
Network analyst firm Streetwave have published the results from a new study they’ve just conducted, which examined the mobile network coverage and mobile broadband (4G, 5G) performance of EE, Vodafone, Three UK and O2 across the entire M4 motorway from Cardiff to London – a 127-minute journey.
The live and admittedly somewhat anecdotal study, which used Streetwave’s portable network measurement and monitoring kit placed in a Toyota Aygo on 3rd August 2026, found that “most” of the mobile operators delivered high levels of ‘Basic Coverage‘ along the M4 motorway (see below for the definition). The M4 is the third-longest motorway in the United Kingdom, running from West London to Southwest Wales
Streetwave typically defines Basic Coverage as reflecting locations where the mobile network provides users with data speeds of above 1Mbps download, 0.5Mbps upload, and below 100ms (milliseconds) of latency (i.e. supporting only the most basic of use cases or needs).
Advertisement
Overall EE delivered the strongest result and were closely followed by Vodafone and Three UK, while O2 were by far the poorest. Most of the poor coverage areas for the latter came in multiple smaller segments as opposed to long stretches – particularly between Swindon and Reading.
M4 Motorway – Mobile Basic Coverage Scores
Time each network fell below the ‘Basic Coverage’ threshold
➤ EE – 4 minutes
➤ Vodafone – 6 minutes
➤ Three UK – 9 minutes
➤ O2 – 28 minutes
Such figures are useful for business travellers who may need to make calls or passengers who want to use the mobile broadband connectivity on their devices. But in that respect it would have been nice to see the ‘Good Coverage’ scores for each operator too, which set a higher performance bar for data connectivity.
Advertisement
so they didn’t cover cardiff to Pont Abraham
Wow the difference between the EE, Voda, Three and then O2 is wild!
They are lagging so far behind right now in so many places.
Oh Wow – at last a results that is realistic to the great majority of users! Results from a their test rig in a car not a truck so we can conclude the antennas will be at the average users height above ground level. Not surprised about the EE result (as ESN coverage is by EE). Not sure if I’m surprised or disappointed by the O2 result as they do seem to be getting worse in rural areas – is that due to 3G switch-off or loosing sites?
Even better would be for everyone to measure it themselves, passively, which would give a much better and wider range of results at different times and congestion levels.
ROAM are doing this – as a decentralised mobile coverage mapper. You can try it out here: https://r.roam.network/SaolaBadlandsFicus
The https:// roam.network website seems to currently be parked by GoDaddy…
I hope that no lives where put at risk during these calls? That I see many cars where I live only have one person in them, so even hands free would compromise the Driver’s attention.
The testing kit does everything by itself, without the driving needing to touch it.
ISPreview has covered Streetwave before, the article about Northern Ireland’s coverage, gives full insight on how it works.
In Northern Ireland it was placed in the regions dust carts.
Believe 3’s results are heavily influenced by VF site merges in the area by MOCN. Usually there last place or second last but not too much better off than O2