Who gets left behind by the UK's emergency alerts?

The UK's Emergency Alerts system is no longer just a concept or something most people associate with national tests. It has been used to warn communities about immediate threats, from severe flooding and storms to the red wind warnings issued across Cornwall and the Isles of Scilly earlier this year. In each case, the technology allowed the government and emergency services to send urgent advice directly to people in affected areas, including instructions to stay indoors, avoid travel or evacuate.

These alerts demonstrate how technology can reach large numbers of people quickly when lives may be at risk. But in a crisis, sending a message is only the first step. An alert can arrive on somebody's phone without reaching them in a way they can understand or act upon.

What happens if English is not their first language? What if they do not own a compatible mobile phone, need information in a different format or require support to leave their home? What if they are staying in temporary accommodation and do not know the area? Or living in a refuge, where the sound of an alert on a concealed phone could put them at risk?

Delivering information is not the same as communicating it effectively.

Reaching devices is not the same as reaching people

The Emergency Alerts system broadcasts messages to compatible devices connected to the UK's 4G and 5G networks. It does not require people to register, share personal details or download an app, removing some of the barriers associated with conventional notification systems.

However, people whose phones are switched off, connected only to 2G or 3G, incompatible with the system or set to aeroplane mode may not receive an alert. Those without a mobile device will not receive it at all.

Device compatibility is only one potential barrier. Once an alert has arrived, the recipient must be able to perceive it, understand the information, decide what it means for them and follow the instructions.

Technology often treats delivery as the end point. A message has been sent, a notification has appeared and the system has performed as intended. Human-centred design asks a different question: what happened next?

If someone is deafblind, the combination of sound, vibration and on-screen text may still not meet their needs. If a person has a cognitive impairment or does not speak English as their first language, an unfamiliar official message may be difficult to interpret under pressure.

For someone who relies on a carer, support worker or accessible transport, an instruction to evacuate is not something they can necessarily carry out independently. Someone who is socially isolated may have nobody nearby to help them understand what is happening or take action.

An alert is only effective if the person receiving it can understand what it means and act on it. That's why emergency communication has to offer more than one route.

Accessible communication needs more than one channel

No single technology will reach everyone. That does not make the national alert system ineffective; it means it should be treated as one part of a broader emergency communications network.

Critical information should be distributed through several complementary channels, including television, radio, accessible websites, landlines, public address systems and direct communication through local services and community organisations.

The message must also be accessible. Information should be written in concise, plain language, with a clear explanation of the danger and the action people need to take. Supporting information should be available in commonly used community languages and formats such as British Sign Language, Easy Read, audio and screen reader-compatible text.

This should not be seen as unnecessary duplication. In an emergency, relying on one channel risks excluding anyone who cannot access it. Providing several routes to the same essential information makes the overall system more resilient.

Trusted intermediaries also matter. Local authorities, housing providers, health and care services, charities, community groups and refuge staff may understand the needs of particular residents better than a national system ever could. They can help interpret warnings, identify people who need assistance and help people understand what the instructions mean for them locally.

Accessibility includes safety and privacy

Inclusive design is not just about making information easier to read or hear. It also means understanding the circumstances in which technology is being used.

For someone experiencing domestic abuse, a concealed mobile phone may provide a private way to seek help. An emergency alert sounding loudly, including when the device is set to silent, could reveal its existence to an abuser.

People can switch off alerts or turn off a hidden phone, but that creates a difficult trade-off as protecting their immediate safety may mean losing access to future emergency warnings.

There may not be a perfect technical solution to every conflict of this kind. However, services can work with specialist organisations, communicate potential risks before planned tests and provide discreet guidance alongside alternative ways to access information.

Voluntary registers can help emergency services and local organisations identify people who may need additional support, but they also have limitations. Records become outdated, people move between temporary homes and some individuals may be unable or understandably unwilling to disclose their circumstances or location.

Nobody's safety should depend entirely on whether they completed the right form before an emergency happened.

Design for the response, not just the alert

Emergency technology must be tested with the people most likely to encounter barriers, including disabled and older people, survivors of abuse, people who speak different languages, those without reliable digital access and people living in temporary or insecure accommodation.

Testing should look beyond whether a device received the message. Did the person notice the alert? Could they access and understand it? Did they recognise it as genuine? Did they know what to do next? Crucially, were they able to do it safely?

Those questions should be asked while systems and emergency plans are being designed, not once a crisis is already unfolding. Co-designing services with affected communities can reveal practical barriers that technical testing alone will miss.

The UK's Emergency Alerts system is an important public safety tool, but its success cannot be measured purely by the number of messages delivered.

The true test is whether people know what is happening, understand what they need to do and have a safe and realistic way to respond. If our crisis communications do not work for those facing the greatest barriers, they risk leaving the people most in need of support in the greatest danger.