I was really excited when Naintara agreed to take part in a UX Insider interview. We had a great chat about the role of user research at GDS. I hope you learn as much as I did from this interview.
Can you tell me a bit about your job title and your responsibilities?
I am the lead for the GDS User Research Community.
My focus is on the internal GDS user research capability. I recruit and ensure that all teams who need user researchers have those user researchers and that I have an understanding of what the needs for these resources are. I make sure that we have a set of clearly defined standards, in terms of what good user research looks like. We have ongoing research challenges, so I continuously work with user researchers here to figure out these problems and issues.
“We fundamentally believe that user experience is everybody’s responsibility… No-one within GDS has UX in their job title.”
GDS are seen as top of the tree in terms of good UX and the way things should be done, why do you think that is?
We fundamentally believe that user experience is everybody’s responsibility, so this very much follows on from what Peter Merholz says. No-one within GDS has UX in their job title. The minute you label a particular set of individuals as UX, it can absolve everybody else from the need to ensure that they are meeting the needs of their users.
I think that we have our reputation because we are seen as having a collective remit and agenda to meet user needs. I think this is quite refreshing. I obviously lead the research team, but there are all the other digital specialisms, such as content, design, product, tech… . Everybody has quite a clear role to play but they are also working together as part of multidisciplinary agile teams. I think in those terms, we can make big things happen, and that is really exciting.
A lot of organisations have a very commercial focus and they have very clear metrics. How do you measure success?
In different ways; we have a performance analytics measurement programme, that looks at quite granular stuff. It is very useful in terms of finding out what content is working for us, and what is not. We have a satisfaction survey on GOV.UK, which is currently being worked on by performance analysts and user researchers with regard to how we make that data more useful and to ensure we get better quality data.
On GOV.UK we also run a quarterly ‘usability’ benchmarking programme, looking at task completion, success and those kinds of measures. We scale it and moderate it so that we have enough detail and context to interpret the data accurately – this allows us to gain more granular feedback for immediate design changes and iterations.
The other challenge for us, and something that we are very much working on now, is how we measure overall service performance. We are beginning to look at end to end service design, and how we measure it. My opinion is that there are probably lots of different data points across channels, and we need to look at how we can use those effectively.
“The main thing we learned was that fixing one service at a time wasn’t going to get us where we needed to be.”
Can you talk about anything you are working on at the moment?
Our big drive at the moment is to support end-to-end service delivery across government.
Previously, we built GOV.UK, and transitioned the 24 ministerial departments and then over 300 arms-length bodies. This in itself was a huge task in terms of migration and now in reconciling the information architecture because of the volume of content that we have in a single domain. Our strategy, based on Martha Lane Fox’s original report to Francis Maude, was to have a single domain for government, so users don’t have to understand how government operates in order to interact with it.
Another key focus for GDS was the transformation programme. This involved 25 exemplar services, such as Carer’s Allowance and Register to Vote that are owned by the departments that have the policy remit to deliver them. We helped these departments build those services. The aim of the transformation programme was to seed capability – as government had no real history of delivering digital ourselves, so GDS brought it in-house as we believe we can deliver better services and help to save money. I feel that this has been a revolutionary step.
We learned quite a lot while we were building those services during the transformation project. The main thing we learned was that fixing one service at a time wasn’t going to get us where we needed to be. This is why we’re now looking at end-to-end service design, including guidance and reusable tools and components that can be built once and then shared across government. So this is what we’re doing in programmes such as Government as a Platform (GaaP) – building agnostic service components such as GOV.UK Pay, a payments platform, and GOV.UK Notify, our notifications platform. We’re also looking at our data infrastructure in our Data Programme, which includes building canonical data registers, as well as improving the usability of publishing tools on GOV.UK, and much, much more.
What would you say is the biggest challenge that you are to face in the next 12 months?
The big focus is on supporting service design across government, so that rather than focussing solely on the digital components that are often transactional but not a self-contained end-to-end service in themselves, building in other touch points. We have to ensure that services meet user needs and are delivered in alignment with users’ mental models rather than organisational silos.
At GDS, it means that a lot of our focus is on understanding the needs of other teams across government and colleagues within the Civil Service. For example, the real challenge for aspects such as the payments platform, GOV.UK Pay, is understanding the back-office workflow for payments within departments. There are departments with dedicated finance teams, and others that are so small that when they take money, the transaction is processed by one person. A lot of our attention is focused on reconfiguring the relationship with our colleagues in government, and realising that they are our primary users.
