The Definition of a Nudge: Separating the Impact from the Hype

Phil Guest:
Great.
Well, good afternoon, everyone.
Welcome to the next in a series of events across the summer being hosted by Redflags.
We’ve got another great panel lined up for you today on a topic that’s becoming increasingly important as organisations look to reduce human risk.
After all, hackers don’t holiday and neither does the challenge of helping people make better decisions in the moments that matter.
Today’s topic of discussion looks at the definition of a nudge where we’re going to separate the impact from the hype.
Now, if you’ve got any questions that you’d like to ask the panel, please drop them in the chat.
If you’re in Zoom, drop them there.
If you’re on LinkedIn Live, please drop them in there.
And Danny in the background will pick those up and we’ll try to do our best to cover those.
I’m Phil Guest, Chief Go to Market Officer at Redflags.
I’ve spent most of my career in technology, marketing and cyber security.
One theme that keeps coming up is if you want better outcomes, whether that’s growth, security, compliance, or culture, you ultimately have to understand how people make decisions.
That’s why today’s conversation is so important.
We often talk about people being the first line of defence, particularly in cyber security.
But whether we’re trying to improve security, health outcomes,
workplace culture or employee engagement, we’re fundamentally trying to influence decisions.
We want people to make better choices more consistently in the moments that matter.
Behavioural science is helping us understand how those decisions are made, and that’s becoming increasingly important because the same principles are being used by both sides.
Organisations are looking for ways to encourage safer, healthier, and more productive behaviours.
At the same time, threat actors have become remarkably adept at exploiting human psychology, using urgency, authority, trust and social proof to influence behaviour for their own ends.
So the question is no longer whether behaviour matters, it’s how we influence it effectively.
I’m delighted to be joined by three people who bring very different perspectives to that challenge.
Juliet Hodges is a behavioural scientist with more than a decade of experience applying behavioural science in real-world challenges, particularly within healthcare, where her work has delivered measurable improvements in behaviour and outcomes.
Dan Bennett is Senior Partner and Global Head of Behavioural Science at Ogilvy Consulting.
Dan has led more than 1,000 behavioural change projects across major brands and organisations worldwide, and is also the driving force behind Nudge Stock, the world’s largest behavioural science festival.
And joining us from the technology side is Tim Ward, CEO and co-founder of Redflags.
Tim has spent years exploring how behavioural science can be applied to reduce human risk in organisations, helping people make better decisions in the moments that matter, while providing organisations with evidence that behaviour is genuinely changing.
Thank you, panellists, for joining us today.
So let’s start with a term at the centre of today’s discussion.
The word nudge gets used a great deal these days.
And my question to the panellists is what makes something genuinely a nudge and what role should a nudge actually play in shaping people’s decisions?
Dan, perhaps you could kick us off.

Dan Bennett: We need to remember that language is marketing.
And I think it’d be a great thing if the world knew more about how to change behaviour.
So you always get English lessons, maths lessons, history lessons, geography lessons, but you never get behaviour change lessons.
You never get told how to actually get somebody to do something.
And if you went to a car mechanic, you’d want them to know how the car works.
And if you went to a heart surgeon, you’d want them to know how the heart works.
So if you go to a room full of people whose job it is to change behaviour, you’d expect them to know how the brain works and how behaviour changes.
But we’re often quite blind to that.
So I think it’s a great thing if more people in business know more about psychology to know how people really work.
And if the language of nudge does that, then I think that’s great.
Do you know what I mean?
I think that anything that people need to attach onto would be a great driving force.
And we can certainly be more technical about the language, about exactly what is a nudge and what isn’t a nudge and what is behavioural science.
But for me, it’s just a better thing if more people know more about psychology and business, really.

Phil Guest: Right.
And of course, we’re talking about nudge in the context of security today.
But as you touched on there, we use nudges for all sorts of reasons within our work and in our daily life.
Juliet, what’s your view on the use of the word nudge and what makes a genuine nudge?

Juliet Hodges: So maybe I’m a little bit more of a purist than Dan.
I don’t know.
But if you go back to the book Nudge, the 2008 pinnacle of nudging, they are very broad in the way that they define it.
So it’s an element of the environment that changes behaviour in a predictable way without ruling out
other options.
So if you’re someone who feels strongly, you can still go down the path you already wanted to.
It’s not going to actually change what that looks like for you.
And I think you can see how that plays out in a lot of different environments where maybe
people are not as well versed in behaviour change and how the brain works and how to actually influence behaviour.
But they’re calling things nudge because it still fits under that big umbrella term because you’ve got an informational campaign, for example, that does, in theory, change the environment that people are operating within.
But of course, we know that often information isn’t enough to change behaviour.
And I work in health, nine times out of 10 information is not going to change your behaviour.
We know what we should be eating.
We know what we should be
drinking less, we should be sleeping more, all of those things.
So if you have a campaign that is informationally based rather than changes to the environment or something that really is built on why people behave the way they do, then you’re going to be limited in your success.
So I think it really comes down to the actual change that you see in behaviour and whether that could be a nudge or not.

Phil Guest: And Tim, I know this is something quite close to your heart because you often talk about telling people what to do is not a nudge and is not enough.
Perhaps you can give us your definition of a nudge.

