computing).
Today, the economic equation is unfavorable: in 2026, the cost ratio would be 1 to 20 in favor of Earth. This may already make sense for military applications where functionality outweighs price. For intensive computing such as AI, parity is expected by 2035, when the cost of energy on Earth and the cost of launching into space (less than $100/kg) could converge (ASCEND study). The saturation of the electrical grid could also limit the installation of new data centers and be a development factor. Following this vision, Elon Musk chose to merge X.AI and SpaceX in a highly speculative bet.
Challenges Ahead
Many challenges lie ahead for LEOs:
- The role of states and their sovereignty needs, which could restrict uses/licenses.
- The impact of space debris (in 2025, Starlink performed 300,000 avoidance maneuvers for 5,000 satellites), which could drive up costs.
- Potential chronic congestion could become a bottleneck.
- The sustainability of the economic model for equipment that needs to be replaced every 5 years.
- Maintenance and update difficulties, with the impossibility of physical intervention.
- Sensitivity to cyberattacks or solar flares.
Conclusion
Satellite markets are highly segmented: by target (demanding B2B and unconnected populations), by use (industrial IoT and broadcast TV), and by geography (empty oceans and rural areas in developed countries). But above all, connection values (ARPU) differ considerably: from fractions of euros for intermittent connected objects to tens of thousands of euros for ultra-critical military drones.
For these reasons, depending on the use case, conclusions can be very optimistic (mobility) or very cautious (data centers). These markets are part of a technological, geopolitical, and industrial battle that is far from over, and whose outcome will depend not only on demand but also on the strategic choices of states and future technical performance.