What would you say is the biggest threat to the success of UX at GDS?
At GDS there is no intrinsic threat. The researchers, designers and content teams here have a huge amount of support at a very senior level. The challenge is trying to change some of the deeply embedded thinking across government, and breaking up silos.
In delivering services that reflect the mental models of the users of those services, we are facilitating cultural change – not just making some interfaces, but also changing the way that we think and work together across government.
“we are often deeply moved and surprised by some of the feedback that we get”
What is the most valuable thing you have learnt during your time at GDS?
The most recent is that there has been a bit of an epiphany that our colleagues across government are our users too. As a community of researchers, we have been focusing a lot on empathy-building exercises, to get the wider teams that we work in to also see that.
I think that it can be easy to say: ‘We can revolutionise this, we can fix this.’ It is also easy to wade in, but actually we are often deeply moved and surprised by some of the feedback that we get. Sometimes the answers are a lot simpler; they don’t necessarily need something overly engineered. The problem is often nowhere near the symptom that you are seeing. It is better to be willing and just say: ‘I am genuinely open to understanding more about this situation before I fix it.’
What would you say is the best part of your role?
I get to work with amazing people across the board at GDS, and that feels like a real privilege. Within the User Research Community we have some amazing people. When I first started here, I was very happy to be working with Leisa Reichelt, but was equally thrilled to be working alongside people like Caroline Jarrett too. I used to read Caroline’s blog when she used to work on HMRC forms 15 years ago, and now I am working with her! It’s great having people of that calibre and with that experience. If you have a question about forms you can go and talk to an internationally renowned expert as they are on your team – and that is amazing.
What is that hardest part of your job?
One aspect is just trying to supply the demand to meet the need for user researchers. One of the things I seem to spend a lot of time doing is recruiting and ensuring that we have the resource to deliver the things that we have committed to. I think that it is one of the most interesting challenges about UX. Some of the challenges are quite specific to government here, so I think that all practice leads think quite carefully about the kind of people that we bring in – because a lot of work has particular areas of expertise.
A lot of the skill and aptitude that we are looking for surrounds the ability to navigate the complexities of working on this kind of environment. It is hard to find people that possess both the research expertise and that particular ability.
When you have two candidates, is there a way of telling which one possesses the correct aptitude?
At is the civil service we have to follow a fairly structured protocol. This involves asking everybody the same questions, filling in score sheets and giving them back to recruitment. We get candidates to talk about their work. In the same way as a user researcher would, I always focus on getting behavioural data, and try to elicit as much behavioural, specific and granular data rather than opinions or attitudes. That is how we approach interviews: “Tell us about your work.”
“It feels as if there has been a reinvigoration of the market in terms of having more dedicated UX roles.”
One of the things that I am constantly told in these interviews is that finding people who have the correct skills in the UX community is actually really hard. There are lots who do a bit of UX but few to fill the roles that they need. Have you found the same thing?
Yes, we have. It’s interesting, as we don’t have UX roles here too. The last time I tried to look for a job in UX, about 10 or 11 years ago it was very different. You had very strict usability testing jobs, and then in the early 2000s and during that decade everything became quite generic UX. When I first started my career I noticed that there were still ‘proper’ information architects who had information science degrees or library backgrounds. One of the things that we have struggled to find is individuals with that kind of specialist background, which is sorely needed given the complexity of the issues that we’re dealing with.
So, one of the things that I enjoy here is focusing on a particular area and really developing expertise. Many of us have come from a ‘jack-of-all-trades’ background; we have been designers or undertaken some form of dev work. It’s really refreshing when an individual says: “It’s really nice just to be able to do research.”
It feels as if there has been a reinvigoration of the market in terms of having more dedicated UX roles. Rather than a generic hybrid role that conflates a number of disciplines, there are genuine user research roles or design research roles elsewhere. I think it’s quite hard to find somebody who can demonstrate a particular corpus of work or just an understanding. I believe that roles that are more focused will generate more people with those specialisms.
“Focusing largely on usability testing works very well, but I also think that you still need to do some sort of contextual research up front.”
Do you have any favourite research methods or research tools that you rely on?