Tim Ward: Yeah, I mean, I think both Dan and Juliet picked it up there in terms of if you don’t understand the behaviour, then your lay approach, certainly in security as well,
is we just need to tell people what to do.
We just need to give them the knowledge.
And in fact, we were chatting to someone in the defence world and
one of the generals said, well, can’t we just order people not to click links?
It’s like, why don’t we think?
So for me, the reason it matters is that if you do this properly, it can be really effective.
And practitioners out there deserve to be able to cut through the noise and work out how to use the behavioural science effectively.
And so, to Juliet’s point about the official nudge definition,
if it’s contextual, it’s in the moment, it preserves that agency and it’s focused on changing the choice architecture,
it actually changes behaviour.
That’s what defines a nudge for me.
And so a warning isn’t automatically a nudge, a pop-up isn’t.
And something happening afterwards, after you’ve done the behaviour, is not particularly nudge-like because it’s not changing the decision that’s just been made.

Phil Guest: Yeah.
And touching on that geographical cultural piece that you mentioned, because we’ve talked about the military being told what to do and doing it in the hope they will.
And having come from that world, I can tell you that nine times out of 10, we didn’t.
But there’s also the geographical culture, the national culture.
So one of the things that we’ve seen a lot of recently is what’s happening in markets like Japan and how accessibility to the language, thanks to AI, has meant that they’re under greater threat.
But culturally, getting people to change in different countries is a challenge in itself.
We’re not all going to act the same way.
Juliet, have you got some experience on that from your perspective, the regionality of it?

Juliet Hodges: Yeah, absolutely.
And you see huge differences across different cultures, different populations.
And it’s a really big problem, actually, for the field as a whole. You’ve probably heard the term WEIRD, where all of our samples are overrepresented in Western industry.
Now I’m not going to be able to remember what’s in it.
Western, Educated, Industrialised, Rich, Developed nations.
And obviously, we’re missing out a huge amount of the world and how people operate,
how people in more collectivist cultures tend to approach problems.
And I think one thing that really stuck with me was when you see Japanese football fans at the World Cup and they’re really keen on being respectful and cleaning up after themselves.
Imagine a group of Brits doing that.
How on earth?
We’re not even talking nudges there.
We’re talking
I don’t even know what.
And I think it just shows how big a difference there can be between different cultures.
And so different things will vary greatly in their effectiveness.

Phil Guest: And Dan, you’ve run projects all over the world with Ogilvy.
You must see this, particularly when it comes to branding and positioning, and nudging is a key part of what advertising is all about.
And behavioural science is a key part.

Dan Bennett: And that’s what marketing is really good at, looking at the differences between people.
I think for many decades it was, we’re amazing at segmentation, we’re amazing at hyper-personalisation, we’re amazing at having deep data on masses of people and then changing how we respond.
So actually marketing is the opposite.
Marketing is very good at looking at the differences between people and it hasn’t historically been very good at looking at what’s common between us.
So actually when behavioural science came to marketing, it was really helpful because it reminded us that there were hundreds of thousands, even millions, of years of evolution that were worth paying attention to because, in marketing, we’re programmed to look at what’s different about people.
We’re not programmed to think about why we’re all the same.
So absolutely. But then when it goes back to the differences, there’s some amazing data out there that psychologists have gathered about
how different countries and different cultures will respond differently to authority.
There’s the power index where countries like Australia care less about hierarchical structures than countries like Korea really do, and the positives and negatives that come with both.
Which massively plays into security and the things you were talking about.
So it’s both in marketing.
One is that we underestimated how similar we all are and therefore were missing very simple levers to pull to change behaviour because we were trying to
look beyond them.
And also the psychological data gives us so many new things to look at.
If behavioural science is used well, it should be refreshing.
It should be going, there’s more levers to pull in your brain than you knew.
Therefore, this should creatively stimulate you to think about more ways to change behaviour, rather than trying to tell you what’s wrong or right, really.

Phil Guest: So is it fair to say then that nudging, really in this context, is about pattern breaking?
Is it about getting us to stop and think when perhaps we might be on autopilot?
And is that a broad, simplified explanation of what a nudge should be?
Does it break the pattern of thought?
Anybody got a view on that?

Tim Ward: I mean, I think we see that as a type of nudge. Certainly if you’re about to make a mistake, it can be that kind of knock you out of
that very automatic pilot, system-one thinking into something slightly more conscious because I think pretty much every awareness campaign around the world has a slogan of stop and think.
And I’m always concerned about the fact that that’s probably not actually how people’s brains are working.
If you’re caught up in a really good phishing lure, for example, that’s playing to all these cognitive biases, you’re literally not thinking, you’re just going, oh my God, they’ve locked my account, quick.
And so something that can just be a bit of a speed bump, a little bit of gentle friction.
But I’m not sure that all nudges have to be like that.
I think there are other nudges that are less about friction and more about the easy route.
Like making the secure route the really easy route because we’re all a bit lazy and we’ll take the easiest path.
So I think it depends on the situation, really, how you’re going to use a nudge and what a nudge really is.
I suppose, back to Juliet’s point, the broad definition is it’s something that changes the choice architecture, which might be friction, it might be ease.

Phil Guest: Well, let’s build on that because I’d love to get a sense from you as to what you think an effective nudge looks like.
So you’ve started us on that path, Tim.
So thank you for sharing that.
Juliet, do you have a view, or even examples from your own work, where an intervention has influenced behaviour because it connected with someone at the right moment rather than simply telling them what to do?

