Is Starlink becoming a new kind of hybrid telco?

Starlink began as a niche satellite broadband provider serving customers beyond the reach of traditional telecom networks. Today, SpaceX, the company behind Starlink, appears to be pursuing a far more ambitious goal: becoming a fully-fledged hybrid telco combining satellite coverage with terrestrial spectrum and infrastructure. Is the telecom industry’s next major player another operator, or a space company?

In 2020, when Starlink was still in its infancy, Elon Musk insisted it was “not some huge threat to the telcos” because it intended to serve customers beyond the economic reach of fibre and mobile networks.

Today, the picture looks very different. SpaceX has filed a trademark application for Starlink Mobile, and executives have confirmed plans to build a terrestrial mobile network alongside Starlink’s satellite layer. Gwynne Shotwell, the company’s president and COO, has also indicated an ambition to sell mobile contracts directly to consumers and compete with AT&T, T-Mobile and Verizon.

The important question is therefore no longer whether satellites can replace terrestrial mobile networks – they cannot, at least not in the foreseeable future. It is whether SpaceX can combine satellite coverage, terrestrial spectrum, selective radio infrastructure and its own backhaul into a fundamentally different kind of telecom operator.

Starlink has already moved well beyond its original rural broadband niche. It now serves enterprise, maritime, aviation and direct-to-device markets, while SpaceX’s reusable launch capability supports the continued expansion of network capacity. It is the biggest ISP in rural New Zealand, has partnered with GCI to extend connectivity into remote areas of Alaska, and has been used by operators such as Liberty Latin America to provide emergency direct-to-cell connectivity following Hurricane Melissa in Jamaica.

Starlink is increasingly both a competitor to, and a component of, terrestrial telecom networks.

SpaceX’s spectrum acquisitions from EchoStar gave the company access to approximately 65 MHz of valuable mid-band spectrum: 50 MHz through a $17 billion transaction, followed by a further 15 MHz for approximately $2.6 billion. These frequencies have long been described as “orphan spectrum” because they have lacked the handset ecosystem enjoyed by more widely deployed cellular bands. Together, they provide enough spectrum to establish a network, although not necessarily enough to replicate the capacity of the incumbent MNOs in dense markets.

Until now, Starlink’s direct-to-device model has largely depended on partnerships with operators such as T-Mobile and spectrum already supported by their customers’ smartphones. Owning spectrum gives SpaceX something it did not previously have: the option to move beyond its role as a satellite infrastructure partner and become a hybrid operator and potentially a direct retail competitor.

Recent dealmaking involving Amazon, Globalstar, Rocket Lab and others suggests that control of spectrum suitable for direct-to-device services is becoming a strategic asset in its own right. SpaceX can remain a wholesale partner where that model makes sense while preserving the option to launch direct retail services where regulation and economics are favourable. Simply having that credible alternative also strengthens its negotiating position with incumbent MNOs. This hybrid setup would greatly weaken the boundary between networks; a customer could begin a session on terrestrial 5G, move onto satellite coverage while travelling through a remote area and return to terrestrial service without noticing the transition.

SpaceX’s regulatory strategy reinforces this interpretation. Ahead of the planned upper C-band auction, the company has asked the FCC to allow satellite coverage to count towards certain build-out obligations attached to those licences.

Traditionally, spectrum licences require operators to demonstrate that they are putting those frequencies to use through terrestrial deployment. SpaceX is arguing that, in some circumstances, coverage delivered from orbit should count too. If regulators agree, the implications could be significant.

A conventional operator may have to build expensive rural sites to meet coverage obligations even where customer density is low. A hybrid operator could potentially satisfy some of those requirements from orbit while deploying terrestrial infrastructure only where capacity or economics justify it.

That would not eliminate the need for towers, but it could reduce the number required to establish a credible national footprint.

In effect, SpaceX is testing whether a nationwide mobile network still has to mean a nationwide tower network.

None of this means terrestrial networks are becoming obsolete. Dense cities will continue to favour terrestrial infrastructure because it provides superior capacity at lower cost. Up to 80% of data is consumed indoors, where concrete, steel and glass significantly attenuate satellite signals; indoor traffic is therefore better served by fibre, Wi-Fi and terrestrial cellular infrastructure.

