A constellation operator planning ground infrastructure tends to budget carefully for antennas and modems, and to underestimate everything else by a considerable margin. The equipment is a procurement exercise. The site is a project — and in some geographies it is a multi-year one with a meaningful chance of failure.
What actually takes the time
Site and permissions. Land with clear sky and stable access, in a location that is useful for your orbits, with the regulatory approvals to operate radio equipment on it. In an unfamiliar jurisdiction this is where most of the schedule goes, and it is not a problem that yields to spending more money.
Power. Not a connection — reliable power, which in much of the region means generation, storage, transfer arrangements and fuel logistics. A ground station that goes down when the grid does is not a ground station.
Backhaul. Data arriving from orbit has to go somewhere. In many candidate locations the achievable backhaul is a constraint that shapes what the station can usefully do — there is no point downlinking faster than you can move it onward.
Staffing. People who can be on shift at the site, who will stay, and who can act rather than report. Recruiting and retaining that capability in a new country is a years-long undertaking.
Against those four, the antenna is the straightforward item.
Why the Horn of Africa is worth a station
Two reasons, one geographic and one infrastructural.
Mogadishu sits close to the equator, which for certain orbits produces useful pass geometry that higher-latitude stations do not offer. Coverage of this longitude band is also comparatively thin, and a gap in a ground network is a gap in a constellation’s ability to deliver data promptly.
The infrastructural reason is more specific. Our teleport sits on the same compound as a submarine cable landing station and a data centre. That means downlinked data can be processed and stored metres from the antenna, and moved onward over subsea capacity without crossing a leased link first. Ground stations frequently end up with excellent downlink and a bottleneck immediately behind it; here the bottleneck is not in that position.
What hosting gets you, and what it costs you
The honest comparison has a column on each side.
What you get: a site, power, backhaul and staffing that already exist and already operate. Processing and storage adjacent to the antenna. A partner who knows the regulator and the practicalities of working here. A schedule measured in months rather than years.
What you give up: it is our site and our footprint, not yours. You hold a commercial relationship rather than an asset. If your model depends on owning ground infrastructure outright, or on a location we do not have, hosting is the wrong answer and we would rather say so early.
Data residency, which comes up more than expected
An increasing number of Earth observation and communications programmes carry conditions about where data may be held — imposed by a customer, a funder, or the country being observed. A station co-located with an in-country data centre can satisfy those conditions without a separate arrangement, which is a smaller point commercially and a significant one contractually.
What to send us
The orbit, the band, the antenna footprint, the data volumes per pass, and what the passes have to achieve. We will come back on whether this site helps your coverage, what we can host today, what would need building, and the cases where the answer is that we are not the right location.
That last one is a real answer and we give it. A hosting arrangement that does not fit is worse for us than no arrangement, because ground stations are a long relationship and a bad one is visible for years.