Juliet Hodges: Yeah, absolutely.
So one of my favourite examples from my time at Bupa was we wanted to increase the rates of flu vaccination amongst the staff in the care homes.
And, to Tim’s point around making it easy through choice architecture, the absolute ideal intervention in this case was getting a nurse on site who could just do all the vaccines in one go.
No one has to even leave the office. They can be having their cup of coffee in the morning and it’s done.
But unfortunately, that wasn’t possible throughout all the sites.
So
we had to think of something a little bit different.
And I think this is where you can see how nudges can take all sorts of different forms.
So we had some messaging around 70% of NHS care home staff being vaccinated.
That’s a high bar that we want to reach.
We had an opt-out system, so people would receive vouchers and they had to say if they did not want a voucher.
And then we also
gave people maps to where the nearest Superdrug pharmacies, or whatever it was, were, and opening times so they could sit and work through making a plan of how they could get it done.
And we were able to double flu vaccination rates even in the centres that didn’t have the nurse coming to them.
So that’s really great.
But I think you can see how ideally you want to make it as simple as possible for people.
Failing that, there are lots of different tactics that you can use and build a multifaceted intervention
to make people more likely to do the desired behaviour.

Phil Guest: Thank you for sharing that example.
Dan, can you give us one nudge that changed behaviour in a meaningful way and was effective?

Dan Bennett: Definitely. I think, on your earlier question, most behaviour change is about making something invisible visible.
Most of it boils down to that.
And there are definitely some behaviours where you want to get someone from being unconscious about something to being conscious about something.
So we worked with Gatwick Airport on trying to get people to pack their liquids correctly when they’re going through security.
And that is a point where you can’t really do that unconsciously.
You need to block the flow they’re going through, get them to be consciously aware of it and then consider it as they go through.
But then there are others
where you want to shine a light on a previously concealed fact in their head.
So we worked with Adobe for their call centres and this was people ringing up to cancel their Creative Cloud subscription.
And they used to do it where you would ring up and the first thing that would be said to you was, why are you looking to cancel today?
And on the face of it, that’s a great question because you can tell people in marketing that people are leaving for these reasons, so you can reformulate the communications of the product so people don’t leave.
From a call centre operative’s perspective, that’s the worst thing you can ask because it’s getting people to mentally rehearse all the reasons why they want to leave.
So instead, you want to try the opposite.
You want to try the priming question, which is
asking them why they were looking to join in the first place.
What was their original intention and goal, and what aren’t they achieving from it?
Because that’s the real reason they want to leave.
Instead of asking for people to rehearse the reasons in their head why they’re looking to cancel, you get them to repeat the reasons why they wanted to join.
It makes them a lot easier to save.
And that again is about making something previously invisible invisible.
It was already in their head, it just wasn’t, that wasn’t the thing where their focus
was about.
So by asking some of those, and we had a 7 to 8% change in the retention rate by asking that question when we ran tests.
So again, this is where I think that, is that a nudge, perhaps?
Is it an intervention that changes behavior?
Absolutely.

Phil Guest: Yeah, definitely. I think it’s a really good example.
And it’s based on the question you ask.
I mean, what’s the trigger?
And Tim, we talk a lot about that, red flags, about every action takes place in the context of, and a trigger has to happen for something for you to act on it.
Have you got some examples?

Tim Ward: Yeah, absolutely.
And I think it builds on that point Dan made because I think generally we’ve solved the awareness in quotes kind of problem.
Like people have heard of fishing
And most people, when they’ve clicked the link, they realize they shouldn’t have done.
So people know this stuff, but that information is kind of at the back of their mind.
And that idea of that, you’re sort of giving them the time to make that visible.
So.
And in cyber, the moments that really matter, as you say, can be these really tiny little kind of, do you click?
Do you upload a file?
Do you enter your password?
And we’ve been trying to tackle that generally as an industry, so miles away from the context where you’re doing it.
So we have a very effective nudge that one of our colleagues, Lucy, calls our cheese and wine pairing, which are two nudges together.
So as people, if they get an e-mail from someone they don’t know, as their mouse moves over that, that link
ready to cook it, they get a tiny little sort of like a table.
to their mouse that says check the sender.
And that check the sender again is a mantra that the previous few weeks we’ve started to talk about this idea of a very simple coping strategy of if you’re not sure who an email’s from, like check it, like if you’re going to click a link, check who the sender is.
And as we talked about earlier, that can be that kind of, that can knock you off that.
that autopilot.
If you know what you’re doing, you’ve got your agency.
So you just carry on and you click it and go, no, I’ve already done that.
And in some cases, I see that as a pat on the back for people to know what they’re doing.
There’s a nice big red arrow pointing to the e-mail address you’ve got to look at.
But if you actually click it,
We then do another nudge once you get in the browser to go, look, when you start entering your password, you go to hold on.
You did click a link from someone you don’t know.
Now you’re on a website you’ve never been to and you’re entering your password.
You really ought to be just kind of stopping and thinking at this point.
And obviously we aren’t doing kind of RCTs here and randomized control trials and all that kind of, but we’ve seen click rates come down by about 35% on average and in one organization about 73%.
So that’s real clicks and real emails.
So feels like we are changing people’s behaviours there just by kind of bringing back that visibility of information they already knew.

Dan Bennett: To embarrass my dad who recently got scammed.
So he was flying a Canadian airline, we’ll say, and was downgraded from business into economy.
He’s absolutely furious, you know, he’s retired, absolutely furious.
and he posts it all over his Facebook page how angry it is.
He then gets a message from that Canadian airline saying, whoever says sorry, we want to give you a conversation and just, you know, gives your bank details and send you the cash.
Obviously, that then takes more money out of his bank account.
And it’s absolutely genius sleight of hand, isn’t it?
Because his goal is to get retribution to the airline.
someone comes out to him without those flags in place.
Sorry, Keith, for mentioning this one to thousands of people online.
But it’s one of those ones where it’s absolutely genius because they are absolutely acting on the goal from the hot state.

