April 11, 2020
| Season

A better alternative to classic personality tests and how to apply them to commerce


Jon Puleston




Bart van der Meer


Klik Process


We learn about the development of a more robust measurement tool that can provide a clear picture of human decision making at scale, all applied to (e)commerce.
Recorded during
Marketing Insights Event
This is some text inside of a div block.
Audio only:



Episode guest


A challenge that we’ve all faced when trying to understand customer behavior is that the data just isn’t quite there yet and it’s also a challenge to make it viable for commercial use. But - since you’re listening to this podcast - you are probably very much aware that insights into purchasing behaviour is key to your success. This is where Jon Puleston comes in. He is the Vice President of Innovation at the data insights and consulting company, Kantar and he is a multi-award winning market researcher known worldwide for his thought leadership and new perspectives on the way in which we conduct user and product research. My co-host Bart van der Meer and I spoke with Jon at the beginning of February at a conference and we spoke about his recent paper on developing a more robust measurement tool that can provide a clear picture of human decision making at scale, as well as the techniques he learned along the way that can be applied in the wider world of research. So this is the audio recording of that live session, if you want you can also view the live video recording of my talk with John when you head over to cro.cafe/youtube.

Book(s) recommended in this episode


Please note that the transcript below is generated automatically and isn't checked on accuracy. As a result, the transcript might not reflect the exact words spoken by the people in the interview.

Guido X Jansen: [00:01:40]  You have a session this afternoon at the MIE conference: "the real why and hidden who" so let's start with that. What's your session about?

Jon Puleston: [00:02:09] some work we've been doing with a colleague of mine in WPP. Ogilvy called Christopher. Grace is a very fun, boring character.

And Ogilvy have got quite an advanced viewpoint on using behavioral science in communication planning. And they very interested in understanding human decision making processes, how people think and link to things, things like people's personality. and how they see the world. And and, the papers about the story that, the journey we went on to try and work out a better way of measuring all that in traditional market research, because SSH is actually quite a challenge to measure just the basics of people's personalities.

Guido X Jansen: [00:02:55] It's not as easy as MBTI

Bart van der Meer: [00:03:00] he tries to slip in MBTI whenever he gets the opportunity.

Guido X Jansen: [00:03:04] Yeah.

Jon Puleston: [00:03:06] Yes, lots of there's lots of challenges on lots of levels. the paper that we published around the methodology it's worth anyone reading thing, get hold of the copy. It just tells the series of experiments that we, the undertook to try and explore better ways of getting people to really reveal who they are and how they think.

and trying to overcome some of the frailties in that process and some of the classical ways that, people measure personality in particular, fraught with error. and we dismantled some of the classical techniques and then try to stick them back together again, using different approaches.

Guido X Jansen: [00:03:47] so what are those findings?

Jon Puleston: [00:03:49] in a nutshell, if you were to say the biggest challenge, when I'm trying to measure say your personality in particular is self assessment bias. Yeah. most personality as function of things like, how agreeable you are. And very few people are going to admit that they're not very agreeable or, there's, this level of conscientiousness is one of the classical big five personality traits.

No one admits to being unconscious.  disorganized people just refuse to acknowledge that they are. And so you

Guido X Jansen: [00:04:21] to develop a question that somehow people can answer, honestly.

Bart van der Meer: [00:04:27] Yeah. Because yeah, because it's viewed

Jon Puleston: [00:04:30] as a negative. So what we, what we were able to demonstrate was how it, how that completely corrupts a personality test.

Cause what you get is everyone's in the top half of some sort of. Yeah, everyone likes it. So what we had, we developed quite a lot interesting techniques to try and resolve that simply by making people aware of their biases. We found that she quite effective. You basically, we use the grumpy cat mean, it was seriously factor thing.

We ever tried to work out how to get people conscious in a nice, friendly way about the fact that people aren't suffering selective about some things. So we showed the grumpy cat. Yeah, I'm happy. Then ask people to tell us some of the, how they feel they are compared to other people. And we doubled the amount of moodiness, for example, lots of people when they were exposed to that type of thing, we also discovered that people are able to more objectively look at their past their youth, their growing up period, more than that, they're there now.

