A mobile signal feels weightless, but it stands on concrete, steel and secured land. On 6 October 2022, Safaricom Ethiopia began commercial service, and behind the invisible network is a very physical build: towers, fibre routes, power, and the land and permits that decide where any of it can go. For the property and engineering reader, the launch is a construction story wearing a telecom badge.

The Footprint: A network is a distributed real-asset portfolio

A second national mobile network entering a market previously served only by the state incumbent, Ethio Telecom, is, in physical terms, a large distributed infrastructure programme. It requires thousands of base-station sites, backhaul fibre along roads and corridors, power connections or on-site generation, and retail premises in the cities. Each site is a real-asset decision — land access, a lease or wayleave, a foundation, and long-term maintenance. The coverage race that begins today is therefore also a race to secure and build sites, under terms and coverage obligations administered by the Ethiopian Communications Authority. In practical terms, the coverage promises a customer hears on launch day are a forward schedule of civil works, land agreements and equipment installs that must still be executed on the ground.

The takeaway: a coverage map is really a construction and land-access programme in disguise.

The Delivery Constraints: Land, permits and engineering capacity

The local tension for delivery is practical: land, permits, engineering capacity, compensation and maintenance. Acquiring or leasing site rights, navigating municipal permitting in Addis Ababa and Dire Dawa, and mobilising enough qualified civil and telecom-engineering contractors all sit on the critical path between a licence and a live signal. Where compensation for land or wayleaves is contested, or where skilled installation and maintenance capacity is thin, the coverage timetable slips regardless of capital.

This is the ordinary friction of building physical infrastructure at national scale, and it is reportable precisely because it changes execution. A network’s advertised reach means little until the sites beneath it are secured, powered and maintained. Reliable power is its own constraint: where grid supply is intermittent, sites depend on generation and fuel logistics that add cost and complexity to every location.

The takeaway: the binding constraint on coverage is rarely money; it is land, permits and engineering capacity.

The Ownership Question: Who holds and maintains the asset

Infrastructure sharing is named as one of the competitive fronts ahead, and it is fundamentally a property question. Whether an operator owns its towers, leases them from a specialist tower company, or shares sites with the incumbent determines who holds the asset, who maintains it, and how efficiently capital is used. Shared passive infrastructure — towers, power, fibre ducts — can lower the cost of coverage and reprice the economics of building in less dense areas.

The pattern also reshapes locations. Sites, exchange points and retail footprints concentrate value where the network densifies first, which in Ethiopia means the primary cities and the corridors between them before the rural periphery. A tower or exchange site that sits at a natural aggregation point can carry traffic for more than one operator, which is precisely what makes shared passive infrastructure a valuable, long-lived asset rather than a cost.

The takeaway: how the passive infrastructure is owned and shared decides both the cost of coverage and where value concentrates.

So what

For a regional operator, contractor or property investor, the decision implication is to read this launch as a pipeline of real-asset work rather than a consumer event. The bankable opportunities sit in site acquisition, civil works, power, tower ownership and maintenance — the physical layer that any coverage promise depends on. The locations that reprice are those where the network builds first. The question to carry forward is not how fast the brand grows, but how fast the sites go up, because in telecom infrastructure the concrete always leads the signal.