There’s no such thing as a universal digital service

17 Oct

In a world where Meta has more users than most countries have residents, it seems odd to say that digital services aren’t universal – but universal services need to work for everyone, not just for people who are digitally connected.

Using the recent UK government announcement of digital ID as a jumping-off point, this blog post explores why universal services need to be much, much more than digital-by-default.

The founding principles of the NHS do as good a job as any in outlining what counts as a universal service. Written by Aneurin Bevan in 1948, they state that the NHS should:

Meet the needs of everyone;
Be free at the point of delivery; and
Be based on clinical need, not the ability to pay

Setting aside the third principle on Bevin’s list as specific to health, the first two feel like a good test for universality. Is it possible for a digital-by-default service to meet the needs of everyone and be free at the point of delivery?

Digital by Default

The concept of “digital by default” is now taken for granted in UK public service delivery, but as plans for a mandatory digital ID service for right-to-work checks are announced, it is important to remember that a digital service is not a universal one. Digital is still just one channel. Face-to-face events and conversations, telephone, SMS, and mail are all still a part of the delivery mix, while dependence on digital devices is being more routinely raised as having negative effects on both societal connections and individual’s mental health. In fact, there are more and more indicators that people in their 20s are starting to log off. In the rest of life, digital is an option, and it’s not always optimal.

If we’re going to linger over semantics here (and why not) it’s also worth saying that government’s use of the word mandatory to describe digital ID creates some additional challenges. Mandatory is not the same as universal. “Mandatory” means you must, while “universal” implies you can – the first word conveys a responsibility, the second denotes a freedom. As such, something that must be used by everyone needs to be extremely accessible: in a fair and just society, it should have no barriers to entry.

“Do the hard work for the public”

Although it may feel that many things we see and do online are free, the reality is that all digital interactions come at a price. I don’t mean this in the metaphorical “If you’re not paying, you’re the product” sense, but practically: having access to an up-to-date personal digital device with reliable connectivity costs money.

If you’re relatively affluent, you might not notice the cost of your smartphone bill and your superfast broadband connection might feel non-negotiable. If you have an office job that provides you with a modern laptop and a reliable Internet connection, then those things might seem like basic necessities, but for many households digital connectivity is expensive. Research by Promising Trouble, the social enterprise I used to run, found that in 2023 the poorest households were spending a massive 5% of their disposable income on broadband, and that’s before you even get to devices.

However, the Blueprint for Digital Government, published earlier this year by the Government Digital Service (GDS), starts from the assumption that digital services are the most accessible option – offering a vision of digital government in the UK that delivers:

transparent, next-generation public services that do the hard work for the public, can be accessed and used by everyone who needs them, and are designed around the user.

Recent figures from the Digital Poverty Alliance figures show that between 13 and 19 million people over 16 in the UK live with some kind of digital poverty, meaning that being digitally included – having a usable personal device and a reliable, secure connection – is not affordable. People who are unemployed are also two to three times more likely to experience digital poverty than people in employment. Research by the Good Things Foundation shows that 7.9m people lack basic digital skills and 21m cannot complete basic digital tasks for work. The intersections of these groups are not clear, but even when bundled together they do not represent a trivial number of people, and are a significant proportion of the 43m people of working age in the UK. These inequalities will not be remedied with the announcement of a single scheme or the redistribution of a few devices; they are symptoms of a society that has failed for decades to include some of the most vulnerable people in social and economic progress, and those opportunity gaps are now entrenched as a feature of modern life. Digital exclusion is not a bug that can be fixed with a few million quid, but a feature of the standard operating model of a digital economy driven by market forces.

The government’s Blueprint also assumes that convenience for more people means more digital personalisation and joined-up digital service delivery, but in reality convenience is also often about choice. Letting people choose the best way for them to use or access a service at the point of need is enabling convenience for the public; mandating a single digital delivery channel is enabling convenience for government . And while the Blueprint does commit to “ensuring as many people as possible can access public services digitally” and promises to “support the digitally excluded”, it is unclear how exactly that support translates into real life.

Meet the needs of everyone

The minimum viable interpretation of meeting the needs of everyone is a service that:

is easy for everyone to access and use
does not deepen existing inequalities
does not introduce new societal or economic harms

When a service is mandatory, it should should also:

be clearly and effectively limited in purpose

Purpose limitation is important because any mandatory service has implications for the rights and freedoms of the people mandated to use it. As such, the potential for the scope of any mandatory service to increase over time – particularly in shifting social, political and economic contexts – should be understood at deployment, and controls should be put in place to ensure that any increase or expansion is rights respecting, effective and proportionate.

I would also argue that any digital by default service that builds infrastructural capabilities – including the ability to collect and process public data – should also be forward-looking by default, with an understanding of both the future capabilities being created and the future maintenance needs. Increasing use of AI in public-service delivery makes this essential: understanding how even incremental changes in technological capabilities could make a significant difference to the potential uses of a mandatory or universal service is now a minimum capability for due diligence.

Not deepening existing inequalities

“Not deepening existing inequalities” is a fancy way of saying “not make things worse for people”. The very minimum a universal public service deliver is an improvement, but it cannot be taken for granted that the roll out of digital service will lead to improvements for everyone. As well as creating better experiences for some groups, it is also likely that a switch to digital-by-default will worsen some existing inequalities and also create new ones.

Each product or service will have its own segmentation of people’s needs, but at the very minimum there should be a distinction between the fact that:

For people who are already digitally confident a well-designed, reliable digital service should create a better experience, including increased convenience where appropriate
For people with low or growing digital confidence a good onboarding experience is essential, with additional support available as needed. Ideally this onboarding should move at the speed of trust, rather than in an expedited timeframe.
For people who are digitally excluded, appropriate and accessible non-digital means of using or engaging with the service should be easily available.

Beyond that, the specific impacts of digital delivery on particular communities should be worked through and mitigated. If you start by designing for the affluent, able-bodied majority then exclusion is being built in from the start. It is not only more equitable but more cost-effective and efficient to start by understanding the impact a service will have on people with the highest needs or the most critical dependencies and then adapting it for the majority.

It should also be the case that service delivery obligations are taken seriously, and that any digital-by-default service should be supported by realistic assessments about its availability and up-time, be subject to robust testing and have failover mechanisms in place to ensure all products are fit for purpose. Much of this is covered by the 2014 digital by default service standard but it cannot be taken for granted in the delivery of essential services. For instance, the recent delivery of eVisas has shown that fixed deadlines are problematic when critical services are rolled out when they are simply not ready to be used.

Meet the needs of the furthest first and leave no one behind

In a commercial setting, or for a specific service where the needs of users are well understood, digital-by-default might be the answer. There are circumstances where it is admissible to make trade offs, to deliver something that makes things better for some people and worse for others, or where suitable mitigations can be put in place to alleviate any new deficits. But digital-by-default is not a practical, equitable solution to either a universal or a mandatory service. For universal services, the starting point should be to meet the needs of the most excluded and vulnerable people first, and then work towards mainstream adoption – doing it the other way round builds injustice and exclusion in from the start, and commits you to solving the most complex problems towards the end of the project, when it’s already too late.

Rachel Coldicutt