So I'm not going to admit that I'm commonly late for meetings. I'm prepared to admit that I was late for school. And so what we realized is that navigating through to the past, it was a good way of, and a lot of our personalities established in the teenage years. And so quite a good rich way to mine, real behavior out of, out the past where people are able to process it.

Bart van der Meer: [00:05:57] and so basically by. Having people tell you about their behavior from the past. Yeah. you're, it's a lot better indicator of what they will do in the future.

Guido X Jansen: [00:06:07] We also know that memory is not the best.

Jon Puleston: [00:06:10] Yeah. One of the interesting we do, for example, to predict whether you're a smoker or not has got a lot of personality traits linked to that.

what am most revealing questions that you could ask is, when you're growing up, did you cross the line much with, your family? Or your parents or school and the people that readily admit to that a 35% more likely to be a smoker. Cause it's, you can understand where that's coming from, they'll push things.

And one of the things they went into what they ended up smoking it on the back of that. And then that's a good example of how. a small facet of your personality can have quite significant implications to your life. In a sense. We also discovered that we were always, if you ask people, how could you use job?

Everyone's better than average.

Guido X Jansen: [00:07:02] Yeah.

Bart van der Meer: [00:07:04] Yeah.

Jon Puleston: [00:07:07] If, but if I were to say. Can you point someone in your document? If are you compared to your brother? I might be a little bit more prepared to admit that I'm quite well all compared to near-field family members. I can be a little bit more honest about who. You know what I'm actually like then the general, what we found is that by asking people to think about themselves in the context of real, tangible people, they know their family, they were, again, they were able to be a bit more honest.

and I think the biggest leap forward was when we started to break down the problem into little bite sized chunks. So I'm not gonna say I'm disorganized, but I might prepare to admit that my bedroom door. Drawers are a bit messy. and I might, I could ask you, when did you do the washing up straight away or do you leave it a bit, or I might say how you park, do you just drive straight in or do you think about it in reverse park, in, those tight, those little we can be honest about those and in themselves, they're not going to be, it's been unfair to it.

Doesn't feel like it's. Honestly, I feel like it's personal. Yeah. Those tiny little clues together. And you get actually quite a more reliable assessment or so much calls in each. And so what we did was literally we deconstructed every one of the big, say the big five personality traits. Literally we tested out for each of those personality dimensions.

We tested hundreds of behavioral based little indicator type questions. And then we did technical principal component analysis to pick apart and isolate the things that are most revealing of each the primary personality traits that people could answer my phone, honestly. And then the survey is made up of lots of little questions like that people ask quite enjoy answering in a funny sort of way, because you can just bring it to life a little bit more human based questions.

Bart van der Meer: [00:09:10] if you get a question, do you park right in the front or. Backwards. It's like, why could they ask this? I have no idea what they can conclude from this, but actually you can,

Jon Puleston: [00:09:20] if you edit it tiny little clue, you can over-interpret some of these things. And so that was one nice breakthrough that we were able to really demonstrate that we could be a bit more rounded.

And then what we discovered is that, How many questions did you answer? we took

Bart van der Meer: [00:09:37] a hundred thousand

Jon Puleston: [00:09:39] for the big five personality traits. There's five primary person. I think we did five over, more than four. We probably did about 20 experiments. And we might test out a batch of 50 or 70 questions for each of them, and then you whittled them down and it would be literally AB testing in a sense, you'd be testing out all these different approaches then working out, which is the most reliable, which is the most comprehensive.

There's lots of subtleties to it because a lot of these questions you realize how dominated your white Western sort of humanity. You're my perspective on the world. We want you to, we want you to develop a test that works in lots of different languages and lots of different cultures. And we first rough of our questions, how all his stuff about recycling, because we discovered really revealing about your levels of conscientiousness, how, whether you can be bothered the cycle, it's actually a really good indicator.