For a long time, certainly on the transformation programme, we were simple and straightforward, doing usability testing as part of the agile sprint cycle. We had a really clear rhythm. If you are building a defined service, it helps to have a defined rhythm.
These days we spend a lot of time doing discovery (going into people’s context of use) and using more ethnographic methods. We undertake lots of field-based stuff and look at the broader environment; seeing what users are doing, and looking at aspects that you may be redesigning to understand how they are thinking about it. How is it being enacted and what are the problems, points of pain – and what are the opportunities?
I would say that even if we were settling into a usability testing-based rhythm you would still have a piece of discovery at the beginning. Focusing largely on usability testing works very well, but I also think that you still need to do some sort of contextual research up front. It is important to ensure at the early stages of the project that you have defined the proposition appropriately through a thorough exploration and understanding of user needs.
Working in government is fundamentally different. You can’t compare it to commercial sector organisations. The culture and some of the things that you need to take into account makes it quite different. When I started here, I heard somebody say: “I am blocked today because the Queen hasn’t given royal assent.” Leftfield things like this happen, or a minister suddenly makes a decision. This could happen in a commercial environment, for example when you may have a director making a decision, but I think the ramifications here can be very different.
I am always spotting UX issues in daily life and wanting to fix them – be it door handles, car parking tickets or whatever I come across. Do you share that general dissatisfaction with design around you – and how do you deal with it?
I used to. If you had asked me seven or eight years ago, I would have said: “yes!” Now I am not as bothered, perhaps because I am distracted by other things. I think that because I have worked in this domain for so long, it has been long enough for me to surround myself with better design decisions.
I think it has become more tacit than it is explicit and that is probably why it doesn’t feel like it is an urgent issue, whereas in the past it has been buttons in lifts and other things! I remember I used to have very conscious thoughts and quite strong feelings about buttons in lifts…
It does change the way that you perceive the world, but it is probably more of an unspoken thing. I also think it silently informs a lot of the decisions that I make in terms of the things I have in my life and that I have to interact with and decisions I make. This boils down to (in terms of UX) my daily choices – how I commute, and choosing to take the longer but much less unpleasant commute because ultimately I want to have a good experience.
I saw there was a blog post where you mentioned you have seen a correlation between teams that are genuinely user centred and team morale, can you tell me a bit more about that?
That observation first came out of sitting on our service assessment panels. At GDS, and also across government now, we run assessment panels when a service is either at the end of alpha, so going into beta or we run the beta assessment, which is before they can go live. The team has to come in and talk through the points on the service standard to demonstrate how they have delivered on these points. If they pass the service standard then they can proceed to the next stage. Being on the panel of assessors, we usually have a person there who is assessing the user research because user needs is one of the key points.
There was a difference between the teams coming in for assessment, when talking about their users and the research that they had done. The teams come in to present their services and you have the service manager coming with the product owner, often bringing a user researcher. They also bring someone who can talk about the technical architecture and tech issues. I noticed that it was teams where the service manager and product owner could talk about users and their needs as confidently as the user researcher that really excelled in their understanding of user needs.
You could tell that they worked much better because it felt like the whole team had come on that journey and everybody understood their users, which is one of the things we really wanted to make happen. You could notice how different the dynamic in the team was when there was more eye contact, and when people were talking about their service and the work that they had done in a collaborative and inclusive way. It was really noticeable that there was a different energy in the room.
My background is in psychology and it makes me think about a lot of the more recent developmental neuroscience. There is research now that shows a baby’s mother’s brain changes. Just as the newborn baby after birth grows most of their pre-frontal cortex in the first three years. Being a mother your brain also goes through fundamental changes and it feels like there is a little bit of that too. As teams get closer to their users they also change, they have more empathy and they embody that in other aspects of their ways of working.
It comes back to the fact there is empathy, but it is also being willing to not know the answers and solutions too. It comes back to having some humility too, everyone being genuinely willing to learn from their users and question some assumptions. Be willing to test hypotheses and for it to be okay for them to be a hypothesis rather than, “This is what we are going to do.” and be very solutions orientated.
“As user researchers we have objectives about ensuring that our teams understand our users as well as we do.”
You also mentioned in the same post that everyone needs to be incentivised to meet user needs. Have you got any examples of how that might work?
