Social Science-Led User Research in Tech. With Rosie Webster

Our guest today is Dr Rosie Webster. Rosie has a PhD and an MSc in health psychology. She’s currently Science Lead for Zinc’s venture builder programme. Prior to Zinc, Rosie worked as a UX researcher at digital health company, Zava, and was Lead User Researcher at Babylon Health. 

While at Babylon, Rosie established the foundations of an effective Behavioural Science practice, which is partly what we’re here to talk about today.

Rosie explains that if businesses are interested in delivering impact and making a difference, then social science can be really key.  She says that research, in similar ways to design, is often underestimated and under-utilised in tech.  Our power, she says, lies in understanding the problem and what the right thing to build is. This is a truly user-centred approach that requires trusting in the process and being willing to scrap an idea when the research points in a different direction.

Often people don’t know what social science is, says Rosie, and equate it to academic research, with the corresponding but erroneous perception that it’s slow, when in actual fact it provides answers much more quickly.

Rosie explains how she established the beginnings of a behavioural science practice at Babylon Health, with the support of two managers who understood its value and importance. She shares why she wanted to ‘democratise’ behavioural research, the benefits of that approach, and how she ‘marketed and sold’ behavioural science within the company.

User research should utilise the existing academic literature more, “building on the shoulders of giants”, as Rosie calls it, “supercharging” primary research, and using evidence to understand what the solution might be. It’s an approach she says results in understanding people deeply, while increasing impact and reducing risk, and without slowing down the fast-paced product development environment.

As our conversation draws to an end, Rosie has a final piece of advice for businesses that are genuinely open to achieving impactful outcomes, and recommends two books for people who are looking to bring behavioural science into their work: Engaged by Amy Bucher, and Designing for Behaviour Change by Stephen Wendel.

Follow Rosie on Twitter @DrRosieW, and connect with her on LinkedIn.

Read an edited version of our conversation which you can read online and also download as a PDF.

Transcript

Dawn Walter: [00:00:04]

Today we are delighted to be talking to Dr Rosie Webster. Rosie has a PhD and an MSc in health psychology. She’s currently Science Lead for Zinc’s venture builder programme. Prior to Zinc, Rosie worked as a UX researcher at digital health company, Zava, and was Lead User Researcher at Babylon Health. While at Babylon, Rosie established the foundations of an effective Behavioural Science practice, which is partly what we’re here to talk about today.

Welcome Rosie, thank you for joining us today. How are you?

Rosie Webster: [00:01:28]

Thank you. Yeah, I'm really good, thanks. Very happy to be here. 

Dawn Walter: [00:01:34]

It’s probably a good idea for me to start with how we met and what sparked the conversation we’re about to have. We were recently introduced by our mutual contact, Rachel Carey, at a Zinc event in Bristol and we started talking about social scientists working as user researchers. 

We both agreed you didn’t need to have a PhD in social science to be a user researcher and that social scientists with PhDs weren’t being properly utilised by tech companies, I think, in part, because most don’t really understand what social scientists do and the value they bring to the design of technology. So social scientist PhDs end up doing things like usability testing which is a serious under-utilisation of their research skills. It’s a bit like seeing a doctor to get your flu jab when a nurse can do this. 

So my first question is: Can you share with us how you feel about this and why you share my belief?

Rosie Webster: [00:02:28]

So yeah, I agree with you. I think that beyond the research methods that are often used in user research like usability testing or like, you know, some one-to-one interview stuff, social science can bring a lot more than that. And I think if businesses are open to using it, it could deliver a huge amount of impact. And I think if businesses are interested in delivering impact, so not just making money, but also making a difference, then I think that's where social science can be really key. 

So, my background primarily in health and health psychology and now sort of working with Zinc on various social good missions including health and more broadly, it's very much all about impact and that's something that I really care about. So I think that social science can really help with that. What I would say is, I think that in the field of Behavioural Science, which is my discipline, is actually think we're starting to see a bit of a shift. So when I was applying for jobs, so, two and a half years ago, there weren’t applied Behavioural Science roles, or at least they were very rare, particularly in London in the UK. I think in the US they were more common. 

Whereas this time before I started at Zinc, a few months ago, when I was looking for jobs, you see a lot more applied Behavioural Science roles and it's just seems to have suddenly, snap, like come from nowhere and now like an enormous amount of companies are hiring for this kind of role. So I think that it is changing. What I would hope is that that sparks a shift across the field in terms of bringing social sciences in more broadly. What I hope it doesn't do is just lead to more just pop science and buzzwords and people kind of saying they're doing it even though they aren’t, which is always a risk. 

