Home / Blog

Feb 12, 2020

5 things a service designer in government needs to be good at

Here’s our take on the 5 essential things a service designer needs to do when designing or transforming a public service


This article was originally published by apolitical.

We work with policy teams to design new public services, mostly in wicked problem spaces. Policy officials are often very interested in the “secret sauce” of service design, so here’s my take on the 5 essential things a service designer needs to do when designing or transforming a public service. It’s based on watching and working with some of the best service designers in UK government.

It’s just my view of course, you could also ask people like Emma Gasson, Kate Tarling, Chris Atherton, Lou Downe or Martin Jordan — in fact, seeing as they’re on the end of a DM, you probably should. I also recommend reading this article by Audree Fletcher.

1. Ask a better question
In a perfect world a service designer would crisply identify the “exam question” the service needs to answer during the discovery phase, evaluate the research evidence and progress inexorably to a beautifully crafted solution. Unfortunately — in my experience at least — it’s much more messy than that. The question itself begins to change as you do research and start building knowledge of the problem space. In fact, I now think your objective should be to find a better question than the one you started out with as fast as possible.


2. Have ideas
A while ago James put a post-it up on our agile project wall which said:

“Money follows ideas not problems”

For “money” read ministerial interest, political will, programme funding etc. His product managers at the UK’s Department for Education need ideas than can be prototyped and evaluated. This is my happy place — I’m not as good a critical thinker as Kate, I don’t know as much about interaction design as Chris and I’m not always able to ask a better question like Emma but I am quite good at thinking shit up. Service designers need to be constantly coming up with ideas that might work.

The “might work” bit is critical: ideas are cheap but execution isn’t — especially in the public sector. Evaluating an idea probably means committing a team of people to 2–3 sprints of work. If you’re working in a wicked problem space you will most likely never have enough evidence to unequivocally validate your hunches, so being prepared to stick your neck out and operate partly on instinct is one thing that distinguishes a good service designer from a lame one.


3. Have strong opinions, weakly held
If the evidence is pushing me towards a hunch or hypothesis I’m going to run with it and I’m going to socialise my view energetically with the policy team and stakeholders because I want to bring them with me. But if that idea turns out to be wrong I totally reserve the right to drop it like it’s hot and grab another one. This can look like inconsistency to people who like to reach solutions in a linear fashion but in wicked problem spaces I much prefer Paul Saffo’s process:

“Since the mid-1980s, my mantra for this process is “strong opinions, weakly held.” Allow your intuition to guide you to a conclusion, no matter how imperfect…then prove yourself wrong. Engage in creative doubt. Look for information that doesn’t fit, or indicators that point in an entirely different direction. Eventually your intuition will kick in and a new hypothesis will emerge out of the rubble, ready to be ruthlessly torn apart once again. You will be surprised by how quickly the sequence of faulty forecasts will deliver you to a useful result.”


4. Apply technical skills in design and user research
In her blog post Audree likens service designers to architects. In the same way that an architect needs to have an expert understanding of construction materials, the raw materials of a service are content and interactions. Usually the actual design of a service will be executed by skilled interaction and content designers but a service designer needs to have a very good understanding of how both work because that’s how users experience your service — it might be clicking an interface element on a screen, navigating a call centre menu or completing a form and handing it over a counter.

To work effectively with other designers, service designers need to be able to create prototypes to demonstrate a concept and be able to do good heuristic evaluation of content or interactions and make recommendations for how to improve them.

And because good service designers need to have ideas, they need stuff to have ideas with. I’m fanatical about getting involved in the user research and if possible meeting users face-to-face. This goes double if it’s contextual and I get to see the users in their own environment.


5. Understand what government does
It sounds obvious but it took me about a year of working in government to understand why designing public services is different from the private sector. For a good explanation watch Kit’s Tedx talk from about 1:00–2:15.

The most important thing about many government services is that the market will never provide them. Government gets handed the hard, shitty problems no-one else wants to do. And that’s as it should be, that’s why we pay our taxes. Any service designer who doesn’t love that challenge should go and work in retail banking. The money’s better for one thing.