Falling in the midpoint between the return from the summer holidays and the Christmas period, this Salon was a popular one. In fact, it ended as standing-room only, even after going in search of more chairs. It was rammed.

Clearly, the idea of designing behaviour change (or ‘nudge’) is one that divides opinions. This is not just because of the big philosophical questions it strikes at such as individual agency v. social compliance. It is also about the limits of design and what should be prioritised in its practices. These are not necessarily a new questions. Behaviour change thinking really stepped up in the UK under New Labour, as Kevin McCullough of product strategy consultancy Plan had discussed in Blueprint magazine back in 2007. But it was high time we re-appraised nudge, designing nudge and design that nudges.

So, to return to Friday 20 November 2015, and this write-up, I’m going to depart from the usual template I use for the blog. My behaviour has been changed! This isn’t because of any design interventions that took place on the night. Rather, I was busy finding seats for late-arrivals and ended up jotting haphazard notes from the debate.

In broader terms, a set of background issues (the need for more chairs) were altering my behaviour that evening. And this is something that has always concerned me about nudge. Design can articulate a clear choice architecture, but it can’t necessarily take into account all those other social practices and everyday realities lurking behind the scenes that are going on that come to bear on everyday life. (This is discussed by Will Leggett in Policy and Politics.)

So this blog pulls out some short ‘takeaways’, from the speakers and then draws on some crowd-sourced material. In other words, I asked three attendees if they would write a short, personal response to the event. First, an abridged version of key points made by our excellent chair and speakers.

l to r: Jessica Pyketty, Phoebe Moore (hidden), Dan Lockton, Peter John, Alison Powell, Simon Blyth.

Dan Lockton, Research Tutor, Innovation Design Engineering, Royal College of Art, as the Salon’s chair, kicked off firstly by asking if behaviour is something you can design with and what kind of medium is it, then? He showed that thinking about influencing behaviour is by no means new, but basing this on an assumption that there is an ‘irrational’ set of behaviours that need ‘correcting’ is. This needs to be explored rather than accepted a-critically. There is a new degree of quantification of the self that was previously not possible. This isn’t just about individual fitness data, for example, but found in the rise of Smart Cities, Smart Homes and Big Data. What are the assumptions that are being made here? Is it down to who is instigating the behaviour change and for what?

Jessica Pyketty, Senior Lecturer in Human Geography, University of Birmingham, followed up on this issue of scale. Is it about very small, quite trivial changes to action or does it involve a wholesale involvement of the state, and, indeed, neoliberal systems in the correcting or construction of the self? It’s probably both and therefore it’s equally important to explore the in-between spaces between these two extremes. A number of questions then ensue. What are the rationalities behind and what are their accumulative effects? Is the behavioural intervention open to challenge? What institutions and organizations are delivering these and for whom? What is the legitimacy of paid-for experts (e.g. designers) here? Is behaviour the actual problem to be solved? Again, as I noted above, what background issues are at stake that should or could be addressed instead? What are the unintended consequences of nudge? Does this lead to a fragmented, over-therapeutised self? Does this lead to hierarchies of rationality and irrationality?

Peter John, Professor of Political Science and Public Policy, UCL, viewed behaviour change as offering a fresh set of tools for policymakers that are, in fact, not top-down. Thinking through how a citizen interacts with services forces the policymaker to think like the citizen. Rolling out nudge approaches doesn’t take much in terms of resources. It’s the light touches that matter, he argued. But the challenge is to design nudges that involve some civic participation. There is an experimentalism in nudge within policymaking. The government’s nudge unit has been very transparent with the way the work. By being open about how you’re working and what the benefits may be for all (for example, saving local authority money so it can be re-invested elsewhere), you are treating the public as grown-ups and generating greater empathy.

Phoebe Moore, Senior Lecturer in International Relations, Middlesex University, drew on Gilbreth and Taylor’s time and motion studies in the early part of the 20th century and then Packard’s 1957 text The Hidden Persuaders to illustrate the deep historical background for design for behaviour change in workplaces. To bring this forward, data technologies allow one to see parts of behaviour that are not otherwise knowable – think worker tracking and its analysis to reveal patterns of actions. In 2014, 10,000 companies began to introduce worker-tracking devices that can track steps and heart-rate to provide metrics on activity. The creepy possibility emerging then is whether new forms of tracking will may be linked to appraisals and performance indicators in future, leading to work intensification. Big data accumulation at work can also be used in ways that are less about surveillance and more in helping workers to identify their own best ways of working.

Alison Powell, Assistant Professor, Department of Media and Communication, LSE, pushed the debate away from data as a way of analysing what has happened to data as a way of anticipating a set of potential future possibilities. Futures can then be framed and structured. Nudging design is precisely to do with this. It’s about establishing the range of possibilities and within this, desirable and less desirable possibilities are also established. This then opens up or closes down what is normal in what we can even imagine. Through this, self-trackers, for instance, don’t just represent the body but construct the body. What are the politics of the data that comes from this? Who owns it? When does representing ourselves in terms of data become creepy? The creepiness is when we don’t know what to do with this information. It’s in the design of the system or in the feedback loop between the quantifiable and the possible. You don’t know the extent of knowing about yourself and you can’t control its boundaries. How are you measured? Against what? And what lies outside what is quantified?

Simon Blyth, founder of Actant design research consultancy, spoke from the context of working with the analysis ordinary practices as an innovation consultant to various large corporations. Simon works with consumer packaging and FMCGs. He views the companies as ‘brand bureaucracies’ who are obsessed with marketing – they love abstraction and reduction so that core messages and procedures can be expressed and understood easily and quickly. More recently he has been getting briefs that are not about creating new products but are directed at changing consumers’ behaviour. These can be quite simple things like getting children to wash their hands after going to the toilet. Sometimes the work is then about continuity – how to get people to carry on the same behaviours which may be positive when it comes to things like health. Sometimes these are about compensating for other innovations.

