Interview with Gareth Harvey: former professor of consumer psychology turned behavioral science consultant

In this episode we talk with Gareth Harvey. Gareth was previously a professor of consumer psychology but now advises companies about how they can apply findings from behavioral science. In this episode he discusses some of his work with retailers showing how they can better frame special offers, use music more effectively and adjust the shopper journey.

Transcript

MAF: [00:00:00] Welcome back to Behavioral Science for Brands, a podcast
where we bridge academics and practical marketing. Every other week we go
deep behind the science of some of America's most successful brands. I'm
MichaelAaron Flicker.
RS: And I'm Richard Shotton.
MAF: And today we're sitting with Gareth Harvey, a chartered psychologist, a
director of consumer psychology at Decide, the UK's oldest creative agency.
Let's get into it. So Gareth, welcome to Behavioral Science for Brands. Richard
and I have made it our little mission to be on the hunt for the best applications,
uh, of behavioral science in the world. And, uh, we are talking to folks in the
industry, people who share a passion, uh, like us for this, uh, practice.
And, um, we were, uh, so looking forward to our conversation with you today.
Um, so if you'll indulge me. I'll share just a little bit of your [00:01:00]
background with our listeners. Correct. Build, uh, and, uh, and we'll go from
there. Um, so your current role is the Director of Consumer Psychology at
Decide. At Decide. Uh, I've heard you talk about how you advise clients on
optimizing their marketing strategies using principles of behavioral science, and
also setting up your own.
Tests in market to support those clients. Before that, you were an aShottonciate
profeShottonr and lecturer at a number of universities, and I've enjoyed listening
to you talk about supermarket design, eye tracking, how consumers find items
on the shelf. Um, this is gonna give us a lot to talk about today. Um, so as we
get into it, um.
Our listeners love stories. Maybe you could tell us a little bit, when did you first
hear of the concept of consumer psychology of behavioral science? How did
you get into this field in the first place?
GH: So I kind of got [00:02:00] into this field probably back somewhere in
2007 ish, I think. So I kind of started to do the way that most students do.
I did an undergrad in business, sort of undergrad, master's. How is that really
annoying student? So someone would say something and I'd be like, why? So
they'd say, if you want to grab consumer's attention, you need to do this. If you

want someone to be able to learn your messages, you need to do that. If you
want to understand how they decide, this is how it happens, and be like, okay,
but, but, but why?
And the answer suddenly to all these questions was, oh, that's psychology. So
after about four years of studying, I thought I've done the wrong subject. So,
um, I told my parents, it's like, yeah, about four years was wasted and they were
delighted by this. And luckily at the university I was studying at, in my master's,
there was a, uh, a lecturer or profeShottonr who specialized in consumer
psychology.
So ProfeShottonr James Ator. And rather than doing, shall we say, more
traditional research, although he did a lot of that, he did a lot of consultancy
[00:03:00] because companies, um. Spent a lot of time with Unilever and as
such, and I thought, okay. Went to chatting with him and said, okay, this is what
I want to do actually.
I want to understand how do people decide what's going on. So yeah,
somewhere around 2002, 2007, 2008, I switched over and kind of abandoned
the traditional marketing and did a PhD in consumer psychology specializing in
atmospherics, in effect, how does the lighting, the aroma, the music, all those
things that people are weren't aware of, how do they influence behavior?
And ever since I started that I've been kind of doing commercial work and
building on it and trying to find experiments where I can test in real
supermarkets because it's lovely testing in a lab environment. Lots of nice
controlled things, but uh, it's a little bit sterile when we try and roll those things
out in the lab, outside of the lab.
It doesn't always work the way that the papers say it will. And that's what really
excites me.
MAF: I. It's such a interesting, uh, story. Thanks for sharing it, [00:04:00]
because so much of finding excitement and interest in academia is like
understanding the why behind it. But take, but getting out into the world and
applying it in real supermarkets with real brand is a different level of
application than.
The great work that happens when you write a paper and you, and you publish
results, but taking it to that next level, uh, is certainly gotta be exciting.

