Satellite communications: build, buy or hybrid

Every government satellite-communications contract is also a dependency decision. This page maps the four main procurement routes, what each one actually protects, and where the precedents sit on cost and time.

The dependency you already have

Most governments communicate via satellite today. Almost none own the satellite doing the work. That arrangement functions well until it does not: a foreign operator suspends service under export-licence pressure, a constellation provider re-prices at renewal, or a cable cut leaves an island group with no fallback. Tonga experienced the cable scenario in January 2022. The lesson was not unique to Tonga; it was a demonstration available to every procurement committee on earth.

Since 2022, the landscape has sharpened in other ways. Commercial operators have restricted coverage over active conflict zones under home-country government instruction. Shutter-control provisions, long a feature of imaging contracts, now have a communications-sector equivalent in practice if not always in name. A government that leases capacity rather than owning spectrum rights and ground infrastructure has limited recourse when those decisions are made. The question for a procurement committee is not whether to care about this. It is which rung of the sovereignty ladder is proportionate to the country's exposure, budget and timeline.

Four routes, four different answers to 'what do you own?'

Leased capacity on a foreign GEO satellite is the fastest and cheapest entry point. A government can be operational in months. What it owns is a service agreement, not an asset. If the operator withdraws, the capacity goes with it. This route suits interim connectivity or traffic overflow, not strategic resilience.

A sovereign gateway on leased capacity is a meaningful step up. The government owns and operates the ground infrastructure: the gateway earth station, the network management system, the spectrum coordination filings at the ITU. Traffic flows over a foreign satellite, but the routing, encryption and access control sit in national hands. Interruption of the space segment still breaks the link, yet the ground investment is not lost and can be repointed to a different operator or, eventually, a sovereign satellite. Several smaller nations have taken this route as a first move precisely because it builds institutional capacity without committing to a full space programme.

A hosted payload on a third-party satellite gives a government a dedicated transponder or antenna aboard a bus it does not own. The payload is sovereign; the platform is not. This is a reasonable middle path for nations that need guaranteed capacity and spectrum rights but cannot yet justify a dedicated spacecraft. The constraint is that the host satellite's orbital slot, inclination and lifetime govern what the payload can do. Hosted payloads on GEO platforms typically commit to a fifteen-year orbital position; that is useful if the slot suits you and limiting if it does not.

A fully sovereign GEO satellite or LEO constellation is the top rung. The government owns the spacecraft, the spectrum filing, the ground stations and, after handover, the operational team. No foreign operator can withdraw the service. The trade-offs are real: capital cost is higher, programme timelines run to years, and small nations may find that a single GEO satellite represents a concentration of risk rather than a distribution of it. A LEO constellation distributes that risk across multiple spacecraft but multiplies operational complexity. Neither option is obviously correct; the right answer depends on traffic volumes, coverage geometry, threat model and institutional capacity to operate.

Cost classes and timeline: what the public record supports

Sovereign gateway infrastructure, meaning a purpose-built earth station with network management and ITU-coordinated spectrum filings, has been delivered by several Pacific and African nations for figures in the low-to-mid single-digit millions of dollars, depending on throughput requirements and site preparation. That range is defensible from publicly reported programmes; specific figures for any given project require a site survey and traffic model.

Hosted-payload programmes vary enormously with the host spacecraft. Where a government is contributing a communications payload to an existing commercial or partner satellite, the payload hardware and integration costs for a modest transponder complement have been reported in the range of tens of millions of dollars, with timelines of two to four years from contract to in-orbit commissioning, depending on the host's launch schedule.

A dedicated small-to-medium GEO communications satellite, covering a national footprint with Ka- or Ku-band capacity, sits in a cost class that publicly reported national programmes, including several in the Asia-Pacific and Middle East regions, place in the low-to-mid hundreds of millions of dollars for spacecraft, launch, ground segment and initial operations. Programme timelines from contract signature to operational service have historically run four to seven years for first-time operators, though that has compressed as commercial satellite manufacturing has matured.

LEO broadband constellations of meaningful scale, say thirty or more spacecraft, carry higher aggregate spacecraft costs but lower per-unit manufacturing costs at volume. The IRIS2 programme, the European Union's multi-orbit governmental constellation, has a published budget in the billions of euros across its full scope. That is not a relevant benchmark for a national programme of ten to thirty satellites, but it illustrates that LEO constellation costs scale with ambition faster than GEO costs do. A small national LEO constellation of six to twelve satellites for connectivity and resilience is a different order of magnitude and has been attempted by several emerging space nations, though published out-turn costs remain scarce.