Tim Ward: And he’s in a heightened kind of state as well, isn’t he?
So he’s not thinking particularly clearly because he’s, yeah, wow.

Phil Guest: Great example, yes.
Juliet, any more from your work that you’ve done in the health sector?

Juliet Hodges: Just feeling sorry for Keith now.

Dan Bennett: We should  do a wit round, one of those, when you raise the money back. It was 700 pounds, if we can.

Juliet Hodges: Yeah, we should do. Just one of my favourite ever examples, I talk about this all the time, because I…
I think it’s lovely.
There’s one called Temptation Bundling from the guys at Wharton in Pennsylvania who do a lot of brilliant work.
And what the concept is, that you pair something that you know you ought to do with something that you actually like doing.
So in this case, they asked people to only listen to their favorite podcast while they were at the gym.
And what they found is that people not only went to the gym more frequently, they also spent longer when they were there and they kind of went more consistently over the course of a week.
So I talk about this a lot because
I think sometimes, if there is that thing that you don’t want to do, and a lot of the more tricky behaviour changes are when the behaviour is hard and boring, but how can you then add something a little bit nicer to kind of lure yourself into doing it?
So that’s just one of my favourite examples.

Phil Guest: Yeah, so that sort of reward mechanism there, but you often talk about that.
that priming activity even the night before going for a run in the morning about how making it easy by putting your kit next to the bed.

Tim Ward: Yeah, I mean, and we talk a lot about that kind of thinking about the order in which you want to try and.
help people because with security behaviors, they tend to be things that people don’t really want to do.
And yet as an industry, we’ve always been focused on just the motivation lever, which is the hard one.
And we can all reflect on that being hard when it comes to fitness, like want to eat cake, I want to go to the gym, I learned which one’s going to win.
And so, I mean, it’s not the most purist kind of behavioural model, but the simplicity of BJ Fogg’s models, I quite like that idea.
Well, if you’ve got this idea of the motivation, the ability and how easy it is, but also the prompt, if just the prompt works, you don’t need to worry about all the other stuff.
And so the great example he uses and I have tried is that kind of put your gym kit next to the bed and then when you wake up and trip over it, that’s your prompt.
you put it on and then going running is really easy.
And then all you need is the motivation to actually kind of walk out the door.
And so it’s quite an interesting idea.
But I’m not sure that plays quite so well to that kind of how do you bundle it with something you really want to do?
But I like that one.
And I think in cyber we could do with that kind of how do you add a bit of pleasure to the pain of having to do a cyber security behaviour?

Phil Guest: So we’ve talked a lot about what good looks like.
But nudges, they sometimes fail.
If we see a lot of endless alerts, reminders, warnings, at what point does a nudge stop influencing behaviour and just simply become noise?
What are examples of bad use of the nudge?

Dan Bennett: Things definitely distinguish, don’t they?
And to go back to the previous point, sleep in your gym kit, it’s even better.
You have to take your trainers off, but that really makes sure you’re there.
Sleep at the gym.
And then you sort it.
No, there’s definitely, we’ve definitely found wear out effects.
You know, we’ve put buttons on websites like
“Two flights left at this price”, and we’ve found that’s lost some utility over time, but still maintains a good amount.
So there are definitely wear-out effects. The magic is in making sure that, and this is where behavioural science shouldn’t be the only seat at the table.
It is a missing seat at the table, but it shouldn’t be the only seat at the table.
You want to be using behavioural science as a lens in conjunction with all the other disciplines that you’re already working with.
It’s not some kind of magic separate discipline.
It’s something that really enhances and gets you to think differently about problems.
It should help you find more solutions to the things you’re trying to solve, not fewer.
And really this is about making sure that all the thinking that we do is always trying to solve a real customer need.
And so it doesn’t matter what technique you’re using, if it’s not tapping into a real customer need, then it’s not going to work.
So yes, “two flights left at this price” works because of the scarcity bias.
And all of us are descendants of people who benefited from the scarcity bias.
That’s why we have it within us.
But if people don’t care about
getting the lowest price, which they don’t always,
and if they don’t care about not missing something, which they don’t always, then that nudge won’t always work.
So none of these things are panaceas.
You add the science and it works all the time.
We have to make sure that they’re tapping into real customer needs.

Phil Guest: But that’s the relevancy and the timing again of, is this something that’s even going to work for you?
And we know that most people get caught out by security incidents because they’re rushed.
They’re under a deadline and they’re on autopilot. They miss something and then the cascade starts.
Juliet, from your perspective, why do nudges sometimes fail?

Juliet Hodges: I think, to build on Dan’s point, as well as understanding what the customer need is, you really have to understand what is going wrong for them, what the actual issue is.
And I think there’s one classic example of
a huge trial that was done in Australia to try and reduce teen pregnancy.
And they gave, I think it was all the teen girls in the trial, those robot babies that cry and that you have to look after.
And the whole point of the trial was that they’ll see how difficult it is and they’ll be much less likely to get pregnant.
Yeah.
And what they found was that actually that group of girls were twice as likely to have a baby before the age of 20 than the other group.
And what they decided was the most likely reason for that is that these girls tended to be maybe not so academic, not so sporty.
They didn’t have a lot in their school life and in their life in general that was giving them self-esteem.
And then they had this baby that they could actually look after.
It gives you a percentage score of how well you’ve looked after it.
And it felt good, right?
It was something that they
could feel they could do well.
They felt like they were nurturing something.
They got attention from their peers who would come and check in.
And then they were twice as likely to get pregnant in the end.
So I think actually understanding what is going wrong. It’s not that teenagers just don’t understand how difficult it is.
There’s a whole host of other reasons why they might make that choice.
So I think when you’re just applying techniques at random and not really understanding what’s going on, that’s going to lead you into bad territory.