And then we test. Is it, our insurance different countries discovered the, in most parts of Asia, no one recycles. And so you can't that those questions are precluding. Yeah. And when you discuss, when you deconstruct a lot of the personality tests, scarily American, Western culture bias, the questions just simply don't translate across culture.

So there's

Guido X Jansen: [00:10:54] one level of psychology or social

Jon Puleston: [00:10:57] Eastern week. There was one particular element of a personality. It's measuring people's attitudes towards risk taking, and you take a lot of the refugee control. You take a lot of the classical tests and you start looking at the questions. What should I choose to whitewater rafting or, Skype diving.

And you can imagine a retired lady in Singapore asking her that just doesn't mean anything. So that was one of the challenges is trying to think. things that could transcend culture and were stable from one culture to another anyway, to tackle that and the other real, evolutionary stat we discovered is that if you, if you all these personality tests side by side, you start to realize that.

All lip interwoven. and the different cognitive thinking style techniques and different ways. So we think they're all, it's our brains have a mashup of lots of different factors. All these different tests, often have very similar questions. and what we discovered is that if. If you take out the constitute con the core constitutional parts of each personality framework you want to measure and measure them all side by side, they start to inform each other and you could use the results from one part of the test to validate another part of the test.

And when you start doing that, you might yeah. To say, to measure consciousness, I can actually validate it from maybe 20 or 30 data points from other parts of the personality. Framework questions. They start to introduce a degree of like blockchain style stability, because I'm not relying on one particular thing to determine your whole personality measures made up of lots of little ticks here.

So if one particular factor doesn't particularly cost culture that doesn't fall over my personality assessment. So what we found is by combining multiple measures into one survey actually gave us a great deal of stability and cross. Cross correlation around cultures was higher than we were finding with some of the traditional techniques.

So having a quieter, like an Uber measurement protocol, when we think about what we developed as a measure of essentially the big Fife, personalities, but, a range of cognitive thinking, style, fish, things like, locus of control. my sense of control. I feel have a, my life, and the people.

Have feel like they've got a lot of control over that I've actually tend to be slightly more conscientious as well.

Bart van der Meer: [00:13:33] so is it like, you have a, I don't know what the English term is, but you have some, some things you can slide and if you slide deck one, a lot to the right then the other one automatically goes a little to the left.

Jon Puleston: [00:13:43] Yeah. The inter interrelated. Yeah. Everything. and we take every scrap of detail from just actually the way you use the system.  is an indicator of certain personality types. So some people use the far a lot more enthusiastic in the way that these scales are. The people are very measured and that correlates with, regulatory control, risk takers tend to,

and they, and so those that metadata, we also weave into our assessments.

Guido X Jansen: [00:14:23] about something else. Yeah, it looks, it's not about the questions itself. It's about how fast you answer or something that's from.

Jon Puleston: [00:14:31] and then the final part of the story, as opposed to just the way that, we realized is that if you've ever taken a personality test, it's essentially 40 or 50, agreed to squeeze statements, it just drives you nuts off the wall and you get really bored doing it.

And what we were able to observe is after about the 10th time, you were asked to agree to sewage peoples. To patronize their answers and you start to lose concentration. And so what we put a lot of thought in is to how to keep people's attention, going over the court. So maintaining that, and what we did is we broke the survey up into what we call thinking chunks.

We're about three minutes of thinking and we theme them each in a sense around a sort of a construct. and then what we did is we're at the beginning of. We have, what's called a thought starter, something that would like arrests their attention a bit. So we have a,

Bart van der Meer: [00:15:24] if you will, if you win a million euros.

Jon Puleston: [00:15:26] Yeah, exactly. Yeah. We actually have a question in the seventies. If you won a hundred thousand pounds, what would you rather take the money or get slightly more money and have it only a thousand pounds a month? Would you want them all the money up front or do you want to impede. And that's, it's a fun question to think about what, but actually does reveal people's speech of having protocols.