Dawn Walter: [00:04:06]

Yeah. Absolutely. You've worked at two different digital health tech companies and you also worked as an independent behavioural researcher. From your experiences, why do you think there's a lack of understanding about social scientists and user research within tech companies?

Rosie Webster: [00:04:20]

Yeah. So I think that I would say that it's research in general there's underestimated, right, like and I think that's quite similar to design. So I've also been kind of based in design teams for the past few years as well. People think that they decide what to build and then they just get researchers and designers to kind of tweak it make it slightly better and that isn't where our power is, right? Like our power is in understanding the problem, understanding what the right thing is to build, but being that kind of truly user-centred takes, first of all, a lot of bravery and kind of trusting the process. 

But also setting aside your ego and trusting the research and the data, and that applies across research and social sciences to be able to kind of trust that the researchers, your scientists, your designers, your product people are going to be able to just build the right thing and something that will be really impactful. 

I'd say also means that it's being open to accepting when your prized idea, the thing that your baby, doesn't work because part of the social sciences and behavioural science is about evaluation and measuring impact and sometimes we measure impact and we find out that actually is not doing anything so we should scrap it. And often people might not necessarily be open to that because they can get quite protective over their own ideas. 

I think more specifically within the social sciences, I think, from what I've seen in the tech world most people just don't know what it is. So they might equate it to academic research, and therefore think that it could be really slow. When actually I think it does the opposite. So I always say it kind of gives us the evidence, the theories, and the frameworks for us to be able to get the answers that we need more quickly but it hasn't got that perception yet, I don't think. Although, as I say, I think this is changing for behavioural science. And I think that is partly actually due to the pandemic. 

So people have realized the importance in human behaviour and understanding it and changing behaviour. People like Susan Michie from UCL being very much in the public eye, which is really awesome. It's really kind of raised the bar there and we've also got kind of more and more popular people in the field, particularly from the US so people like Matt Wallaert and Amy Bucher, who both have great books, are kind of leading the way and showing what behavioural science in companies looks like. I guess on the flip side where I kind of see it having worked well, I think particularly in my time as an independent consultant working with a very small startup. And now, at Zinc working with ventures right from the very beginning, right from their inception. I think that's kind of the space where it can be really, really impactful. Before companies have kind of got set in their ways of what they want to do, and what they want to build, being able to bring the social science and behavioural science in at that stage, when they're kind of very open to thinking about whatever’s going to be best for them in terms of building their product and also very open to being problem-focused. 

So, it's about working with people who kind of want to have and measure impact. So not just vanity metrics, like, daily active users, and things like that, people who actually care about things and companies that are also very focused on research. So at Zinc we have a very strong research and development focus. And Beam, one of the small companies I worked with, I worked with them through an Innovate UK grant, for example, so they were very focused on doing the research to find the right things. 

Dawn Walter: [00:07:34]

So while you were at Babylon, you set up and built the foundations of an effective behavioural science practice. So behaviour change is now part of the product strategy. Can you share how that came about and what strategies you employ to achieve this?

Rosie Webster: [00:07:46]

So I'd say that it was at least partially a bit of luck and being in the right place at the right time. So as a company, they wanted to focus on prevention, they’re an online health care company, they've been delivering online GP services for a long time, but were really keen to kind of think about how we can support people and being more healthy. But I also had a really great manager, and a really great Head of Experience Design above that as well, who were both very invested in this kind of work and saw the importance of it. 

So I came into Babylon as a user researcher, but my always kind of end goal was to be able to bring in the behavioural science stuff. So the role that I even came in on wasn't even a patient-facing team. So I kind of had to work to get into the right places. So I guess initially sort of showed value through delivering training to other teams. So giving them the tools that they need to be able to go bring behavioural science into their work and get people excited about it. I think I was also lucky that I was, I guess senior enough, like with my, you know, all the years of experience I've had as a researcher, I came in at a lead level in the team which meant that I got the opportunity to kind of, once I moved out of the team I started in, got to work on some cross-company strategy projects so that gave me exposure to senior leadership, it got me away from like the more, you know, standard usability testing that we might tend to do when we're embedded in product teams. 

So that was a really great opportunity to kind of get me the respect that I needed for people to be able to think like oh, yeah, Rosie knows what she's talking about, which is always a great start. I've also always been a really strong believer in democratization. So, both on the research and the social sciences side. I think it's important not to be precious about, you know, only I can do my job because that's not the case a lot of the time. This really helps to give me the freedom not to just do the tactical usability testing stuff but enabling product managers and designers to do those things so that I can kind of go off and do those other things. 