Phil Guest: It’s a really good, very strong example, that.
Tim, from your perspective on why nudges fail, and also whether at times organisations are solving more for visibility than change itself.

Tim Ward: I think so.
I think,
we always try and push people to take a bit more of a behavioural lens. To Juliet’s point, what do you actually want people to do?
What’s the behaviour change you want?
Because again, in the cyber world, there’s a tendency to just be going, don’t do that, don’t do that.
And so an intervention that’s telling you not to do something, adding
cognitive load, fatigue, and interrupting you indiscriminately, is just going to start to annoy you.
And we encourage people to think about what’s the behaviour you want and how are you going to steer people towards that?
And that comes back to aspects of Protection Motivation Theory, that kind of giving people a coping mechanism and an out, rather than steering them away from something, trying to steer them towards something.
But also, I think we talked about it earlier,
the effectiveness of a nudge does depend a bit on who you’re nudging.
There was some really interesting research by Silvia Saccardo around COVID-related nudges to get people to take up vaccines, and the exact same nudge in two different situations.
In one, it worked brilliantly and the next time it didn’t.
And so that was really interesting.
So they dug really hard into why.
And in the first instance, they found that actually the demographic they were nudging were mostly people who did want to do it and believed in the vaccines and just needed
tipping over, whereas when they ran it again, it was the laggards and broadly people who didn’t particularly believe in it.
And interestingly, they were looking at the demographics of voters, the types of voters as well.
And therefore they actually needed a different type of nudge.
They didn’t just need nudging over to remember to do it.
They actually needed convincing because they didn’t believe in it.
And so I think who your audience is and what behaviour you want
really matters in designing an effective nudge.

Phil Guest: We touched a little bit on measurement there, or started to touch on measurement.
So how do we really measure the sort of change that we’re talking about here?
So you’ve got some examples there. Juliet, you were talking about looking at data sets and pregnancy in young people.
Clearly you can measure that.
Not all things are as measurable as that, particularly when we come to security
within the workplace.
How are organisations measuring the change?
And do they need to?
Does it matter?
Is it more important that they are exposed to this,
that they’re encouraged to think about security, or do we need to measure it?
So let’s talk a bit about that.
And Juliet, perhaps you can kick us off from your perspective. How do you measure this?

Juliet Hodges: Yeah, absolutely.
So I think
a lot of the time it’s about defining, as Tim said, what behaviour it is you’re actually looking for in the first place.
A lot of my work at Bupa was designing wellbeing strategies for our corporate clients.
So you’d go in and they’d say, “Oh, we want to improve wellbeing.”
Okay, let’s take a minute to pick that apart.
And there are all sorts of things that could underpin that.
You could be thinking about, we want people to brush their teeth twice a day.
We want people to go to bed at 10pm.
There are hundreds of different things.
And then
you have to really pick that apart and understand not only what will have the biggest impact in your situation, but also where you have influence, or the most influence.
So if your employer tells you, “I want you to get more sleep”, sometimes that’s just not going to land, right?
So knowing what your place is and where you have influence is also really key.
And you might want to look at even more distal measures.
So even if
the behaviours you’re targeting are things like physical health, sleep, whatever it is, you might want to do a survey that’s actually measuring overall feelings of wellbeing, how burnt out people are feeling, things like that.
So sometimes they aren’t one-to-one and that can be really difficult.
As well in terms of designing your intervention if they don’t map on perfectly and you might make inferences about not seeing an outcome, but maybe the behaviour is happening.
And it’s a really difficult challenge to get right.
But I think if you’re clear on the behaviour, then a lot of things will fall from there.

Phil Guest: Yeah, you touch on a topic there.
And Dan, I know this is really relevant within the world of advertising and marketing, attribution.
Because what you don’t know is what was the trigger or what was the event that might have changed the decision or changed the behaviour at that point.
I know from my time in media, Dan, we used to wrestle with that all the time.
Like, what was the thing that made it happen?
Now we know from that world it’s not one thing.
It’s a build-up of triggers and stimuli that encourage you to think, or at least think at that moment when you see something that triggers you. How do you measure behavioural change?
Because I guess ultimately you can measure how many air tickets you’ve sold.
That’s the outcome.
But could you attribute it to one particular nudge or steer?

Dan Bennett: So a lot of the time we’ll run trials and tests.
So we’ll do A/B tests and randomised controlled trials and compare one thing to another in relatively controlled conditions and be able to say with confidence that it was.
So when we did that Adobe contact centre work, it was actually in Noida, just outside New Delhi.
There were about 100 contact centre agents.
And we had five groups, each with equal numbers.
We had a control group, and you can measure the outcome.
Obviously what you would love to measure, but it isn’t possible, is to stick a USB stick into someone’s ear and find out what they’re thinking.
You would love to measure more things that are upstream, those antecedents, the things that lead up to the behaviour you’re looking to change.
And you can do that with
different implicit and explicit measurement techniques, but it’s quite expensive.
It can be cumbersome and actually isn’t always practical.
So I agree with Jules, which is radically focus on the behaviour and then be very specific about what you want to happen. Then build a plan around that and do your measurement, look at the barriers and drivers, do your research around that, and be very specific about how you change it.
Because all a culture is, is a series of behaviours.
Everything we do is ultimately behaviour change.
And sometimes we forget to define our challenges in those terms, as Jules was saying.
So I think it’s about saying what the behaviour is, making it explicit, and then making a plan to get there.