This was another thing that came out of my mini internal discovery: realising that most of the teams on the ground here at GDS work really well and are genuinely on board with user research and being user centred. There are always things we could be doing better, but largely that is the case.
I think one of the things that Leisa did particularly well was to establish some really key… I was trying not to use the word principles, because we have got our design principles and these are different, but they are principles in terms of how user research should work. One of them is that user research is a team sport. To me that very much means everybody gets involved, your user researcher doesn’t just go away and report back – everybody makes time.
I did my discovery and that took me about two months, because the other thing I was doing at that time was mid-year performance reviews here – which because it is civil service it is this colossal bureaucratic thing, so I was spending a lot of time doing that. I was simultaneously looking at people’s objectives and looking at how they had been incentivised. This involved how they had informed where they had put their attention, their focus and the things they had been doing.
I inherited a whole bunch of people I hadn’t managed before; it wasn’t so much them because they were all user researchers, but it was the first time I was exposed to the real operational machinery of that process, which is a big thing in government.
I was really curious about the mapping between the objectives that any individual might have and things like the design principles or things like the points on the service standard. These other published lists are what are our expectations and standards are, but how does that translate?
As user researchers we have objectives about ensuring that our teams understand our users as well as we do; that we are facilitating user centred design in our teams. I am quite curious about how we can ensure that is also embodied all the way through the organisation in terms of objectives being set for both individuals at these different levels but also in our mission statements.
What separates a good user researcher from a great one?
I think impact is one of the real opportunities of working here in an agile environment. We can have amazing impact, and that is what the great user researchers do. Leisa wrote a really good blog post about this, that a lot of our job isn’t about doing research in itself. I think in her blog post she said, ‘It is probably about 30% of the time we are actually doing the research and 70% is communicating.’ That is completely true, but the communicating isn’t just this very didactic thing. It is understanding the needs of the team, also remembering that they are also our users and if we haven’t met the needs of the team first, then there is never going to be that appetite or openness perhaps to the things that we might be finding out.
As a result, user research becomes integral to what the team is doing. It becomes absolutely essential. It is part of the team, it works well and it fits into the rhythm, it is how we are driving, informing and validating product decisions. Great user researchers have really strong communication skills and are really able to build those relationships.
What is the one thing you couldn’t do your job without?
People usually say coffee, but I don’t drink coffee!
I couldn’t do without my team – in a way that sounds counterintuitive, I don’t know whether it is because I am a researcher or I am a researcher because of this. I am quite reflective by nature, and often I value just being able to talk things through, like idea and concepts.
I like the fact there are great people who are really good at listening and willing to hold that space. It is a very safe place to share ideas and formulate thinking together, that there is a lot of collaboration. Where I get the most reward is working alongside the people I do, the people I work with are exceptional and that is great.
“It is really important that we are advocates and we represent the users, but we still need to do that with sensitivity and diplomacy.”
What is the biggest UX mistake you have made in your career and what did you learn from that?
It’s not really a mistake, but for a long time I worked for a UX consultancy and that can be challenging sometimes because there isn’t necessarily that immediate relationship with your clients. They come in for particular things, have meetings and you will go off and you will do a piece of research and they might come and observe.
There was one time in particular where I think a client was concerned about the way I had phrased a question or whether I was leading the participant. I don’t think I was, but there was clearly some concerns there and perhaps some defensiveness on their part, which happens when research might end up invalidating a particular solution that you have come up with. I think my reaction at the time (and that was before I started working client side or working in internal teams) was not realising that there is always work to do to understand the needs of either the team you working in, for or with.
It is really important that we are advocates and we represent the users, but we still need to do that with sensitivity and diplomacy. I think that was a learning curve for me at that time – I have seen it happen with other people and I am trying to facilitate it with others.
Do you go anywhere for inspiration on best practice in user research or general UX stuff, is there anywhere that you rely on?
I tend to rely on the people that I respect and trust – and those are usually people that I know. Caroline Jarrett gave me a book called ‘Designing for Real Life.’ I know if Caroline gives me a book then I should read it! Similarly Leisa will send me things or colleagues will share a particular article.
Often it’s not about researching or UX-ing per se – it will be about management or organisational culture. I think a lot of what we are doing here is about that, more so than UX stuff because we are trying to fundamentally change how government delivers services. It is probably a few levels up from what we would traditionally associate as the mainstay of the UX domain.