Cross-border operation may prove an even greater challenge; unlike terrestrial networks, satellites routinely cover multiple countries simultaneously, yet spectrum licences remain national. As users move between countries, particularly in regions such as Europe with many neighbouring states of differing sizes, questions remain over direct-to-device spectrum rights, roaming arrangements, emergency service obligations and regulatory approvals.

In short, satellites are best viewed as an extension to terrestrial coverage, rather than a replacement for it.

That’s also the opinion of T-Mobile CEO Srini Gopalan, who has publicly played down the threat posed by SpaceX, arguing that satellite service is fundamentally complementary to terrestrial networks. On T-Mobile’s Q1 2026 earnings call, Gopalan said T-Satellite was seeing “a lot less usage than we were originally thinking”, adding that most usage, as you’d expect, was occurring in national parks.

Starlink’s broadband business already provides a reminder that satellite capacity is geographically finite. Rival broadband provider Shentel has pointed to Starlink’s demand surcharges in high-demand areas as evidence of those capacity constraints. Shentel is an interested competitor rather than a neutral observer, but the underlying point is relevant: Starlink can extend coverage very efficiently, while serving concentrated demand remains a different economic problem.

The ability to provide a signal almost anywhere on Earth is an extraordinary achievement. Satellites have an obvious advantage in places where terrestrial infrastructure is difficult or uneconomic to build, but delivering huge volumes of traffic across cities, airports, stadiums and other high-density environments is a different challenge. Satellites are exceptionally well suited to solving the coverage problem, while terrestrial infrastructure remains better suited to solving the capacity problem.

SpaceX is assembling the components of a vertically integrated communications business spanning launch capability, satellite infrastructure, terrestrial spectrum, enterprise networking and potentially nationwide mobile services.

Whether the strategy succeeds remains uncertain. The incumbents possess far greater spectrum depth, mature terrestrial networks, established distribution and enormous operational experience.

But SpaceX does not necessarily need to replace them. It could combine satellites for coverage with terrestrial cells added selectively for capacity, use Starlink for backhaul where fibre is uneconomic and rely on MNO partners to fill the remaining gaps. This combination could give its mobile proposition a fundamentally different cost structure.

SpaceX has proposed adding that terrestrial layer by combining small cellular base stations with Starlink equipment already installed at customer premises. The idea has a loose precedent in cable: Comcast uses the Wi-Fi capability of millions of customer gateways to offload Xfinity Mobile traffic from its MNO partner. SpaceX could attempt something analogous with licensed cellular spectrum, while using Starlink itself for backhaul. However,  wireless experts have questioned whether a customer-driven installation footprint could provide the coverage and capacity of a planned national radio network.

Traditional BSS can already manage roaming and wholesale relationships, but hybrid connectivity introduces a more fluid mix of access networks, partners and commercial roles. Future platforms will need to support real-time charging, partner management, converged product catalogues, wholesale settlement and policy-driven orchestration across those combinations. If every hybrid service requires months of systems integration before launch, any competitive advantage will disappear long before products reach the market.

Operators working with Starlink, AST SpaceMobile or Amazon Leo (formerly Project Kuiper) can integrate third-party satellite coverage while retaining the customer relationship. SpaceX may eventually occupy several positions within the value chain simultaneously, acting as infrastructure provider, wholesale partner and even a retail competitor.

This creates opportunities for CSPs and specialist providers to aggregate terrestrial and non-terrestrial networks into a single managed service, allowing customers to buy connectivity rather than a particular access technology. The complexity of selecting networks, enforcing SLAs, managing suppliers and settling revenues would then move behind the scenes.

SpaceX therefore does not necessarily have to choose between becoming an MNO and remaining an MNO partner. Owning spectrum gives it the option to pursue direct retail services where the economics are attractive, while satellites, selective terrestrial infrastructure and partner networks provide the underlying connectivity. It may not displace terrestrial operators in dense markets, but it could redefine a telco as an orchestrator of coverage, capacity and commercial relationships across multiple networks.

For CSPs, the implication is clear: connectivity alone is not enough. They must be able to define cross-network services, manage partner agreements, measure consumption, allocate revenue among participants and explain the resulting charges.

The HINT for Supply Chain Catalyst, demonstrated by Cerillion at DTW26, showed how CSPs can combine terrestrial and non-terrestrial connectivity in a single managed service. If you are looking to monetise hybrid connectivity across a complex partner ecosystem, get in touch to find out how Cerillion can help.