Phil Guest: Thank you, Dan.
And in the world of technology, Tim, is it easier for technologies to measure that?
And what’s the difference between seeing engagement and evidence that actual behavioural change has happened?

Tim Ward: I think there’s a real question about whether you’re measuring the right thing.
And after this, we should get back to Juliet for that example of measuring around donors and opting in, because I think that was a real example of asking, are you measuring the right thing?
But I think in the world of cyber security we’re driven to measurement by regulation, and we’ve been making some mistakes where we’ve tended to measure
measures of performance, quite performative things like, am I doing something about this?
So how many people have seen my nudge?
How many people have done my training course?
And that’s not to say it’s not an effect, but what’s the actual effect you’re looking for?
Have they reduced the risky behaviour or increased the secure behaviour?
And has it persisted?
Can you track it across a period of time?
And so a big focus for us at Redflags is
measures of behaviour change that you can really attribute.
So you can see that you’ve baselined how often a behaviour happens.
You’ve got a really close cause and effect, or as tight as you can get it, between the intervention and the behaviour so that you can start to see that it is having an impact.
And obviously there are going to be lots of other
things that impact outcomes.
And sometimes you see behaviours go the wrong way because of something else that’s going on within an organisation.
But getting that close cause and effect and thinking about it over time so that you can see what’s going on, see the trends and see where it’s not working, because there will be groups where it’s not working and you may need to change tack with those.

Phil Guest: That’s the important thing, isn’t it?
Not just success all the way through, it’s understanding what’s not working and how you change strategy.
And also, Tim, we’re seeing, certainly within industry sectors like finance now, an increasing requirement from compliance to be able to measure behavioural change.
It’s not something you just report on.
It’s not, have you done the training? Tick.
Yes, everyone’s secure.
We know that doesn’t work anymore.
It’s actually, can you show that there’s been a shift in the culture in the organisation?

Tim Ward: Yeah, and I think insurers and auditors have maybe not been particularly in my good books for taking some very easy routes here, where in the past it’s been,
have they done their training? Tick. But actually with some of the new regulations, NIS2, for example, is a little bit more outcome-based, where it’s looking at human behaviours as both a risk and an opportunity for control.
So measuring whether those controls are in place and whether they are impactful.
So that’s a really positive shift, I think, trying to get people away from the simplistic kind of “did you do something?” to “has it been effective?”
Yeah.

Phil Guest: Dan, Juliet, any thoughts to add to that from your perspective?

Dan Bennett: I think it ties to a really interesting piece of psychology around psychological ownership.
There are two types when it comes to persuading people.
There’s externally led and internally led.
There’s that idea of, I do things because people are looking at me and that’s the thing I ought to do.
And then there’s doing it because I think this is the right thing to do.
And I do think that the world of compliance could be accused of focusing on the external a bit too much and not really getting into people’s hearts.
If you want to change behaviour, you really want to show people how that behaviour is already a part of their identity. So the EU Parliament did some really interesting research. If you use the term “I vote”, it’s a behaviour. If you say “I’m a voter”, that’s an identity. “I’m a voter” does a lot better with that social identity frame. So what we really want to be getting to, and I like that direction of measuring culture, is that I want a culture change rather than a click change. Because you want people to internalise doing the right thing, being responsible for both themselves and their company.
You want that to be something that they internalise and become proud of, rather than saying, “That’s the reason I can’t get Spotify on my work computer”, or “I’m only doing this thing”, which is currently my bugbear.
You want people to understand, to internalise, and for it to be something that comes from within rather than something they do because they have to.

Tim Ward: And I think that’s partly why I’ve always grumbled about phishing simulations, which are a very common approach here, because you are essentially slightly breaking your psychological contract with your staff, like you’re trying to catch them out with a honey trap.
Now obviously you can do it in a nice way, but if you don’t, you’re more likely to be creating that slightly frustrated feeling of,
“They’re trying to measure me.”
“They’re trying to trick me.”
And so you’re not winning hearts and minds at all, are you?
And so I think that’s a real risk.
And it means that security is seen as the people who say no to everything and try to trick everybody.
And they call staff users.
And we know there’s only one other kind of thing that calls people users, and that’s the drug industry.

Phil Guest: Yeah, it’s interesting.
We’ve had anecdotal feedback from some of the customers we’ve been working with about how it changed the perception of the security team when you start to be helpful.
But we’ve talked very much about the corporate side of things. Juliet, the human side is that this is as important to us personally as it is to our work, because this is stuff that follows us home.
This is the stuff that puts our parents and our families at risk, because it’s happening to all of us.
How does that play into this from your perspective?

Juliet Hodges: I do think that is a really interesting challenge because a lot of cyber security risk is so abstract. What someone is actually doing with your data isn’t always visible.
I’m working on a project at the moment about digital therapeutics.
And one reason that that people give as a barrier to using these apps.
So whether it’s a kind of CBT coach on your phone or whatever it is, one reason that people say they wouldn’t use it is because of data privacy and they’re worried about what’s going to happen to that data.
But then people will use Meta and just sign their rights away without even thinking about it.
And so I think it’s such an interesting challenge because I think it’s something people say rather than something they genuinely believe is a risk.
I think it’s an easy way to say, “I don’t want to do that.” But if it’s something that you actually are invested in, or you’re already doing, you already have a Facebook account, you’re not going to stop just because Mark Zuckerberg owns all the pictures of your kid, or whatever it is. I think it’s such a System 2 rationalisation rather than a System 1 thing that we really feel, even though it is quite important. So then that’s where you get back to the measurement point.
I suppose it’s quite difficult because you could do survey after survey that surfaces that as a big thing that people care about, but in their actual observed behaviour you’re never going to see it show up in the same way. So then how do you cross that gap to understand what is important to people?

