Technologies
5G and getting online
The measured story of 5G is capacity growing faster than anyone uses it, while 2.2 billion people remain offline for reasons that have little to do with coverage.
The best-measured fact about 5G in a mature market is that its capacity is growing much faster than the demand for it. Ofcom found UK average mobile download speeds up 30% year on year in 2025 while average data usage rose only 10%, and 5G carrying 28% of mobile traffic against 4G's 72% after six years of deployment. Globally the constraint is not the newest network either. About 6 billion people were online in 2025 and 2.2 billion were not, but only 4% of the world's population lives outside mobile broadband coverage — the far larger group is covered and does not use it. Where 5G reaches is largely a function of national income: 84% population coverage in high-income countries against 4% in low-income ones. For a practical example of software used to turn activity into operational metrics, see this page.
Most of the people offline are already covered
The headline connectivity numbers come from the International Telecommunication Union, a United Nations agency and the only body that attempts a global count.
About 6 billion people, roughly three-quarters of the world's population, were online in 2025, with 2.2 billion still offline and around 240 million coming online during the year.
Direction: Increase. Strength of evidence: Strong.
Caveat Modelled from national data of uneven quality, and "internet user" means any use in the last three months, which flatters the figure.
The composition of the offline 2.2 billion matters more than the total, and the standard citation for it comes from the mobile industry's own trade association — which has an interest in framing the gap as a demand problem rather than an investment one.
4% of the world's population lived outside mobile broadband coverage in 2024 while 38% lived under coverage but did not use mobile internet, down from 19% and 49% in 2015.
Direction: Decrease. Strength of evidence: Mixed.
Caveat GSMA is the mobile operators' trade association; the estimates derive from operator data plus GSMA's own consumer survey.
Even discounting for the source, the ratio is roughly ten to one. Building more network is not what closes most of that gap. Cost is a large part of what remains: an entry-level data-only mobile broadband package was still unaffordable in around 60% of low- and middle-income countries in 2025, measured against a target of 2% of income per head (ITU, 2025). The rest is documented in who is still offline and why.
5G coverage tracks national income
Coverage figures are supplied by operators and describe where a signal is theoretically available, not where anyone gets a usable connection. On that basis the technology has spread quickly and very unevenly.
55% of the world's population was covered by a 5G network in 2025, but coverage was 84% in high-income countries against 4% in low-income countries.
Direction: Cuts both ways. Strength of evidence: Strong.
Caveat Operator-reported population coverage rather than measured signal availability, and it says nothing about whether covered people have a 5G device or subscription.
A twenty-one-fold gap between rich and poor countries in a technology marketed as universal is the finding. The same report puts 85% of urban dwellers online against 58% of rural dwellers, and 94% in high-income countries against 23% in low-income ones (ITU, 2025).
In the UK, capacity is growing three times faster than use
The most useful single measurement in this subject is a comparison Ofcom publishes almost in passing. It sets the growth in what networks can deliver against the growth in what people actually consume.
UK average mobile download speeds rose 30% year on year to 285 Mbit/s while average data usage rose only 10%, and 5G carried 28% of mobile traffic against 4G's 72%.
Direction: Increase. Strength of evidence: Strong.
Caveat Speeds come from operator network data rather than independent user-side measurement, and the traffic split shows most usage still runs on 4G after years of 5G marketing.
Two things follow. First, the applications that were supposed to consume 5G capacity have not arrived in volume: usage is growing, but at a third of the rate capacity is. Second, most of what people do on a phone still runs over 4G, which is the practical answer to whether a 5G handset changes daily experience. Ofcom also reports 5G standalone — the version that is not built on 4G core infrastructure — reaching 83% of the UK population at high confidence in the same report (Ofcom, 2025).
Availability is not take-up
The fixed-line picture shows the same gap between what exists and what is used, and it is measured more precisely because Ofcom works from premises-level records.
87% of UK residential premises had gigabit-capable coverage and 78% full fibre by late 2025, but only 42% of premises where full fibre was available had taken it up, and about 44,000 premises still could not get decent broadband.
Direction: Cuts both ways. Strength of evidence: Strong.
Caveat Coverage is derived from operator records rather than end-user speed tests, so it measures availability rather than delivered performance.
Fewer than half the households that could have fibre have bought it. That is not a coverage failure, and it is the domestic version of the global usage gap.
Subscription totals come from the companies selling the equipment
The most-quoted global 5G figure is published by a vendor.
Global 5G subscriptions reached about 3 billion by mid-2026.
Direction: Increase. Strength of evidence: Mixed.
Caveat Ericsson sells 5G equipment; subscriptions count SIM cards rather than people, and holding a 5G subscription does not imply using a 5G network.
Set against Ofcom's traffic split, a subscription count is a poor measure of use. The gap between the two is the same one that runs through counting connected devices, where the market sizing is done by firms selling into the market.
The short version
- UK mobile speeds rose 30% in a year to 285 Mbit/s while data use rose 10%, and 5G carried 28% of traffic against 4G's 72% (Ofcom, 2025).
- About 6 billion people were online in 2025 and 2.2 billion were not, on the ITU's modelled estimate.
- Only 4% of the world's population lives outside mobile broadband coverage, against 38% who are covered and do not use it (GSMA, 2025).
- 5G population coverage was 84% in high-income countries and 4% in low-income countries in 2025 (ITU, 2025).
- Nobody publishes independent user-side speed measurement in either the UK or the US, so claims about what 5G delivers in practice rest on self-selected commercial panels.