So we would have little questions like that. We had a question, it's the queen, and that was coming to tea. How long will it take to tidy up your house?

assessment. And then we would just call it a questions around that sort of theme about your house, your home or weapons. And, and then at the end of each little chunk, we give them live feedback about what we've learned. So you are this type of, little, we draw out this sort of flair, slider Scouts and visuals and stuff.

You're this type of person or that person. And so they, what they felt is it was. The effort each little bit, because they knew that they'd sought to learn that they were, that we can get something out there, find something out about themselves as they went through it rather than right through to the end was a bit too late to engage them.

And boy did that have an impact on improving concentration? We were able to do some tests control experiments, where we stripped out those elements of feedback. And we were, the people were giving feedback like that. We were seeing about a 40% increase in thinking time. Address overall. And so that was just so that the paper just tells all this, what we did, all these different techniques.

So many of the techniques that we did develop actually could be applied in all sorts of different dimensions of research. So I think that it was some good general learning. I think that's why people like the paper, cause everyone read it, take something out of it. It was just spin off methodology from the back of it.

Bart van der Meer: [00:17:33] It does that imply that all the personality tests we have up til now are. Useless or very inaccurate.

yeah. Plus your

Jon Puleston: [00:17:42] research showed up

to pin down the errors. They have to be long. And so they're inefficient, I'd say I wouldn't want to do, the big five, 40 years old, pretty well established, but I dimension the long form is it's 40 or 50 questions per personality. Simply impractical to do that at scale.

So it's more to do with the inefficiency that I think we've closed off. And because all those tests inherently have lots of, things I'm talking about, but just done in a slide  

Guido X Jansen: [00:18:23] big five. So the skills you use are based on big sites.

Jon Puleston: [00:18:27] more to tell about that, because what we realize is that, a lot of the big five.

Constructs are very academic and really don't translate very well into, commercial decision making. So what we did is we actually tried to transpose those measures in somethings a bit more commercial useful. So

Guido X Jansen: [00:18:44] let's find out during this phase, you might find out it's actually in those five dimensions, but six, four or four,

Jon Puleston: [00:18:49] they didn't use the big five is just the architecture beneath that there's like loads of sub dimensions.

a good example would be, Openness. Now we did this project for Ogilvy. They were, they, they had a behavioral economic theory that, people who were trying to lose weight were one of the facets that they were open. and their reasoning was that they were open. Therefore, B if I say what you like a bar of chocolate, you say

Bart van der Meer: [00:19:18] yes.

Jon Puleston: [00:19:20] Okay. Now in actual fact, Openness. And the classical academic measure is a measure of intellectual curiosity. It's not quite the same as open to suggestion. No. And it

Bart van der Meer: [00:19:35] could be the opposite.

Jon Puleston: [00:19:36] Yeah. And so those two things are being confused just by the language. And so what we realize is that a lot of the, these tests conscientious as well, it's like a general construct that actually has three independent variables, that are really important.

To understand the it's a confluence of fastidiousness Snus and the primary questions in the classical test ask things like, how could you spelling is? Yeah. But it's also about how goal-driven you are, which is different. I could be really goal-driven individual where I'm really conscious. I want it.

I'm going to win this thing. I'm going to go and do it. And. I leave a lot of mess along the way, and I'm not very fastidious about the detail. And then you've got another person who's, very careful about . That's a good, this is more focusing on the careful and doesn't take into account this group, people who in many respects, you'd look on them as being very conscientious.

They really work hard, who status stay in the office late, try and get things done, but they may rush over some data. And then there's another element of it, which is level of organization. And that doesn't necessarily, I can be very organized and therefore I don't have to rush to get, and that the result of that open, thinking about paying my bills on time and stuff like that.


Guido X Jansen: [00:21:02] just like it's in the old test, this is all self-reporting. So you're not actually a dusting out organized. Are you compared to

Jon Puleston: [00:21:17] So a measure of conscientiousness is almost useless for me as a market researcher. I need to know the sub facets of that to be able to do something with it. So what we are, our personality measurement framework is actually made up of the primary ones, but some sub dimensions that we think are important for market position, decision making.