These are some of the highlights. Let’s hear from three attendee-participants. All three have been panellists at previous Salons, so it’s nice to feel that some kind of Salon community is building.

First up, Stephen Feber, wrote to me the day after Salon 18 with the following points.

I’ve used behavioural economics in a project funded by the department for communities and local government – applying it to behaviour change in the visitor economy – where it’s simply too difficult to alter the physical infrastructure but possible to alter the relationship between the suppliers and customers via behavioural change nudges. I ran four innovation labs in Bournemouth with Bournemouth University.

I’m sure you’ve covered this elsewhere but the most obvious point is how generally ignorant we are and how poor we are forecasting. Also, several of the panel did seem to be working with a really old version of the state. Really, the question of the century is how we lighten and simplify our organisational structures to move beyond the command control state. Data visualisation and digital everything start to extend our sensing, responding and thinking capabilities both at individual, community, region and state level. We should be looking at design which facilitates and frees, whilst providing some behavioural boundaries solve the larger questions – energy independence, food independence, population et cetera. Digital connectedness starts this – it’s not simply dark but it does need watching!

So shared surface road design, at a very trivial level, is one interesting example of design that works with the grain of human behaviour – which is the really important part of behavioural economics – design which fits us, rather than design that manipulates us.

[We need] really, intelligent design, resilient design provides iterative/feedback mechanisms that inform the user/creator/builder – so that cycles of continuous improvement happen. And the lesson of nudge is not really about state control but about design that works with us – the good and the bad parts of us – to establish a facilitating environment for learning and growth. When I used to talk about interactive science centre exhibit design I used to use a continuum: instructive – reactive – interactive – creative. It’s perfectly possible to design streets, towns, public buildings, housing which follow this “facilitating environment” idea.

Lucy Kimbell (UAL), who recently completed a one year AHRC fellowship in Policy Lab in the Cabinet Office, shared some perspectives from being inside the policy making environment. She said, I’ve heard David Halpern [Chief Executive of the Nudge Unit) talking about the Behavioural Insights Team and he often emphasises that he thinks its major contribution is a new empiricism and experimentation in policy making. In contrast – or rather alongside this – are new kinds of expertise such as those being introduced by Policy Lab within the Civil Service, which involve setting up collective inquiries using inventive methods in which problem and solution are not fully defined at the outset. You could say that all policy aims at changing behaviour and design is being explicitly brought into play to help achieve this. Positivist approaches that resemble hard science remain the dominant narrative but there are plenty of other kinds of experimentation going on.

Finally, Jocelyn Bailey brought another view from the design consultant arena.

Working for a design agency that occupies itself primarily with socially-focused projects, and quite often at the behest of government and other public bodies, I find myself thinking about the ethical, political, institutional, and value frameworks that shape what we do.

In social policy interventions, broadly, you can either try to change the service, or to change the user (or in some cases to change the system). Nudge and other behaviour change techniques are essentially about consciously attempting to change people, and the argument is, typically, ‘it’s for their own good’.

A lot of the time it seems fairly self-evident what constitutes that social or individual ‘good’ – persuading people to give up smoking, or save money, or get themselves tested for some disease, or do whatever it is the institution in question wants them to do. So on the one hand we are helping solve some social ills (assuming that is even possible).

But in another sense, if we just pretend for a second that ‘social good’ isn’t an absolute, we could view many social interventions as an imposition of one (more powerful) organisation’s set of values onto another (less powerful) group, and we are acting as the go-between that helps that happen even more effectively.

As a practitioner, the questions of which interpretation is the right one, and whose agenda we are really serving, become acute. In one sense of course we know we are very overtly serving a client’s needs or agenda – that’s why we’re hired – but what underpins that? Taken to its logical conclusion, we are either part of the next brilliant wave of social renewal and innovation, or handmaidens to the neoliberal dismantling of the humane state.

Design historians have a much easier time of things. Spotting political agendas and cultural narratives at work in design policy and practice is relatively easy with hindsight: the horrendous colonialism of the Worlds Fairs, for instance. But when you’re in a particular political moment, it’s a bit harder to discern the different currents in the water you’re swimming in – and tell which one you’re riding. And even if you do bother to puzzle it out, there aren’t any clear answers.

So, to answer the question, ‘is designing for behaviour change creepy?’ (sinister?), I guess my work-in-progress answer to myself is, ‘only as soon as we stop worrying about whether it’s creepy’.

The problem is, I don’t see that many design practitioners worrying about it all that much. To get anecdotal, I recently heard the chief designer for a multinational manufacturer of medical devices (and other things) say, to a fairly large audience, and with no hint of awareness that this might be a terrifying statement:  “we can see inside 400 million people’s bodies, all of that data is in our cloud, and we can mine that data to come up with new ideas and products.” This was couched in terms of “delivering products and services that make people’s lives better”, with not even a nod to the profit motive of the company in question. Leaving aside the tricky territory of behaviour change, there is a highly prevalent trait across the mainstream design community for practitioners to see themselves as the heroic ‘agent of change’, without stopping to ponder which master that change serves. Perhaps it’s very easy to believe that doing ‘user-centred design’ means you’re doing something good for people

Thank you to Stephen, Lucy and Jocelyn for sharing their post-match thoughts. And thank you to the panellists, the chair, Dan Lockton and to all the attendees who filled out the event. And if you want to add some of your own comments, please join in!

@guyjulier
7 December 2015