GH: Well, what really surprised me is most academics aren't excited by that.
Their systems, their incentives are to publish papers and actually they design the
most beautiful experiments that work in a controlled environment.
But they've, they've achieved their objective. Not all of them, but a lot of them,
they don't want to, they're not interested in trying to take it into the real world.
Whereas for. I'm smugged when I walk around a supermarket like, oh yeah, I
did that. That's my work. I can actually see it. And it's, I kind of joke, it's a bit of
a game.
I, it's a virtual 3D chess where, how can I influence people without 'em
noticing? That's what really excites me and [00:05:00] going through and trying
to work out what studies in the lab setting have a chance at replicating. We're
talking about maybe 10% will. It's trying to work out which ones will and
testing it before we get there.
RS: And are there any lab studies that you've seen recently or, or, or that stand
out that haven't replicated when you've taken them from that? Uh, academic
setting to the, the supermarket.
GH: The classic one is not one I did, and it's one where they did the lab study
and then they tried to recreate it and they did a fantastic job and it was looking
at their food nutrition labels in France.
So there are four different sorts. So we've got the ones which are telling you sort
of A, B, C, and D. There's somewhere they've got bar charts somewhere.
They've got numbers and they test it in a lab setting and they tried to work out
which one worked. And in effect, they gave people a catalog. So they gave
people a catalog, they had to select what food they were doing.
They had about three and a half thousand products, but each one had those nutri
stores on. They found out the one system worked. They then replicated it, I
think it was 20 supermarkets across France. [00:06:00] 10 had them on 10 were
control, and they tested different ones on different products and they got the
same result.
But the effect size of that experiment was 17 times small in the lab. And it's
certainly a huge case of like when you're making public policy decisions based
on what you think of these lab-based studies are showing. A 17 time and a
smaller effect size is huge. If you are a commercial organization rolling
something out and you were expecting a certain return on investment, you're
never going to get that.

And that's a terrifying thing for retailers.
RS: Yes, having an effect is not enough. It's got to be cost effective and there,
and there are many things, if you, uh, reduce the impact by 17 fold that are still
gonna be cost effective. So that, that's, that's, that's fascinating.
GH: Yes, one of Pierre Chaney's studies, but it's, when you look at those sort of
things, papers will publish things and they talk about it based on, oh, it's a
significant finding, P equals 0.01.
But when they're so overpowered, it's very easy to find those effects. [00:07:00]
Actually, what we need to be more focused on is what those effects are. Does it
matter? And that's especially when we throw in extraneous variables like you've
got screaming toddler next to you, you're stressed, you're at the time pressure.
That certainly reduces all of these things even further. I. And it's so important,
which is, this is why I love doing the field trials. When
MAF: you're advising clients and you're looking for new studies or studies that
you, you think would apply to the business case, how do you go about deciding,
well, you know what, these are the ones that are worth testing in the real world
and these aren't given that there could be such a difference.
Just, just how could you approach
GH: it, even if it varies. That used to be the main job, main role I had for one of
the companies I worked for. Um, it was basically reading all the literature and
work out which ones actually look cool and worth trying. So the first thing you
do is actually, is it something that will apply to a bit broad, uh, broad set to the
population?
So many studies will focus, say like, oh yes, [00:08:00] extroverts who are
under the age of 30 will behave in a certain way. And it's like, great in a lab
study, I can do that. Actually in a supermarket I'm influencing trying to shift to
population level. So anything where we kind of nailing into that level of detail,
we kind of ignore.
Then the next thing is actually going through the method and saying, okay, did
that work? Is it the sort of thing where I don't see any obvious flaws? And then
it goes down to look at the effect size. I'm not looking at medium, medium
effect size. I'm generally looking for large effect sizes. So we kind of ignore the