So it's built around that language because everyone understands what the big five are, but we're actually. I'm more interested actually, in some of the sub dimensions,

Guido X Jansen: [00:21:48] then the question is what can we do with it? as a commercial company we spoke yesterday with, was the gaming company, the first cytoma and, also, they had this two dimensional scale for, four images and other big image DataBank, and they have the different magazines.

So which image represents best or works best speaks to the audience of a certain new magazine. And. They create this machine slash DPI system, to use this huge DataBank of, images and put it in one of those four boxes, which fits a certain audience

Jon Puleston: [00:22:24] or for certain,

Bart van der Meer: [00:22:25] the basically help themselves and advertisers predict what image would work best on which,

Jon Puleston: [00:22:30] yeah, exactly.

The same sort of idea is just to Ashley. we're talking to loads and loads of farm work where. You might show people different advertising messages and look at which personality groups, each message appeals to what you can see. It's quite distinctive differential appear between just subtle changes sometimes in a message have more resonant with more, different sub genres of people based upon their personality.

So what the way the root of the Ogilvy methodology for developing thing, any form of marketing. In communication. It's trying to understand the behavioral science of the decision and how people think about it, what and what the emotional roots to that process. And so the test measures that, so what they can do and do is apply theories to communication development, strategy development, and that's essentially where we've had some really great early wins in doing that.

And for example, we did this fantastic piece of work where we tried to deconstruct the person. Personality of people trying to give up smoking. And we, we identified some of the core group of people struggling to give up smoking had, the struggle to process emotions a bit. And that using smoking is a bit of a timeout for, an emotional break they're incredibly fatalistic.

They had  there were, they controlled them. They didn't really control. They didn't have a sense that they could actually get. Give up smoking for that, with their own internal motivations. They've given up that they what's called an external locust of control. They felt that it should, that things would happen.

Yeah. They didn't have much I could do too. Okay. And then we also discovered that they were very, it was called promotional orientated and that they were only, they were, they had a sort of, it's not just risk taking is that they were so getting. Goal-orientated and a bit rebellious. that's the route of going into smoking.

It's got an association with rebelliousness and then the most critical thing was avoidance of negative emotion and that, and they, and the way they used smoking to process that. And then you start to look at your typical anti-smoking ad and it's just like a picture of someone dying of cancer. And we did some really interesting experiments where we show people, ads, just random ads.

And then we drop in one of these antiques. The key messages and compared how smokers and nonsmokers reacted to them. And essentially the non-smokers just click the next button. And we timed how long those look at them. And they were literally clicking the next button almost twice as of 50% faster than the smokers, because it was setting off alarm bells internally.

And so the way that we're dealing with those negative emotions is to blank out almost. And so these it's not say that those sorts of messages not effective for certain groups of people to encourage them to give up smoking. But this hardcore group of people are really struggling. I'm obsessed that and the way they do it's just Oh no.

Don't tell

Bart van der Meer: [00:25:28] me so the question is it's a good ad for non-smokers,

Jon Puleston: [00:25:33] but it's a bad ass.

Guido X Jansen: [00:25:34] Oh, websites we talk about at banner blindness is something you learn. Ignore.

Jon Puleston: [00:25:40] So what we then to do, we trying to then build from the ground up and different approach to communicating the opportunity to give up smoking, around solutions around exploiting or called external locus of control reasons.

Bring up you're getting married or you've just had a grandchild. These are wonderful external motivators to give up smoking and showing people achieving that those, those moments in life where people were able to actually actively achieve giving up smoking seeds, the opportunity in their minds of someone.

And then for the capturing this idea that emotional orientated mindset we developed some messaging around. positioning, giving up smoking is a bit of a challenge that these people had overcome and fuck you. I gave up smoking, they're smoking, a really like the rebellious,

Bart van der Meer: [00:26:30] but then the other way around

Jon Puleston: [00:26:32] reverse engineer and then the fatalism of it all, we developed some messaging around tomorrow's the day sort of thing.