And also, we're enabling people to use the behavioural science stuff, it gets them excited about it. Like, if they just come to me and I do it for them, like, yeah, they're like, oh great that's useful. But ultimately is when they actually do it that it’s when people think, oh, yeah, this works, this is interesting and they can feel a lot more ownership over it. So I think the democratization really helped.

I then got various opportunities to work on projects for others. So once I'd kind of done some of the those cross-company strategy projects, I got the opportunity to move a centralized role focusing primarily on the behavioural science stuff. And that was kind of acting as an internal consultant. So I sat centralized and I worked with various teams supporting them on their projects, helping them to kind of think about how they design various bits and bobs of this, the solutions in the product journey, and how they might measure impact. So, for example, I worked with the team who do a lot of our proactive communications about how we can inform the messaging around inviting people for smear tests, for example, or Covid vaccinations. 

And then I think like a broader theme was just getting people excited. So it's just getting people interested in talking about it. So it's as much marketing as it is anything else. And I think, because I was kind of only starting off the function before I left, that is essentially what I, what I spent a lot of my time doing, so having the right conversations with people, working out who to speak to, who to sit down with and kind of tell them about this. 

I got the opportunity to present a weekly company stand up, which is Babylon being a big, you know, 2,000 people company was quite a big released online meeting to be able to kind of really pitch it and market it and sell it as, look, this is something that can really transform what we're doing, but also doing that in a very accessible way. So, rather than kind of talking about stuff that people go, oh this is science, I can't, this is too much and it's too confusing for me, bringing in accessible, simple frameworks, like the COM-B model, for those familiar with behaviour change work, things like that, that people can kind of pick up and run with, and they can think, oh, yeah that makes common sense to me, can really help. 

And in addition advocating outside of the business, so speaking at various events about what we're doing at Babylon. But I also think I was really lucky to have the support from the whole experience design team. So I mentioned earlier, the head of UX and my manager, the Head of Design Research was very engaged but the whole of the experience design team were like super interested, they were a brilliant, bunch, awesome team, really miss working with them. 

But what they did was essentially was act advocates for me across the business at all levels. So I was working with service designers, product designers, content designers, and the other researchers, of course, and they're kind of like, spread across the organization. So they can be like, oh, actually, yes, you should go back and talk to Rosie about that. So they could be in places that I wasn't in. 

And I think I find that design teams are a really good place to start with getting engagement for science in a business because they really share that mindset of starting with the problem and doing what's best for the user having impact those kinds of things, so yeah, they're good place to get buy-in. So yeah, overall I think it's about getting that respect, getting engagement from senior leaders, starting to show value, offering people a way to get the outcomes and answers that they need right, like, you know, so they're like, you want to have some impact, you know, I can help with that, and kind of going along and offering them an opportunity which tended to excite people.

Dawn Walter: [00:13:08]

Wonderful, thank you very much. So I just want you to, for me, to define behavioural science  because I think as an anthropologist I wouldn't necessarily define myself as a behavioural scientist.

Rosie Webster: [00:13:18]

Yeah. So behavioural science is the systematic study of human behaviour essentially. It includes lots of different fields. So psychology is a big part of it and that's the background that I'm from but also behavioural economics will also fall under behavioural science as well. So, anywhere where it's providing us with the theories, the frameworks, the methodologies to understand human behaviour and how we change it, in particular, it is kind of like where it draws from. It's essentially the science of behaviour change, at a fundamental level, I would say. 

Dawn Walter: [00:13:50]

Yeah, so you worked with an anthropologist there [at Babylon Health] and it's not necessarily about who it was, but but the discipline itself and I just wondered how you found anthropology and it's kind of social theories and frameworks and everything, compared to what you know about psychology, how that, how the two are different for you, and do you think they can work together. 

Rosie Webster: [00:14:08]

So I mean, I guess I'll start by being completely honest. We didn't bring a huge amount of theories and frameworks for anthropology in. So I'm not actually familiar with a lot of them and I wonder if that is a lost opportunity, you know, I think that's something that could be really valuable. I think, one thing that I did learn from kind of working with someone who was an anthropologist is, and I don't know, because obviously there's just one person, so I don't know whether it like, how much of it is just personal differences, but like we thought very differently and I think that that brought really, really good strengths. 

So she had the opportunity to kind of like think a bit more creatively and kind of like, yeah, really embed herself in the people side of things which I would also do but I'm a much more of a systematic thinker. So yeah, so we kind of had to work out the best ways for us to work together, and I think that that brought us the fact that we think quite differently brought us strengths as a team. 