The decision framework

Three questions structure the choice. First: what is the minimum service level that must survive a foreign-operator refusal? If the answer is 'government-to-government secure voice and emergency broadcast', a sovereign gateway on leased capacity may suffice, provided the lease is with a non-aligned or treaty-bound operator. If the answer is 'national broadband for defence, civil aviation and disaster response simultaneously', the gateway route is insufficient.

Second: what is the country's spectrum position? ITU filings for orbital slots take years and are contested. A nation that has not filed, or whose filings are lapsing, cannot simply decide to operate a GEO satellite at a useful slot. This is not a technical problem; it is a regulatory and diplomatic one, and it has a long lead time regardless of which procurement route is chosen. Starting the filing process is the first concrete action available to any government considering sovereignty, at any budget level.

Third: what institutional capacity exists to operate? A sovereign satellite without trained operators is a liability, not an asset. Handover to a national team is not a formality; it is a multi-year process that should be contracted explicitly, with milestones, before the programme begins. The countries that have struggled with sovereign space assets have generally underinvested in this phase, not in the hardware.

What you end up owning, and what you do not

Ownership varies by route, and the distinctions matter in a crisis. A leased-capacity arrangement transfers no assets. A sovereign gateway transfers ground hardware, software licences (with source-access terms negotiated before signature), spectrum filings and trained staff. A hosted payload transfers the payload hardware and its associated spectrum rights; the bus, the orbital slot and the launch insurance belong to the host. A fully sovereign satellite programme, structured correctly, transfers spacecraft, ground stations, mission-control systems, spectrum filings and an operational team capable of running the system without external assistance.

The limits of each option are equally concrete. Ground infrastructure can be seized, damaged or legislated against by the host government. A single GEO satellite is a single point of failure; a debris-generating collision or a transponder failure has no on-orbit redundancy unless a second satellite exists. A LEO constellation requires continuous operations, regular re-contact with each spacecraft, and eventually re-launch as satellites reach end of life, typically five to seven years for small LEO platforms. Sovereignty does not mean invulnerability. It means the decisions about resilience are yours to make.

What this mission is built from

Geostationary orbit: Defines the orbital mechanics, coverage geometry and spectrum-coordination requirements for GEO and hosted-payload options.
Communications payloads: Specifies the transponder, antenna and frequency-band architecture that determines capacity, coverage and interference margins for each route.
Ka-band high-rate stations: Provides the ground-segment gateway infrastructure that a sovereign-gateway programme owns and operates, independent of which space segment it connects to.
Export control navigation: Identifies the export-control and technology-transfer constraints that govern which hardware, software and operational systems can be transferred to national ownership under each route.

What you end up owning

ITU spectrum filings and orbital-slot coordination records, held in the national administration's name
Gateway earth-station hardware and network-management systems, with source-access terms agreed before contract signature
Spacecraft and payload hardware (for hosted-payload and fully sovereign routes), with hardware audit rights throughout the programme
Mission-control software and ground-station facilities, handed over to national teams at agreed programme milestones
Trained national operators capable of conducting routine mission operations without external assistance
Contractual audit rights over all third-party components integrated into the programme

Handover is structured as a staged process with explicit milestones: shadow operations, then supervised national operations, then independent national operations, each with a defined acceptance criterion. Source-access terms and hardware audit rights are agreed before contract signature, not negotiated after delivery. What remains with Satellize or integration partners after handover is the option of ongoing technical support; that relationship is the customer's choice to continue or end.

Programme parameters

Sovereign gateway (ground-only route)	1 to 3 earth stations, ITU filing, network-management system; timeline 12 to 24 months
Hosted payload route	1 payload aboard a third-party GEO bus; timeline 2 to 4 years from contract to in-orbit; lifetime governed by host spacecraft
Dedicated GEO satellite (small-to-medium)	1 spacecraft, 1 to 3 ground stations, 15-year design life; programme timeline 4 to 7 years for first-time operators
Small national LEO constellation	6 to 30 spacecraft depending on coverage and capacity targets; satellite design life typically 5 to 7 years; requires re-launch planning from programme outset
Spectrum filing lead time	ITU coordination typically 3 to 7 years for contested GEO slots; advance filing is independent of spacecraft procurement and should begin first
Operator team size at handover	Typically 8 to 25 trained national staff for a single-satellite programme; scales with constellation size and ground-segment complexity
Training and handover phase	18 to 36 months of structured shadow and supervised operations before fully independent national control
Cost class reference precedents	Sovereign gateway: low single-digit millions; hosted payload: tens of millions; dedicated GEO: low-to-mid hundreds of millions (publicly reported Asia-Pacific and Middle East programmes)

Sources

One contract, one accountable engineer

Commissioned as one programme, not a stack of contracts: spacecraft, launch, ground segment, mission control, training and handover are priced together. Source-access terms and audit rights are agreed in writing before signature. Request a route-comparison workshop.