Almost nobody in Africa uses 5G, yet they are already eating a big share of the data.

Kenya's Communications Authority reported that as of June 2026, the country had 2.1 million 5G connections out of 64.26 million mobile data connections in total — 3.27% of the market. Those connections generated 135.38 million gigabytes of traffic in a single quarter, up from 43.86 million gigabytes a year earlier, an increase of 208%.

The regulator's per-user figures show why. A 5G subscriber gets through roughly four times the data of a 4G subscriber and nearly eight times that of someone on 3G. 4G still dominates the connection count at 48.31 million, about 75% of the market.

Nigeria's picture is less granular but moving in the same direction. As of July 2026, 5G represented 4.74% of the country's 195.1 million telecom subscribers, while total monthly data consumption hit 1.67 million terabytes — the highest on record for the period tracked.

In Simple Terms

A small number of people on the newest network are using a disproportionate amount of the internet.

That sounds like a technical footnote. It isn't. Mobile networks are built to handle peak traffic, not average customers, so the cost of running one is driven by how much data moves through it — not by how many SIM cards are active. A network where 3% of users generate 13% of the traffic needs building differently from one where usage is spread evenly.

It also means the number everyone quotes — 5G adoption rates — is close to useless as a planning figure.

The Money

Two kinds of money are moving here.

The first is what operators are spending. MTN Nigeria recorded ₦620.5 billion of capital expenditure in the first half of 2026. Airtel reported $211 million, about ₦287.6 billion, on 4G and 5G tower deployment — implying an exchange rate of roughly ₦1,363 to the dollar. (Derived.) These two figures are not measuring the same thing: MTN's is total network capital spending, Airtel's is tower deployment for two specific technologies. Adding them would overstate what either is putting into 5G.

Both are spending on small cells and mini-antennas in high-traffic areas rather than only large masts — the signature of a capacity problem in specific places, not a coverage problem across a map. MTN has separately proposed acquiring Mafab Communications' 5G spectrum in the 2.1 GHz and 3.5 GHz bands, which would give it more room to carry traffic. That transaction is proposed, not completed.

The second is what consumers spend. A subscriber using four times the data is a subscriber buying considerably more bundles. That is the commercial logic behind the build-out: revenue per user rising faster than user numbers.

Follow the Money

Who this reaches. Beyond the reported figures, this section is Rainingmoney's reading.

Operators gain the most direct benefit: higher bundle revenue per subscriber, without the cost of acquiring four separate customers.

Consumers on older networks carry a hidden risk. If capacity is absorbed by heavy users faster than it is added, congestion shows up as slower connections for everyone — including people who never bought a 5G phone.

Data centres, fibre companies and tower firms sit downstream of the same demand. Traffic that starts at a mast has to travel somewhere, and bandwidth-hungry applications push work onto all of it.

Power suppliers and diesel budgets. Denser infrastructure in high-traffic areas means more sites to keep running in a country where reliable grid power cannot be assumed.

Investors are funding capacity now against consumption revenue later, with limited published data to size the gap between the two.

The Bigger Picture

Africa's first mobile internet story was about access — getting people connected at all. That framing shaped a decade of policy, investment and coverage targets.

The measure that matters is shifting from connection counts to consumption volumes. On that measure, growth is not gradual. It arrives in steps, because each user who moves to a faster network does not consume slightly more — they consume several times more, largely because software quietly stops rationing itself once the bottleneck disappears.

Adoption across the continent remains uneven. Ghana is running a competitive 5G spectrum licensing process. Elsewhere, spectrum availability, infrastructure, financing and coverage gaps constrain rollout, alongside vandalism, network outages and the price of compatible handsets. The result is a paradox: the technology is out of reach for most people, while the minority who have it already reshape traffic patterns.

The Jargon

Backhaul — the connection carrying traffic from a mast back into the core network, usually fibre. A fast mast on thin backhaul is a bottleneck with an antenna on top.

Small cell — a low-power mini base station added in a crowded area to increase capacity in one spot rather than extend coverage.

Fixed wireless access — a router that uses a mobile network to deliver home or office broadband instead of a cable.

Spectrum — the radio frequencies an operator is licensed to use. More spectrum means more capacity to carry traffic.

Terabyte — 1,000 gigabytes. Nigeria's 1.67 million terabytes a month is about 1.67 billion gigabytes.

What Happens Next

Known. Kenya's regulator publishes the generation-level breakdown; Nigeria's does not. MTN and Airtel are both expanding capacity in high-traffic areas. Ghana's licensing process is under way.

Possible. Nigerian monthly consumption could pass 2 million terabytes before the end of 2026 — a projection reported without a named forecaster attached. If Kenya's usage pattern repeats in larger markets, capacity requirements will outrun subscriber-based forecasts.

Unknown. How much of Nigeria's consumption growth is actually 5G, since the breakdown is not published. How much of MTN's capital spending is 5G-specific. Whether the Mafab spectrum acquisition is approved. And whether backhaul, data centre and power capacity expand quickly enough to keep the experience intact.

Rainingmoney's Take

Analysis, not reporting.

The most useful thing in these figures is a correction to how the industry talks about itself. Press releases count subscribers because subscribers are easy to count. The Kenyan data says that is the wrong unit — a network's cost is set by gigabytes, and gigabytes are concentrated in a small, fast-growing group.

For Nigerian readers the practical consequence is straightforward. The build-out being funded now is not really about extending 5G to you. It is about absorbing the traffic generated by people who already have it, so the network does not degrade for everyone else. That is a less exciting story than a coverage map, and a more honest one.

The gap worth closing is disclosure. Kenya publishes traffic by generation and the picture is immediately legible. Nigeria does not, which leaves regulators, operators and investors reasoning from a correlation the reporting itself declines to call causation. Until that breakdown exists, confident claims about what 5G is doing to Nigerian data demand should be read as informed inference rather than measurement.

This article is for information only and is not investment advice.