 Dawn Walter: [00:14:56]

I also wanted to know I guess the majority of user researchers who perhaps don't have PhDs or advanced degrees like MAs or MScs, what were their kind of backgrounds? You said that you, that they were very open to, to learning more and apply more kind of, I guess, scientific or…

Rosie Webster: [00:15:13]

So largely design, I would say. So in the research team we have a few people who have a psychology background, but just undergraduate level. So they've got the research methods training at that level but like, no further and they were so excellent researchers, you know, I don't think that they were missing anything by not having the kind of advanced degrees at all. They were really brilliant.

 I think other people kind of came from various places. So design tends to be probably the most common one, both on the research side, and of course, on the design side, right? So people come into design and they’re thinking about things from from a user-centred way and that kind of leads them to kind of do research and learn a bit more about research on the job. And I think that you all bring different strengths. So, me, for example, I'm terrible visually and I'm not particularly creative so, you know, I’ll acknowledge where my weaknesses are, but I'm really really good at all the boring stuff everyone else hates around like GDPR and consent and those things because I've been so used to, with my academic background, doing research in quite a like not just rigorous, but regulated way, I guess.

So, I think everyone in the team brings loads of different strengths and I find that really, really valuable. I think like working with designers is one of the biggest gifts that I have found, having made the move from academia to tech, I think working with designers is the biggest gift for me as part of that.

Dawn Walter: [00:16:33]

So you were really lucky at Babylon Health because you were supported by the people who are above you, who recognize what you could bring, and then help you champion it. And I just wonder, there’s probably people in other companies who are trying, perhaps doing the same thing and maybe not having quite as much luck as you did in terms of who their higher-ups are. And and so on I just wondered what sort of strategies you would perhaps advise them to use to in order to effectively market and sell behavioural science, or user research science-led, in their organization. 

Rosie Webster: [00:17:14]

There's two things: there's inspire people and get them excited about it, and then there is showing impact early on.

So in terms of getting people excited about it, I think, and this is something that applies to research as well, I think that the people stories can be really, really persuasive. So you being able to go out there and gather data, particularly qualitative data, to be able to come back, and whether that's showing videos or, you know audio, or just telling people stories is a really good way to get people engaged and I think you can use those stories, I guess, on the behaviour al side side, you can use those stories to find a like illustrate, the challenges of changing behaviour and show kind of how difficult it is essentially. So often, a lot of people assume, oh, we just need to give people the information and then they'll change, which I wish that was true, but it's not so, you know, using that, using those stories to demonstrate that the challenges that could be and maybe bringing some of the theory and to help explain that. 

I think, I mean, everyone that I've worked with, maybe this is just because I'm quite lucky, but everyone I've worked with in tech has been very knowledge-hungry and really interested and curious. So I think you can use that, I think when you start to tell those stories and explain them and like, so this is how we can think about behaviour, you know, based on this example, this is what we know from the science and how we can think about behaviour, that can be really engaging for people. 

So I think that's the start, kind of, step, tell those stories and get them excited and secondly showing impact early on. 

So firstly demonstrating like how you can deliver impact and say look, you know, I can do X Y Z, this will probably help us be more effective, bringing case studies from whether it's other companies or, you know, the academic literature to say, look if we bring in this specific thing, if we bring in the theory, the science, then it has the chance to make us this much more effective at what we're trying to do. 

But also just start early so like I was lucky enough to work with one of the product teams quite early on who were really, really interested and engaged in this, so they're like, great, okay, let's just set up some experiments and start running some stuff. It only has to be like initial and scrappy, you're not looking for like enormous studies, often, unfortunately, not even with statistical significance, but if you can get something off the ground to try something and show that it works and have that proof of concept, then that can be really engaging for people. I think, in my experience, the higher-ups tend to be persuaded by numbers and impact, and the sooner you can start to get and show that, the better. 

Dawn Walter: [00:19:45]

So for any technologists who are listening in today, it might be worth summarizing, from your perspective, what value social scientists bring to the design of technology, and how best to make use of their skills to ensure a flourishing research practice in their organization. 

Rosie Webster: [00:20:00]

So I always describe behavioural sciences, and I think probably social science in general, as giving you kind of an additional layer over existing design processes. So design teams, user researchers are already thinking about things in terms of, you know, like discovery, defining the problem, ideating solutions, testing them, things like that. 