Phil Guest: These are all incredible insights.
And would you believe it?
We’ve been at this about 40 minutes now talking about it and the time’s flown by.
I am going to move forward to asking you a question around looking ahead.
And while I pose this question to you and you have a chance to think about it, I’ll brief the audience.
But looking ahead five years from now, where do you think behavioural science will have the biggest impact?
And what’s the one thing organisations should be paying more attention to today?
So while you ponder that thought, let me just add that we’ve got about another 10 to 15 minutes left of the broadcast.
If you’ve got any questions, please drop them either into LinkedIn or into the Zoom chat.
Danny’s in the background, he’ll surface those for me.
Let’s get a few questions to our panellists.
But let me come back to the question, the five-year looking-forward question.
I don’t know about you, I’m not great at predicting the future, but from your perspective, how do you think behavioural science will have the biggest impact and what is one thing that organisations should be paying more attention to today?
And let’s start with you, Juliet.

Juliet Hodges: Gosh, I was hoping you’d start with someone else.
I mean, obviously it’s AI, isn’t it?
We’re going to see how that shakes out and how that impacts the world of work and what people are actually doing day to day.
And I think the short-term interesting challenge is getting people to use it and getting people to use it in a way that’s productive and makes them better at their jobs rather than obsolete.
And I think there’s a huge behavioural science challenge there.
But also more broadly, what interests me a little bit more, because I’m a bit of a puritan from the 1600s, is that I don’t think tech is necessarily making our lives better.
And I would love to see a more considered approach to how we can use this to our advantage rather than, “This is the cool new tool.”
So therefore my screen time has to be 10 hours a day or whatever it is.
And there’s one app that I actually think is really cool.
Going back to what we were talking about earlier about taking a moment, taking a beat and thinking about what you should be doing.
This app is called One Sec.
You install it and you can select certain apps on your phone that it will pause for 10 seconds.
So if you click into one, it’ll pause for 10 seconds before it opens the app.
So if it’s something like Instagram or Facebook and you’re using it too much, it just gives you a 10-second pause.
And actually what they find is people will click out because it’s just muscle memory that led them into it.
And when there’s something that prompts you to ask, “Should I really be doing this?”, you change your mind.
So I hope there will be more innovations like that which actually help us take more control of what our own lives look like and how we divide our attention, rather than us feeling swept along by the changes that are happening.

Phil Guest: And it’s so easy, isn’t it, with the device so close to us, attached to us all the time.
And I love that idea of an app.
I’m definitely going to look that up.
I can’t believe it’s 46 minutes into a discussion before the first mention of AI comes up.
Tim Ward: That’s quite impressive.

Phil Guest: That’s impressive. Yeah, that’s it. We know it’s important and we’re addressing it, but it’s not the dominant subject.

Tim Ward: Focused on human intelligence.

Phil Guest: That’s human intelligence. Yes, thank you.
Thank you.
Dan, from your perspective,
how do you think behavioural science will have the biggest impact and what should organisations be thinking about today?

Dan Bennett: I think we’re going to get a lot more psychological data that we’d be absolutely deranged not to incorporate into what we do.
At the moment we look at people’s demographics and segment them by how old they are or some really basic information about them that we have to hand.
But we’re collecting much more psychological and psychographic data right now.
So we know a lot more about people’s personalities than we used to.
We know whether they’re extroverted or introverted.
And we know that if people are extroverted, they’re more likely to be persuaded by social proof and by doing what everyone else is doing.
If they’re introverted, they’re more likely to be
persuaded by loss aversion and pointing out what they stand to lose rather than what they stand to gain.
So that’s the challenge that we have. We’re going to have all this information about people.
What do we do about it?
And the big platforms all have patents out on psychological segmentation.
So we’re not going to be seen by organisations as ages or genders so much anymore.
We’re going to be seen in much more interesting ways.
And that’s going to allow a lot more
powerful information to be shared because it’s going to mean that we can be much more specific about how we interact with someone.
I think that’ll be a really powerful change over the next five years.

Phil Guest: Thank you for sharing that with us.
Tim, what’s your view?

Tim Ward: Well, that was an interesting one, wasn’t it?
Always one to counter.
Some of that’s quite scary, isn’t it?
In some respects.

Dan Bennett: “I know a lot about you.”

Tim Ward: Yeah, indeed.
It might make Juliet turn off the tech even more. And I might be with her.
But yeah, what’s been interesting in our market is the transition away from “this is a knowledge problem” to
“this is about understanding human behaviour.”
And we came up with this category called Human Risk Management.
But for me, that was still quite negative because it saw people as pesky users who make mistakes and need managing.
What it’s now transitioning into, thanks to Gartner I think, is Secure Behaviour Management.
So it’s focused on how we get behaviours to be secure.
Let’s respect our people, but we need the behaviours to be secure.
And that’s a mix of education, behavioural focus, but also the cultural side of it.
So I think that’s quite a powerful shift. And for me, the takeaway for people is to start thinking in terms of behaviours. What does your risk register tell you is a cyber security problem? And then where do people relate to that?
And how can you help them? And then try to take that positive approach. Well, what do you want them to do?
Because sometimes you just need to put the right tooling in place.
Like if you’ve got people doing something silly, such as sending secure documents home to Gmail.
Go and ask them why they’re doing that.
And it’s probably because you haven’t given them a tool to do what they need to do.
And so it’s business process change. It’s not really a security problem.
So yeah, that behavioural lens is really helpful in the world of cyber security.

