I always enjoy innovation agency Paper’s Think Drink Do events, in which they have a series of speakers focusing on behaviour change and related concepts followed by a practical crowdsourcing innovation exercise.

This time we had Cath Richardson from the Government Digital Service providing insights about how to encourage government departments to focus on citizen services rather than internal bureaucratic processes. She discussed techniques ranging from providing open and consistent standards that are accessible to all departments who need to adopt a new service so that they don’t have to reinvent the wheel, to recognizing that not every person wants to access a single seamless digital service. Some people actually prefer the specific faceted experience of accessing a single service through their own website instead of in a joined-up single-facing monolithic government interface (gov.uk). GDS is trying to work across a range of citizen needs that allow them to access the services flexibly.

The other advantage of the GDS is they have access to the time and budget to run proper controlled trials on what types of interfaces are the most effective. For example, they found that for most people dropdown calendars (assumed to help maintain data integrity because they don’t allow people to select things outside the correct range) are actually counterintuitive for a lot of people. Their service design manual, while focused on how the government can better serve citizens through digital services, also has many useful insights that can be applied by anyone designing a new product. And it’s open source, so anyone can access it. As someone tweeted last night, when did government become cool?

Our second speaker was Simon Fox of Playlab discussing behaviour change through gamification. Specifically Simon was there to discuss Flowy, a tool/game to serve people who experience panic attacks by helping them apply clinically approved methods of breathing retraining. I was reminded how strongly Simon’s talk resonated with me following the innovation exercise in the ‘third half’, when each group presented our ideas about the brief and midway through a sentence I realised oh God I haven’t taken a breath in a full 45 seconds and I am going to DIIIIEEEEE. Even in a friendly and informal environment like Think Drink Do, presenting to a crowd is really difficult—I’m certain that I am not alone in feeling that. I’ve worked hard to overcome anxiety in my life (or at least, to grow more comfortable with its presence), but even when I’m feeling comfortable and confident, anxiety is a constant influence that may not be visible to others. As Simon pointed out, though there are excellent resources out there for people with anxiety and panic disorders, there are a number of barriers to access to those tools which leave a high number of people suffering without support.

Simon spoke eloquently about the difficulty but the extreme importance of conducting scientific testing when designing tools that are designed to foster real change. This is the only way that Playlab’s tools will have a chance of being recognized as having genuine clinical value. One of the challenges here is that the current trend in software development is to build and alter things very quickly, while the scientific method of randomised controlled trials is designed to slow things down and work in very strict conditions in order to produce statistically viable data which limits confounding factors. Randomized controlled trials are not the only way to run a study, but they are the only way to run studies which can statistically demonstrate a causal relationship between variables and outcomes.

It’s very difficult to make the pace of software design match the pace of designing a proper controlled study. But equally, it’s very difficult for people operating in academic, research-based environments to provide research outcomes at the speed required to action change effectively. I would really like to see some more discussions around this: could trials be run alongside agile development, so the software is iterating all the time while there is a separate sandbox for science? Do developers just have to learn the patience of scientists? Or are there ways that innovative faster methods could be brought to scientists?

As a former academic I spend a lot of time thinking about how ponderous academic research is compared to the pace of change in the private sector where I work now. I’m not saying that we should throw out the scientific method, but certainly from my background in the humanities, I think we’re ripe for innovations which could allow more rapid research processes. Universities are quite rightly resistant to change which would decrease research quality, and I do firmly believe there is the need for time and space to conduct deep longitudinal studies and large clinical trials. (And I’m only talking about studies with people; there are plenty of other scientific processes which also require time, precision, and space in ways that mean they must move at a very regulated pace.)  But I think there should also be room to experiment and to learn from sources outside the ivory tower. I’m far from the first to posit this, but I still see universities struggling with how to enact that value—largely because not everyone believes it to be a value.

For the innovation exercise, this time the brief was very close to my heart: encouraging young people to write stories. We received a brief from the Ministry of Stories asking us to focus on one of three behaviour change areas for a case study affecting a particular young person. My group focused on reward mechanisms and we ended up describing a detailed castle in the sky of reward mechanisms that would demonstrate progress—for example, getting a young person to feel more comfortable with the idea of receiving feedback in increasing detail.

One thing I would love to see in future Think Drink Dos is a report back from organisations like the Ministry of Stories who’ve received innovation ideas, how they were able to apply them, and what they found effective. Perhaps a roundup from several past briefs at once would be useful as a comparison to see which ideas translate across a number of different contexts. In any case, I’m certainly looking forward to the next event!