P value and it's large effect sizes may be medium at a push if all those criteria
are met.
First thing I do is I'll recreate it in a lab. Just check, Hey, if I create the stimuli,
will it work in a lab? I might try three or four different versions, see if I can
boost that effect size. Once I've got it there, we then actually roll it out into one
small supermarket or a controlled one, test it about that.
So there's a number of supermarkets I used to work with quite closely. We test it
in that one, and then we'd look at rolling it out as maybe across a county in the
UK or Canton in Switzerland. [00:09:00] If we got it at that level, then we'd
don't go to a national rollout, but it took us maybe. Six or seven different studies
before we'd actually get rolled it out.
So a great one is, uh, looking at different aromas in stores. We found the study,
we found different aromas looking at chocolate. We got the intensity we wanted
in the lab setting. I think that took us about seven different variations and trying
about where we put it. So there was no put, no point in putting the smell of
chocolate in the chocolate aisle because, well actually you've got a giant wall of
chocolate that's a fairly big cue to make you, uh, to prime you.
You want to put it on the door as you went to the store because. Oh yeah. Kind
of fancy the idea of some chocolate today. It's something they weren't thinking
about, the playing around with that. Intensity levels. We want it to be something
where if I tell you, oh, can you smell chocolate? Oh yeah, I notice it.
But if I, it was a case of you would, it wasn't in your face. We're trying to get
that balance correct. So we try around that and then we tried it across, you
know, supermarket in North Yorkshire, tried it, getting that intensity because
the [00:10:00] doors keep opening. Oh, okay. That's washing away that smell.
How do we play it to where do we move it?
So we applied across that. Got it. Working in that one supermarket before it
ended up being rolled out across, um, all of us, all stores for one brand in
Australia and South Africa.
MAF: It's fascinating. Is it fair to summarize that your approach, it starts with
the academic study, then a lab study, then one store, then multiple stores, and so
you're continuing to validate where possible, what you're learning and test
against it.
Assuming

GH: the client has my dream budget, that is the approach we're going for. Got
it. That's the recommended approach. Yeah. Each stage it's a case of learning.
What do they do? Okay. How do we make it better? Is the effect falling more
than we're expecting? Either some extraneous variables we need to be aware of.
So it's all about trying to play around with those sort of things and working out
what tweaks do we need to make for the next study.
RS: Brilliant. Um, one of the things that we always like to chat about on
behavioral sites of brands is practical uses of [00:11:00] behavioral sites. Are
there any examples of brands that have applied this field to, to great effects you,
you particularly admire?
GH: So on the retail side of it, when we're looking at it, I think it's a really
interesting one because supermarkets have become quite sterile in recent years.
So accountants have got some involved and they've said, these stripped out all
the emotion outta these sort of things. So if we go back to the first
supermarkets, they were just warehouses.
Then suddenly people realized like, actually. Shopping shouldn't be a cho. The
more enjoyable it is, the more you go and you at certain supermarkets, they did
a great job, whereas they recreated the feel of a market. So in the UK we have
boots and they do a lovely job 'cause it feels like there's a delicate tester.
There's a fish monger. So they do this lovely story. But a lot of the
supermarkets, in recent years, we've just seen Tescos. They've removed all of
these sort of individual stores with their stores. We've got the data. We know
people spend more when you kind of subdivide it, because rather than sp feeling
like you've spent 70 FRAs in one store, it's like, [00:12:00] oh, I only spent like,
I dunno, whatever.
Is it 1212 FRAs in the lingerie or there's, I've spent another one in the, in the
fish markets. It changes the way you think about it. It's this mental copper, uh,
your carpa copper mentalizing. Yeah. The word I'm nearly getting to. Yeah,
you, you subdivide how you're spending that money. But we've actually had
that taken away because it's more expensive for the stores.
And that's a really interesting story, seeing which stores are still trying to create
the atmospherics, they're trying to create this story. So I, I love seeing how
they've done that.