And then what we did exactly, same thing we showed people's selves about drop these ads in. And what we're finding is that non-smokers these are there were there where there's no reason. Just a different, but the smokers were pausing and looking at them and we realized that we were connecting, they were connecting.

And it's difficult to know whether it's a good example, of how by really understanding the underlying personality traits and the how people think it's a really brilliant starting point for developing better forms of communication. And really, I think that's, it says revolutionized is how you think about developing after the communication.

It's really what could they have really pioneered? And they really are pioneers of that methodology that they're ugly center for behavioral science is really driving forward their business. And what they hopefully this test does is it gives them a platform to be able to implement that more effectively by actually having a proper framework to measure a theory because they're up until now.

I have a construct. I need to be able to measure whether it's true or not. Before I press ahead with the spending. Money on the advertising.

Guido X Jansen: [00:27:47] Sounds a bit like your chase chasing a moving target. Almost like with a, I think when I read the book from Erin Meyer about culture, what really resonated with me that said, okay, culture in itself, isn't something.

it's always relative to another culture. I think it's very similar with personality. Personality in itself doesn't mean anything unless you compare it to something else, those skills. And like you said, smoking's, relates to being rebellious, but I think 70 years ago

Jon Puleston: [00:28:16] it was the norm. It was socially acceptable.


Guido X Jansen: [00:28:20] So how often should you

Jon Puleston: [00:28:21] be revising? I think that you could think about that. there's a whole area of situational relevance to some of this stuff as well, because My level of conscientiousness will vary. You know what I mean? The supermarket buying a washing powder. Yeah.

Compared to buying a car, my level of conscientious  this, that's probably more significant situational. Factors and trying to understand the situational personality variance as much as in a sense that's what a time is. It's a situational factor in a moving decision making framework. the core personality traits, the same.

This time was 50 years ago, but the situation of buying and choosing cigarettes has moved on and therefore you need to understand that. Yeah. So it's really important to understand that the situation framework and what you're trying to measure, and then often what we do discovered that it's actually, these are personality test is a starting point.

You probably, sometimes I have to bolt on extra little things to understand a little. Other little facets. So for example, we've done a lot work in the healthcare sector. there's this thing called optimism bias. that, really dominant in your likelihood to take treatment options up seriously is literally whether you think you're invincible or not, whether you think that you're vulnerable to and it's, and you can see it with people's.

say for example, diabetic stage. Was it stage two diapers before they developed?

Bart van der Meer: [00:29:53] Yeah. One is before two is

Jon Puleston: [00:29:55] when you can get rid of it, then yes. The early stages. And what you,

Guido X Jansen: [00:30:00] luckily I don't know

Jon Puleston: [00:30:01] if you, if you look at, measure people's optimism bias and that you can measure it with a cluster of questions, it correlates with their lightly.

Heard of actually taking the actions, don't think go out and doing some exercise or doing, responding to that thing. Yeah. In particular, we found that, personality is really important to us. How health care applications, likelihood of taking treatments and avenues for encouraging people to take up treatments there, the relationship with our emotions and how they respond to different types of messages.

Some people really block out emotional advertising messaging that are too strong. And other people respond to more the practical aspects of it. And they want solutions and other people that have a barrier to their confidence of getting over that first thing, the first step, they don't feel that.

So if you,

Bart van der Meer: [00:30:48] that's fascinating.

Jon Puleston: [00:30:50] So when applying this,

Guido X Jansen: [00:30:52] for example, to buying behavior, When it's, when I give presentations and I talk about, you have also the experience of that person might very, have a big impact on their expert level of buying something. for example, I'm the experts of buying wine.

Maybe I'm very, influential. Being influenced by my, whatever my friends do or whatever I buy. I look at my friends first. Yeah. But when it comes to wine, I'm the expert. So please go away friends. You didn't know nothing about this and the wine experts. I'll tell you. I'm not going to listen to you anyway.

So how far his personality? A predictor for buying behavior.