Are there any books, blog posts, podcasts or similar that you tend to recommend quite a lot to others?
There are books that have had a massive influence and are real milestones in my development and learning. Things like Beyer and Holtzblatt’s ‘Contextual Design’ – that is how I learned to do affinity sorting and that is what we now do here.
In a previous role, it was the first big piece of ethnography that we had ever done. I thought: “I don’t know how I am going to do this.” It was completely new; nobody in my organisation had done an end-to-end UCD project ever. I read Beyer and Holtzblatt and basically used that to work it out. It ended up being the template for how we did analysis. That book was a real milestone. I am just really aware that my book choices are really archaic, but those ones are classic.
I remember at a similar time, on the same project, which was for a government client, I was in that situation where I was both doing research, prototyping and testing my designs. I was being challenged by a cynical developer who felt that I was going to introduce bias – and I think that was a fair point, but also he didn’t understand the principles of user centred design. I introduced him to the Inmates are Running the Asylum book, as it was was written originally from a tech perspective. And that seemed to miraculously overnight completely shift my relationship with this one person, so thank you Alan Cooper!
The other things I read are the things that we blog about. Either GDS or other government blogs. The real blog to watch right now is the NHS Beta, their blogs are great. The blogs that we write here at GDS are not just from the user research team but still feel like the most relevant, pertinent and useful thing in terms addressing the kind of things that we are dealing with. The sort of ideas that we are grappling with and being challenged by.
“We understand that firstly we really do need to break down silos in government so you don’t have to understand the structure or who you go to for what.”
Thinking about the future then, what do you think will be different in five years in the way we interact with government?
It will be better. We are talking about transforming the relationship between the citizen and the state; that is the big tagline for the strategy for this term in parliament, so for the next four years. Obviously implied in that is we transform it for the better.
There are massive issues around transparency, trust and privacy, and the need for joined up services. I think we are very aware, because we have done so much research that there is still more to do. We understand that firstly we really do need to break down silos in government so you don’t have to understand the structure or who you go to for what. Also that you don’t have to repeatedly duplicate data that you are entering if you have already shared it with one part of government.
When I was on GOV.UK we did a piece of research, which was originally prompted by the change to Child Tax Credits, which were being replaced by a new scheme called Tax Free Childcare. This was two years ago, and we decided not to do a discovery piece about Tax Free Childcare itself, we decided to do a discovery piece just about childcare and what prompted and influenced childcare choices for families. That was quite a deliberate decision to go and understand the domain rather than a single product.
What became really clear is that as parents you don’t know what help is available. You don’t know all the different options in terms of financial schemes, but also options in terms of applying for nursery or entitlement. As a result, we made a tangible change to the structure of content on GOV.UK to bring stuff together. Previously, all of the childcare schemes were owned by different teams, so they hadn’t been integrated. Even though all of the content was, you couldn’t go in as a parent and see at one glance all of the options available and all of the things you might potentially be entitled to so that you could make an informed decision.
Doing that piece allowed us to restructure the content about childcare on GOV.UK. We made a tangible change and brought together a guide that gave an overview. Childcare – the minute you frame it as that, you are not just talking about HMRC who own the tax-free childcare scheme, you are bringing in the Department for Work and Pensions, you are bringing in the Department for Education and the Treasury also. All of a sudden, the minute that you have got all those people in a room, you can genuinely start thinking about a service for parents who have young children.
When we did the childcare research we ended up presenting it to a room of people – mainly with policy roles from those departments I have just named, so Treasury, Department for Education, the Department for Work and Pensions and HMRC. I was sitting at a table with a couple of people from the Treasury and there was this sudden moment, it was just amazing, this guy piped up and said, “What we need is not an education service, it is not a tax service it is a childcare service.” In government that is radical, it is a fundamental shift in thinking. These are the opportunities to genuinely deconstruct these things that get in the way of us being able to meet the needs of citizens.
We are only able to deliver things depending on how we have organised ourselves. We are limited by the way that we organise ourselves in our capacity to deliver change. We can really begin to change that and think about service design. Actual services, not departmental territories – it is a completely different picture of what a government service looks like. It is not that we have made it better; we are doing something completely different and that is the genuine ambition.
Thanks for reading. If you have any suggestions on how to make these interviews better, please let us know.
Check out more in our UX Insider series