Survey claims power cuts are top cause as “broadband outages” hit half of UK

A new Uswitch commissioned Opinium survey of 2,000 UK adults, which was conducted between 29th September to 2nd October 2026, has claimed that 26.7 million people suffered a “broadband outage” over the past year, with power cuts said to be overtaking ISP faults as the number one cause. But 4.6m (32%) are apparently unaware that power cuts can also stop digital phones working.

The above figures seem to be a rather bold extrapolation from a fairly small survey sample – for context, 48% of respondents had experience at least one outage in the past 12 months (up 19% on the prior year). The survey goes on to claim that 66% experienced outages lasting for 3 hours or more and 20% experienced them once a week or more often.

Among those hit by an outage, 42% say they lost their connection to a power cut, while 31% blamed provider faults and 31% said it stemmed from a router problem. Homes in rural areas, which often lack access to modern full fibre broadband networks, were also found to be the hardest hit with 63% of those affected suffering an outage due to a power cut, compared with 28% in urban areas.

Perhaps unsurprisingly, 7% lost their broadband connection to a storm, flood or severe weather in the past year (i.e. power cuts are much more likely to occur during storms), rising to 12% in rural areas, and 43% say they’re worried that such events could knock out their connection in the winter months ahead.

The survey notes that modern digital phone services are also dependent upon your broadband connectivity (handsets typically plug into the back of your router), which means they too can suffer outages as a result of a power cut. In the past analogue phone lines (PSTN/WLR) were often powered from the exchange, so those with old analogue phones could still use their handset during a power cut.

In fairness, it’s worth remembering that power cuts can also hit exchanges (i.e. knocking out analogue phones too) and many people use DECT (wireless) handsets on those lines, which would also be knocked out during a local power cut (assuming you don’t have an old analogue handset to spare for such events).

The survey notes that 26% of respondents instead have a modern digital landline that runs through their broadband router, yet 32% still don’t know that this can also stop working during a power cut, which is a growing issue because the old analogue phone services are due to be switched off on 31st January 2027 (see details of what will happen).

The good news is that most of the major broadband providers will provide a free battery backup unit to vulnerable users (regular users can get one too, at a cost), which will usually last for several hours and then retain some charge for an emergency call outside of that – assuming it works. Smaller providers often do this too, although the kit they offer can vary.

Finally, the survey notes that only 16% of those hit by an outage received compensation from their broadband provider, which is down from 22% last year. But 57% didn’t know they could even ask for this, although not every outage qualifies.

Ofcom’s related Automatic Compensation scheme only applies to signatory ISPs (i.e. most of the major players) and qualifying outages must last for longer than two working days, as well as be the fault of your broadband provider (i.e. power cuts usually don’t qualify). In addition, the customer should notify their ISP of any protracted outages as soon as possible.

Londoners were the most likely to have lost their connection, with three in five (60%) affected, up from just over half (53%) in 2025, followed by the South West (56%) and the West Midlands (51%). But frankly the survey sample is far too small to be able to extract any credible regional split.

Naturally, we have a few additional comments on all this, not least of which is the fact that we haven’t actually seen any hard independent data evidence to corroborate the above findings (in fairness, it’s a difficult thing to study with any accuracy). Opinion surveys are of course also best taken with a pinch of salt, since they often reflect a perception more than reality.

In addition, full fibre (FTTP) networks with greater reliability have grown in both coverage and take-up in recent years, while more people are now back working from an office (i.e. fewer people are at home to spot when outages occur). Not to mention that it may be considered a bit misleading of the survey to term the overall result as a “broadband outage”, particularly when they’re also including power cuts as a cause (something end-users could mitigate with their own backups, if so desired).

The other issue typically stems from whether a “broadband outage” is actually caused by your ISP or is instead an issue within the home, such as a local network, viruses / hacker, router, Wi-Fi or end-user device problem. In fairness, this survey does at least reflect when consumers have at least identified the cause as their router, although not everybody will be able to make such an accurate determination (i.e. many will just blame the ISP).

Lest we forget that modern broadband networks can be disrupted in all sorts of different ways, such as via weather damage (flooding, fallen trees etc.), third-party street works cutting through cables, fires, power cuts at an exchange / data centre and deeper faults within an ISP’s network (e.g. hardware failure, routing / peering or DNS mistakes etc.). Most of these are quickly resolved, but others can take days or even longer (complex incidents in remote rural areas are usually slower to resolve).