Jon Puleston: [00:31:33] how much

Guido X Jansen: [00:31:34] it's not going to be a hundred percent

Jon Puleston: [00:31:35] obvious. One of the things we very specifically woven into our measure is, the sort of head or neck framework, which you make decisions. whether you're using your feelings to make decisions and what information you feel you need to make a decision we've retooled, raking three control.

It was a risk taking thing, actually that the questions we devised centered around how much information you need to make a decision. If you're buying a car, do you need to go and do some research also just go down to the showroom and jump in a car and see what feels nice. Yeah. But what you've just identified is one of the.

the sense of self confidence in your own knowledge and your decision making skills, and that will vary some people have a lot more self confidence. and it's one of the dimensions you mentioned this sort of it's called locust of control, your sense of self competence, your ability to make your own decision, which might be it's incredibly situational.

So if I recognize that I know a lot about a topic, I'll rely on my own opinion. And if I'm not, I then might be, am I the sort of person that would then seek advice or w or am I now the person that would want to go and still work it out for myself, but will take more time to work out. And there are so that's, you pinpointed without that bit of information.

In a particular buying scenario, I'm a bit blind. And so one of the things that I biggest frustrations with all forms of generic segmentation in the sense we've developed generic segmentation model, is that you cannot just rely completely on that generic, separate segmentation saying what all decisions couldn't be flown and predicted out of this.

You're always have to augment. with some extra things that are relevant situation relevance up to think. And I think you pinpointed what really good example of that?

Guido X Jansen: [00:33:22] Yeah. It's the famous saying, all models are wrong, but some are useful.

Bart van der Meer: [00:33:27] Yeah. Yeah. But I think

Jon Puleston: [00:33:29] we're just, I'd like to think about these as lenses almost to look at Christian Christ.

That's his is his language. It's saying they're just lenses to look at the world. And give a different viewpoint of, of how to think about it. They're not answers. They're just like ways to look at people.

Bart van der Meer: [00:33:45] Yeah. And I think it's also true that, I think you can reverse your example as well. So somebody who usually thinks he's an expert, but specifically on wine, he listens to his friends.

So as long as the buyer behavior lines up with, especially ways that with certain types you can optimize for it, or

Jon Puleston: [00:34:04] at least try though. it's almost diverting a different feel, but I've done a lot of work on social influence. and you realize how significant that is. So social proof, social, the certain, in the past, you'd have fashion.

Things like haircuts and clothing. Yeah. music and stuff were deemed fashion driven and that's basically level of social exposure because I can see your heck. I can see a million people's haircuts every day. I know what's and so those markets were subject to fashion changes. Other things like, what wallpaper was in your living room?

I would not get me so exposed to that. It's only when I go around to your house, I get to see. And so those types of choices, they're also fashion driven, but because unlike exposure rate, To your house. And so the  of the fashion is very low on that. There's the fashion cycles longer. And so you can actually plot it out.

The more social exposure, the more churn you get of choice, really. And then as you go down to something like boot, Polish, Where it sits in your cupboard. You only get it out now. There's no social context. Interestingly, the number one boot Polish brand a hundred years ago, it's still the number one boot Polish.

This is not be no social influence driving. There's no fashion in that. Now what's happened with, basically social media has opened up the house to social exposure. And also it's opened up drinking occasions to huge amount of social spaces. So now I'll post a picture of myself having a drink.

And what we've been observing is that I'm more, a lot more comfortable sharing a photo of a cool brand of beer. Then the regular brand of that. In fact, I did some analysis of, for example, whiskey. Yeah. if you scrape like a hundred photographs of whiskey off Instagram, and then you look at the top three or four brands of whiskey, you won't see the material in any of those.

They're all in the obscure little is not cool too. And so that's the essence of fashion is that it, no one wants to be mainstream. And so that. Fashion social influence is deacon's basically attack it, attaching all sorts of market sectors. That once was pretty stable. And so the beer market is turning fashion driven.

The light, the lifecycle of beer brands is dropping down to maybe five, six, 10 years to down to two or three. And maybe I think I suspect there's going to be faster and faster. And so I think that's a really good example of a new thing. You have to understand human decision making. It's the influence of the social context and how it invades, and particularly vulnerable to that really freaky things.

