Skip to main content

We’re thinking about updating the design principles

Posted by: , , and , Posted on: - Categories: Experimental

The Design Principles have been in Alpha since April 2012. It’s probably about time they went into Beta. But every time we start doing that we ask ourselves ‘Are the principles right?’

We’re genuinely split on that.

Mostly, the design principles work

Between releasing GOV.UK, starting the transformation of 25 of government’s biggest services, and growing from a hundred people in Holborn to more than six hundred people working in eight departments, it’s fair to say we’ve had a busy couple of years.

The design principles have been at the heart of that work. They’ve been presented to ministers and foreign governments, been picked up by other teams and companies and they’re on posters above digital teams up and down the country.


We’ll always start with user needs, we’ll always design with data, we’ll always believe that making things open will help us make those things better.

So there’s a strong argument for keeping them stable. Updating the evidence that underpins each principle is a bit of a no-brainer, we’re doing that and we need to make the page much more accessible than the current design. But the principles themselves pretty much work.

But they aren’t perfect

Thing is, some of them don’t quite hit the spot.

Build for inclusion isn’t as strong a statement as it could be. We’re huge fans of Tim Berner’s Lee’s tweet from the 2012 opening ceremony... This is for everyone is a huge reminder that a service like GOV.UK isn’t for any particular audience or particular ‘type’ of user. It’s for the whole of the UK (and a lot more besides).

Build digital services, not websites sort of works... but it gets wrapped up in questions about making APIs which, yes, is part of the point. But also it’s about knowing GOV.UK and its services are going to be delivered in a lot of different ways, whether that’s device type, people using our APIs or assisted digital support. Build things people can build on has been doing the rounds here in a its place a fair bit.

Understand context has always been problematic. We expect makers of digital services to have empathy for the people using the things they build, but the context of that use is something you should be drawing from data and research. At the moment the principle implies there’s a ‘context phase’ where you work out if it’ll look good on an iPad, which wasn’t what we intended.


Finally, we aren’t explicit about one of the most important reasons for the success of our projects; the teams working on them. Multidisciplinary teams that evolve and grow (and often shrink again) over the course of a service’s development are critical. And we don’t mention those.

We can solve those problems by explaining the principles better, or we can change the principles themselves. And on any given day we’re pretty much 50/50 on that.

So, we should start with users

And you’re the users, or some of them, so we’d love to know how and if you use them, for example:

  • how do the design principles appear in your team’s work? (Or maybe they don’t...)
  • how do you explain them to new starters?
  • do they play a role in the things you build?
  • if we just deleted them all and wrote USER NEEDS NOT GOVERNMENT NEEDS on the page on 100pt comic sans, would anyone notice?

We’re after stories (and photos, ideally) of how you use the principles and what changing them might mean for you. Facts, not reckons.

Sharing and comments

Share this page


  1. Comment by simonfj posted on

    What a nice post!

    "We're genuinely split on that". Goodoh. So we can agree that something is amiss.

    It's the first one which leads us astray. "User centric" may give us a focus. But only so far as developing single services, and in GDS's case, only single National government services. That, for a starter, doesn't work. One only has to investigate how the IER (inter)operates to see the intricate relationship between Local and National government (databases).

    The "shared schizophrenia" which splits a mind is caused by loyalties to two levels of government. If we were interested in a addressing hearts AND minds then the first principle would be "Citizen centric".

    The reasons we naturally shy away from this approach is quite obvious. We are afraid of repeating past blunders. But they have been necessary steps which have bought us to this point.

    There is no progress here without a citizen having a personal dashboard to aggregate "their" public services (both edu and gov). One only needs a Google account to understand that. Without it, we are condemned to starting a discovery at Stage 3,and not Stage 1.

    We risk making the same past mistakes as occurred when outsourcing "individual learning accounts" by doing the same with the verification of a citizen's ID. "The assumption that the private sector is always superior in know-how and efficiency" is a hard one to overcome.

    Re: Groups (what you call "teams". I must use your long-suffering network manager's terminology). It certainly is the "unit of discovery" as well as delivery.

    How does one provision for inter-organizational groups? The question is particularly hard in the when "groups" and "agencies and other public bodies" are treated as though they were different animals. Perhaps "Be consistent" (in your categorization) should be another principle. (as opposed to "understand context")

    We really do need to take a systematic approach to how citizens discover their groups, so they might build their communities of interest.

    Enough. The only other comment has to do with "transformation". The talk at present is about "transforming services", whereas I'd be talking about 'transforming citizens". We are moving to a more participatory form of democracy (I'd like to believe). That really isn't possible until a comment (like my own) is given the same respect as any of the "internal posts" which tumble from

    I'd never expect to have my comment(s) make the slightest of difference, or receive a reply, from people who are busy "delivering a service." Like every citizen I have been conditioned to be an inert object (regardless of the time spent in educating myself about how inter-networks are constructed).

    But I would be interested in whether, as a citizen, you share this philosophy. It's the only one that gives me, as a citizen, some hope.

  2. Comment by Will Roissetter posted on

    Hello i've been thinking a lot about this recently...I think 2 that I use daily is

    Show the Thing


    User Research is a Team Sport

    They REALLY help in day to day design/research/work at GDS.

  3. Comment by Etienne posted on

    Hi there,

    Working for gov in Australia, I've led a UX team to create a design framework to define and share the functional patterns we use in our online services. I've included an inspired version of your design principles in this framework, mostly to educate new starters and less educated designers.

    Unfortunately, it's more something that we know we should do, but we're far far behind. I guess you guys have a first world problem thinking they could be even better 😉