Phil Guest: Yeah, definitely. And just looking to see what questions we’ve got coming in at the moment.
Just one for the panel to answer around. Somebody mentioned the Nudge book.
If people want to learn more about behavioural science, are there any other resources, training, reading or other things you’d recommend?

Dan Bennett: I defer to Juliet. You read more than I do.

Juliet Hodges: Well, I was going to ask if you still have your Elan. But in terms of books,
don’t read Nudge, it’s boring. I wouldn’t bother with Thinking, Fast and Slow either.
That’s probably worse. I really liked Misbehaving by Richard Thaler. It’s a really nice history of behavioural economics as a discipline, written by the man who invented it.
And it’s part story, part theory, and it really brings you along with it, which is lovely.
I’m looking at my bookshelf to see if I’ve got any others I’d recommend.

Tim Ward: I’m doing the same.
Juliet Hodges:  Gerd Gigerenzer is often overlooked.
I think he’s very good.  I’d recommend Gut Feelings by Gerd Gigerenzer, a bit of a nemesis of Daniel Kahneman, but I think they both have very valid points about the evolutionary value of System One.
So I’d recommend that as a very accessible introduction to how the brain works and why it changes our behaviour in the ways that it does.

Phil Guest: Brilliant.
Thank you, Juliet.
Tim, anything you’d want to add from your bookshelf?

Tim Ward: I quite like the Richard Shotton books because they’re very applicable. They’re easy to read and think, yeah, I can see a way that might be useful.
I think the Groom and Velacott book, Ripple, is quite good as well.
Those are almost write-ups of Ogilvy projects, aren’t they?
Is that correct?

Dan Bennett: It was, yeah.
The founder of Ogilvy Change, as it was called back then, was a man called Jez Groom, who hired Jules and me.
And we did lots of interesting projects and, yeah, they’re in that book.
Really interesting.
Yeah.

Phil Guest: Brilliant.
Thank you for those.
So we’re coming close to the end of our time together and I wanted to say thank you everyone for joining us.
The one thing that’s really stood out for me today is that behavioural change isn’t simply about giving people more information.
Whether we’re talking about healthcare, marketing or cyber security, the interventions that seem to have the greatest impact are the ones that recognise context, timing and how decisions are actually being made in the real world.
We’ve heard a lot about nudges, but perhaps the bigger lesson is that behaviour change rarely happens because someone was told what to do.
It happens because the right intervention reaches the right person at the right moment in a way that helps them make a better decision.
As organisations continue to grapple with human risk, that’s likely to become even more important.
Not simply creating more alerts, more training or more awareness, but finding ways to support better decisions before a risk becomes an incident.
On that note, I’d like to thank Juliet, Dan and Tim for sharing their insights and experience today.
Before we sign off though, Juliet, Dan and Tim, if people want to learn more about your work or continue the conversation, where’s the best place for them to connect with you?
Dan, perhaps you can kick us off.

Dan Bennett: Shout-out to Keith.
Go and see what else he’s up to.
But after you’ve spoken to Keith, we’ve got coming up in five weeks’ time, and we’re super excited about it, the world’s largest festival of behavioural science, Nudgestock.
And this year it’s all about trust issues, so it’s incredibly relevant to these topics.
We’re featuring some of the world’s best behavioural scientists, along with some really interesting characters like Louis Theroux and Matt Hancock, who have intimate knowledge of how trust can be built, rebuilt and broken.
You can still buy tickets for the 17th of September event in London at Nudgestock.com, and you can also stream it online on the day if you can’t make it in person.

Phil Guest: Brilliant.
Thank you, Dan.
I know that we’ll all be dialling into that.
Yeah, thank you.
Juliet, how can people follow your work?

Juliet Hodges: Well, I’ll be at Nudgestock, not speaking obviously, not against the likes of Louis Theroux, but I will be there, so you can say hello to me there.
LinkedIn’s probably the best place to find me.
I’m working on a series of papers at the moment about how we can apply behavioural science to improve digital therapeutics.
So it’s a huge industry with really exciting things going on, like video games that can help
manage symptoms of ADHD, but they’re just not getting the traction.
They’re just not quite working.
So I’m really excited about that.
That should be coming out in early September.
So yeah, follow me on LinkedIn, add me on LinkedIn, and that’s the best place.

Phil Guest: Brilliant.
Thank you, Juliet.
And Tim, how can we follow the great work that you’re doing at Redflags?
Because you’ve got quite a following already on this.

Tim Ward: Well, yeah, the same. LinkedIn really.
Reach out on LinkedIn.
And obviously on our website, we list all the webinars we’re doing and you can sign up for our newsletter and then we tell you about the webinars.
I think we’ve got a few more things in the summer series to go.
And I need to go and buy a Hawaiian shirt, probably.
LinkedIn is definitely the best place.

Phil Guest: Yeah, brilliant.
Well, thank you everyone again and thank you everyone who joined us.
I know we had people from all over the world.
There were some people who stayed up really late in Australia to listen to the conversation.
If we didn’t get to all of your questions, apologies.
We’ll make sure that we cover those within the stream, which is obviously going to be available on LinkedIn.
I hope you enjoyed the discussion.
I certainly learned a lot from our guests today.
Thank you, Dan. Thank you, Juliet. Thank you, Tim.
Do continue to connect with us on LinkedIn.
Follow Redflags.
We’ve got lots of great content coming up there.
And of course, more events like this over the summer period.
So thank you very much for joining us, wherever you are in the world.
Have a great afternoon.
And we look forward to seeing you again soon.
Bye for now.