For example, we discovered dogs. I have a dog dogs, the number one, social media, accessory on

Bart van der Meer: [00:37:05] Instagram dogs or cats.

Jon Puleston: [00:37:06] Yeah, exactly. Dog sales have gone up. If you actually look at, the rise of Instagram and the rise of dogs, I'm going to ship. Yeah. It's correlating.

Bart van der Meer: [00:37:15] Oh, that's awesome.

Jon Puleston: [00:37:18] That's

Guido X Jansen: [00:37:18] all depends on how well they're taking

Bart van der Meer: [00:37:21] care of dogs.

Jon Puleston: [00:37:21] Yeah. Freaky ones, Christmas decorations. In the old days, you have the Christmas decorations that you get out of your faith that is reading your family for generation. You put them on your Christmas tree. No one sees them about your family. Now, wherever you are. But you, can you take a photograph of me, next year I can't take a photograph of, I photographed that last year.

I needed a new Christmas decoration. So literally Christmas decoration sounds, I've literally just gone through they've climbed about 30, 40% year on year, the last few years, because. I decided to be able to show something new and fresh.

Guido X Jansen: [00:37:54] My mom has the same color for 30 years, but she is not on Instagram.

Bart van der Meer: [00:37:57] Yeah.

Jon Puleston: [00:37:58] That's the thing. Anyway. Diversity. Yeah, presentation.

Guido X Jansen: [00:38:05] What are you going to do research next?

Jon Puleston: [00:38:08] where, I'm particularly, I talk to you about, at the beginning about that. The grumpy cat. Yeah. Yeah. we're in the midst of a whole load of experiments, looking at how to get people to be built, honestly, declare certain things like environmental attitudes and stuff, which really fascinating because some things like you asked you through was, recycling and so on.

If you ask people straight up, do you do these things? What you feel about this? Would you do this in supermarket? Everyone says yes. yes. And it's not true. And so we've been doing a whole lot of experiments, try and understand how to moat Pete motivate people, to be more honest, actually about.

Their behavior and what they would do in different situations using. And I've got a fantastically smart intern working for me. And who splits, you spent the last few weeks digging up these really wacky sort of Mimi type visuals. If I just tried to test it, if I had this fantastic animated meme of this woman.

Going to throw a bag of rubbish into the recycling bin and she swigs it around and it breaks just as this way, but it all spreads all over like that. And we'll use that as a, sort of the start of a dialogue. we always face someone's face challenges, recycling. Tiny bit of humor, literally doubles the amount of people who then prepare to open up a meant that they don't bother recycling.

We'd be doing a lot of fun with that.

Bart van der Meer: [00:39:32] So you're just, you're still gonna keep on basically improving what you found to make

Jon Puleston: [00:39:38] it better. We're looking at, one of the other is, income. Everyone says in the survey that they're rich

Bart van der Meer: [00:39:45] or.

Jon Puleston: [00:39:45] everyone inflates their wealth as wealth questions in their Heights.

It has no correlation with wealth. Almost certainly no correlation with disposable income. So what we'd be looking at using the same idea of behavioral techniques, we've been trying to work out. How do you measure thinking about how to measure someone's disposable income by their behavior? So for example, Did you take a pack lunch into work or do you go out and buy it is a subtle tell about your level of disposable income.

So try and piece by piece to get that type of thing. So that taking that thinking and applying it different way to different challenge. It's something that we're looking at. At doing, we've had lots of spinoff methodologies come out of this work. So I'm just like exploring different.

View complete transcript

Get your cup of CROppuccino

Through our mailinglist we periodically inform you when a special episode goes live, when there are professional events you might be interested in or when our partners have an exclusive offer for you. You'll receive an e-mail roughly once a month and you can unsubscribe at any time.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Some other recordings from

Marketing Insights Event


Some other episodes you'll (probably) also like:

Some other episodes you'll (probably) also like: