On the 16th November 2013, I delivered a presentation at a Drinkaware conference designed to answer the question ‘Can Technology Drive Behaviour Change?’. What follows is a transcription of the talk along with the images of the slides at appropriate points.

At various points throughout this talk I was rudely interrupted by an iPhone-style app notification. I used this device to underline my point that technology used for behaviour change must still be fundamentally insight-driven and human-centred and the behaviour change must be sustainable and empowering.Notifications often feel like rude interruptions when they are not sympathetic to context. It’s not possible to recreate the ‘interrupt’ effect in a blog post, but hopefully you’ll get the gist. Here we go…


If the question is can technology really change behaviour, then clearly the answer is Yes! From primitive tools and the wheel, to the light bulb and internal combustion engine, to micro chip and mobile devices…


Apologies, that’s my drinks tracker app. It thinks I’ve been dry for 12 days when in reality I’ve not been filling in the data. Seemed like a good idea and kept it up for a few days, but then it became something of chore… and, to be honest, I was lying to it anyway—it’s very easy to lie to technology. Ironic then that many tech for change innovations are based on collecting self-reported data.

Anyway, where was I? Ah , yes…humans and technology.

In some sense the history of the human race can be mapped using its technological innovations—we’ve almost defined ourselves by our ability to invent or create stuff that extends our human capabilities further into the world, addressing the limitations of the human condition—both physically, cognitively and emotionally. So, yes—of course technology can change behaviour.

And throughout today we’ll see some amazing presentations that underline how digital technology in particular can change behaviour—a series of fantastic examples and successful case studies. In a sense, a series of resounding Yes’s to our question Can Technology Drive Behaviour Change?

So, Yes! When we see behaviour change through a technological lens its vast potential to drive positive change is clear.


But! I’m not a Techy, I’m a ‘Changy’ (just made that up). My work is about behaviour change, not technology. I sometimes use technology to help drive behaviour change, but I’m not a technology expert in any sense of the word.

So what I want to do today is provide a bit of counterpoint and, rather than seeing behaviour change through a technology lens, look at technology through a behaviour change lens. And in doing so, rather than answer our question with a resounding YES!, answer it with a more cautious. Yes…but.

Slide06So Yes! Technology can drive behaviour change. BUT! let’s make sure we’re aiming for the right sort of technology and the right sort of behaviour change. Specifically, is the technology insight-led and human centred? And is the behaviour change sustainable and empowering?


This is the fanciest car I’ve ever had. The Audi A4 Avant Estate. A feat of German engineering, it not only served its core function of getting me around, but it was replete with a range of technological features designed to influence my driving behaviour for the better.

As soon as it started raining, the windscreen wipers would come on. As it went dark, the headlights would be activated. The rational logic behind these safety features is clear, but for me it begs the question: if I don’t know it’s raining or it’s dark, should I really be driving a car?

The handbreak had also been redesigned: rather than the long handle with the button on the end, the very physical upward tug, the feeling of it tightening and the affirming sound of the teeth clicking and wrenching… there was an electric button; a button that triggered a sort of electric tightening sound. But more importantly, a button that you pressed down to put the handbreak on and lifted up to take it off.

Again, the logic is clear. But the reality? Well, I kept jumping out of the car leaving the hand break off because I was fumbling to find the button and then lifting it up, rather than pushing it down.

This digital technological assistance was so out of kilter with my analogue behavioural habits (pulling up) that it created a serious safety hazard. It simply wasn’t human-centred or insight-led. It thought I was a piece of digital technology that would seamlessly and perfectly integrate this innovation into my life.

The cherry on this cake of techy behavioural assistance was what my wife technical refers to as the bips. The audio-visual warning system that tracks proximity to other vehicles and let’s you know if you are getting too close—it helps you park.

Well, sure enough when I swapped the A4 for my Vauxhall Zafira—mockingly referred to as the dad-mobile by my friends and completely bereft of bips—I crashed it twice in 3 days trying to reverse park.

And if you think the A4 is a trivial example…

Slide08This is the wreckage of the Asiana Flight 214 airplane after it crashed at the San Francisco International Airport in San Francisco earlier this year. It’s one of an increasingly number of aeronautic accidents that are being put down to automation addiction—or put another way, pilots forgetting how to fly due to over-reliance on digital technology.

So YES!…but! Yes, technology can change behaviours, but is the tech human-centred? Is the change empowering? In these cases, there’s a clear argument that, at least in certain respects, it’s not.

One of the common pitfalls with using technology for behaviour change, is that the technology itself often requires a behaviour a change—it expects us to do new things, or old things in new ways. Rather than being based on an understanding of existing user behaviour patterns and lifestyles, the tech is designed—or simply developed—with an assumption that users will willingly, happily and sustainability bolt new behaviours on to their existing behaviours in order to change some other behaviours.

Meal, exercise and drinks tracking apps for instance—unless you are already keeping a written journal of these activities, starting tracking is a behaviour change in itself, and the reason why much of this sort of technology fails.

In other words this is innovation driven by technological capability, rather than human need or insight. It’s done because it’s possible, not necessarily because it is needed or effective. In other words, its innovation that is led by a presupposed solution, rather than driven by a well-defined problem.



Ooops! Apologies. Grrr!!! Not another one!!! Oh, I see, it’s my stress buster app reminding me that I’m due a 30 second chill out.


Anyway, where was I? Yes. If the technology you’re designing to change behaviours requires a behaviour change in itself, then you need to think carefully about how you motivate adoption and develop new habits.

So YES!…but! Yes, technology can change behaviours, but is the tech human-centred? Is the change sustainable? In cases, where the tech requires behaviour change in itself, data is suggesting that, in many cases, it’s not. Obviously, we’re all eagerly waiting for the internet of things to take care of the data collection for us, but for now the fact that many tech-for-change innovations rely on self-reported data is a fundamental barrier. Not only does it fail to take account of how adept we are at bending the truth (or bending reality to fit our ‘truth’), but it is only relevant for the thin slice of society that is already knowledgeable, engaged, empowered and motivated enough to make a conscious effort to use technology to drive a personal change.

As outlined earlier, Another way to frame this point is to ask whether the innovation is solution-led or problem-driven?

Slide11In 2008 we asked to help address a very serious, complex problem within intravenous drug-user communities.

To cut a long story short, people (lots of people) were dying of heroin overdoses for no other reason than the people with whom they were using—their mates in many cases—were not calling emergency services and getting help in time when an overdose incident occurred.

Why not? Well, yes, the ambulance would arrive to assist, but so would the police. And when you have a pressing heroin addiction, a stretch of cold turkey in prison is a prospect you simply can’t cope with. So the first thing that we required was policy change: an arrangement was made between the constabulary and the ambulance service such that, in an overdose situation, only the ambulance would attend.

However, the main problem was getting this message through to the user community and getting them to trust it. To address this challenge, our client came to us with a full communications plan already worked out—budgeted for so many bus sides, 48 sheets, radio spots etc etc and expected us to simply come up with a communications concept to throw down all these channels.

Essentially, the client had come to us with the solution, rather than a problem and expected a solution-led behaviour change intervention, rather than a problem-driven one. In some sense, they’d come to us saying ‘we need an app’!!

To cut a long story short we convinced the client that this wouldn’t work and that an insight-driven approach was necessary. What do I mean by an insight-driven approach? Well, essentially a creative solution driven by a considered understanding of the problem and the people, And on this, I want to share just a couple of the 10 principles that I use in both my behaviour change training and my projects.

Behaviour is complex… and it happens over time.

Slide12This is the Foresight Obesity Map—a diagram mapping the full range of behavioural influences that determine an individual’s weight. This is the level of complexity that we are dealing with when we work towards behaviour change—this is the messy reality of real people’s lives. Human behaviour is unfathomably complex and if our work doesn’t reflect (or at least respect) this, we are heading for a reductive solution.

So, if we are going to reflect and respect this complexity—and therefore work towards an effective, sustainable solution—we need some way of managing this complexity. Some way of making it actionable. One highly effective way of doing this is to organise influences into personal, social and structural categories.


There’s more detail on this simple tool on my website, but one point I will make today is that in using this tool we resist the temptation to focus disproportionately on the personal influences—to focus disproportionately on the individual as the locus for change—without proper consideration of the social and structural contexts that are often more instrumental.

So when it comes to developing technological solutions to behaviour change challenges, it’s important to understand how that technology relates to this wider behavioural context; which influences you’re aiming to address and how it supports what should be a wider, holistic intervention.

Fundamentally, behaviour change is a journey, not an event: it happens over time and goes through a number of different phases. Prochaska and DiClemente’s Stages of Change model is a classic framework for developing interventions that respond to changing needs over time.

Again, more on this on my website, but to underline the point here that people are likely to need to different sorts of support at different stages of their journey. The type of support required for someone who doesn’t even realise a change is needed is different from someone who acknowledges the need for change, but doesn’t want to. Different again for the person who wants to, but doesn’t have the ability. Different again for the person who wants to and can change their behaviour, but keeps defaulting back to old habits.

So when developing a piece of behaviour change technology, it’s important to consider how it can be designed to respond to different user needs as they shift over time. If this is not possible, at least ensure you understand which stage of the journey your target user is at and design according to those needs.

Ok, back to the IV drug users. So, we persuaded the client we needed to take an insight-led approach, which for us means working intimately with the community we are designing for in order to properly understand the people and engage them in developing solutions to the problem—co-creation based on empowerment principles.

So this is the approach we took in order to deepen the insight with our user community—working directly with that community to create the solution. The intervention that this process led to was wide-ranging and holistic, covering ambient and guerrilla communications, training for support staff, volunteers and users and extensive word-of-mouth, peer-to-peer work. The piece of this jigsaw that I want to focus on to make my point is the small orange card in the following photograph.

Slide15During one of the sessions we ran with our user community, I handed out some contact cards so people could contact me outside the bounds of the session. One of the participants picked up the card, flicked it to check its weight and thickness and then said, “I’m going to keep this. It’s perfect roach material”. Bang! An epiphany—retention value in the chaotic life of an IV drug user. [A ‘roach’ is a rolled up piece of card used as a makeshift filter in a cannabis cigarette.]

So, a crucial part of the wider intervention was this small orange card that we distributed amongst the user community. A business-card-sized card made up of 10 ready perforated roaches, each with our key message and action printed on it. It’s valuable, kept on the person and is present at the time and place that the behaviour needs to be executed… a bit like a smart phone.

Now, the only way to get at this depth of insight, is to get as close as possible to your user community—understand the people, understand the problem. The point that I’m making is that we’d flipped the project from being solution-led to being problem-driven—from being technology-led to being insight-driven—and, in this context, that little piece of card was the technology.

So the next time your client or your boss comes to you declaring ‘We need an app’, make sure you ask ‘why’? What’s the insight? How is this relevant to our the people we need to engage? Is this the right solution, or are we just doing it because we can?


Ah, finally an app that is useful—context specific, relevant and helpful.

So YES!, of course technology can drive behaviour change. BUT! Let’s aim for the right sort of technology and the right sort of behaviour change. Insight-driven, human-centred technology to support behaviour change that is both sustainable and empowering. 

For more information on any of the themes covered in this presentation, please drop me a line at steven@considered.org.uk