But what behavioural sciences  I found, gives is it just super chargers that, it gives increases your chances of that, you know, building the right thing and building it right. But more importantly, it helps you to stand on the shoulders of giants. So, whereas in my experience in user research teams, we would normally just do that by always collecting primary data, I think we need to be doing more of like bringing in what other people have found so looking at the evidence and that's something that we do a lot of at Zinc. So look at the academic literature to see what other people have found and that really helps you to help you to know what to build, essentially, like, it can be really hard if you start with quite a knotty problem of, you know, of like improving health outcomes in people with diabetes, for example, it can be so hard to know when to start. Especially when you know that a lot of other like companies products, apps, services have all tried to do it and are struggling. 

So like looking at that literature, kind of bringing that together with what you have of your own knowledge and research, can really, really help you. So like find what is that insight of that's going to kind of, you know, give us the impact and subsequently hopefully make us the money, right? So I just think it gives you more information to be able to do that. And it does that generally through like lots of theories and frameworks for understanding behaviour is how I would define it.

So often we can find things in research, but then we can again supercharge that by bringing in a theory to be like, oh actually, yes, like, you know, we've seen from the literature that these things are related in these ways and that's how we could think about it and conceptualize it. And also then once you kind of move on to them, what the solution should look like bringing in the evidence to understand what works. 

So in behavioural science we have a concept called behaviour change techniques and there's lots of evidence for various different ones to say, you know, if you want to improve knowledge, you should use these ones, if you want to improve motivation, we should use these ones and that can be really, really valuable in kind of once you're narrowing down the solution, think, okay, what is going to be the most effective, and then once you've done that, it's about those evaluation methodologies, so you can kind of measure, learn and it's kind of just applying the more scientific stance to that, particularly at the early stages. 

So I think businesses are very good at A/B tests, but an A/B test usually requires you to have built something, so kind of early stage, pilot experiments, get an MVP out there and measure what works. 

But I think essentially it just opens you up to really understanding people at a much deeper level. So, rather than just being, I'm not saying that normal research is superficial, but I think sometimes you risk that, right, like if you're only having a conversation with a few people and you don't often get time to sit and think about it more deeply because you're pushing to move so fast in product development environments, the science can kind of give you that extra layer without adding on the time, if that makes sense. And once we get that deeper understanding, that's where we can find those true motivations in the things that are really going to have an impact.

Dawn Walter: [00:23:10]

So it's about sort of investing at the front in order to save money because if you don't define, don't do all this research properly and define the problem, then you're going to pay for that later on, effectively. 

Rosie Webster: [00:23:24]

Absolutely. Yeah, absolutely. And I think that, I guess relating back to your earlier question about how to convince people, I think framing it in terms of business risk, right? Because like if you don't put just a little bit of extra time in up front, then you risk building something land, doesn't land, it doesn't work, like doesn't actually have the impact that you intended it to. That's an enormous amount of waste of time and money, and you’re, especially if you know you're relying on software developers to build that, their time is expensive. Like why would you waste that essentially, when you could just put it in that little bit of extra time to make sure that you're going in the right direction.

Dawn Walter: [00:23:56]

So before we go, is there anything you want to leave us with that we haven't covered today?

Rosie Webster: [00:24:02]

So what I would say is, I think all businesses will, maybe not all but business, the people who are in the business of human behaviour, which tends to be a lot of businesses, should be thinking about bringing in social scientists. But only if you're open to being problem-led. I think if you already know what you want to build, then it's not necessarily going to be the most effective use of your time or their time or you know, or the money that it will cost you to hire them to bring them in. 

So I think if you're actually interested in achieving impactful outcomes and you're open to building something to achieve that, then I think bring in social scientists, they will really, really help you, like they'll give you a much better chance of having an impact. So yeah, I think that would be my kind of my key piece of advice to businesses.

And to people looking to bring in behavioural science, particularly into their work, there's tons of stuff out there. Everyone's got a book and a framework, but the secret is that they're all the same underneath. They’re all just different iterations of the scientific method, but that's not saying any of them are bad, right, they've all got different angle. The ones that I tend to recommend, Amy Bucher does a really good combination of the behaviour change world, which is a framework out of UCL, and combines that with design knowledge. And her book Engaged is fantastic, I recommend it to everybody. 

And another book that I recommend, which is a slightly different framework, but feels much more relevant to product development, I think, in the way that the book is written, is Stephen Wendel. I think the book’s just called Designing for Behaviour Change. 

But they're probably the two main resources that I would recommend to people who are looking to bring behavioural science into their work. 

Dawn Walter: [00:25:43]

Thank you very much, Rosie, it's been lovely talking to you.

Rosie Webster: [00:25:46]

Thank you.