RS: I mean, that's, that's a fascinating one. 'cause the idea of, uh, this com
compartmentalization mental accounting. That's, you know, a, a longstanding.
Um, set of experiments would back up that principle. Why is it, do you think
businesses have ignored that? What's, what's driven Tesco's and the like to
remove those,
GH: uh, mini startup businesses? It's a really hard one to show on a balance
sheet, so it's a straightforward case of, on all Tesco stores, as far as I'm aware, in
the UK anyway, they've done this way.
So it's a case of [00:13:00] they've not got experimental data to be able to show
like, oh, when we've taken this away, we can see there's been a fall in overall
span. They can look at the club card data of and see the individual receipts.
They can see where people go, but it's, we don't have a before and hour after.
There's no, there's just no data to look at it. And it's a case of when they're under
increasing pressure. As far as they see, they're just seeing it as a cost. They're
not seeing it as a benefit, and that's a real challenge to play around with.
MAF: It occurs to me that, uh, that we're always, uh, edging between eking out
profit, quarter by quarter, uh, to to, to try to hit our goals and the business' needs
and creating enough meaningful.
Uh, evolution for the brand to actually make bigger strides forward. And that's,
uh, it's not just a classic ball battle between CFO and CMO. It's also this idea of
how much risk will a brand be willing to introduce into its model in order to get
GH: more advancement. I [00:14:00] think it's worse than that. It's not just the
balance between CMO and CFO.
It's within the brand manager with one section and another. I've done a project
for some supermarkets and I was, uh, a category manager from the one retailer,
and we were able to demonstrate, I think it was a 7% uplift in sales. They refuse
to share that research with other parts of the organization because their
performance is measured on what uplifting sales they get versus the rest of that
store.
So we did this research, really proud of it. Quite smart, if I'm honest. Then
about four months later, the same supermarket, same client approached us to
ask, they said, can you replicate the project that you did over there? And it's

like, um, yes. Could you not just borrow it? I will do it again for you. But NDA
is their things.
Okay. Can you not ask them to share 'em? They're like, they were check, so
they paid to have the same project debt again. Just because they wouldn't share
internally, and I'd love to tell you that's a one-off. It has happened [00:15:00]
numerous times across different stores.
MAF: The bigger takeaway for our leaders that are listening is, do we have the
right incentive set up for our staff to create the conditions for knowledge share
and for and, and, uh, the incentivization to help the overall company win, not
just individual departments
RS: to win.
Don't incentivize one thing and then expect a completely different result. I think
that's a, yeah, that's a fascinating, Eric.
GH: It's a, it's an obvious one, but the idea, yes, competition drives things, but it
also hinder sharing. And actually that's fine to a certain extent, but when you are
paying. External people to come in and conduct research and find what works
and they refuse to share it.
That's a real worry.
RS: So the closing down of some of these shops within a shop is an example of
a retailer misapplying ideas from psychology. Are there any areas within retail
where you've seen someone very positively [00:16:00] applying a psychological
behavioral science principle? So the other one I
GH: go for either is Boots.
It's a classic example. And so if you're not familiar with Booth, they are one of
the more upmarket supermarkets in the uk. It's more a northern one, kind of
comparable to waitress. If you're in the south, uh, of England and they take you
on most of the stores an emotional journey. And we always talk about mood
being so important, so.
Often you kind of say retail therapy, you're in a bad mood. I'll buy something to
cheer myself up. Actually, the data doesn't quite back that up. Generally
speaking, when people are in a positive mood, you buy, you spend about 12%

more. So if you're in a bad mood, you might buy yourself a box of chocolates,
bottle of wine or something, but you buy one specific item.
If you're actually in a good mood, you just spent globally more, which is a
really important thing to look at. Boobs. It's an enjoyable experience. It does
feel a bit like a delicate Tesla. It's actually the atmospheres. It's all focused on
the consumer experience. Yes, they're slightly higher end products.
There is a price [00:17:00] premium for it, but they have this experience where
the customer comes first. The other thing I like is it is a bit of a journey. If
everything is duly positive in the entire store, you kind of could become
accustomed it. You normalize it and you don't really notice it. You've got to
have these highs, and I'm not gonna say lows, but more normal parts for you to
actually suddenly get to, whether it's the, in my case, the lingerie, or you get to
the bakery and you suddenly realize, oh, right, I'm going for it.
You need to have those highs and lows because otherwise everything just feels
normal.
RS: That's interesting. So that's almost like the a story arc for a classic novel.
You know, you don't have everything at the same hitch all the way through.
You have. Tribulations and then you have achievements. If someone just
succeeds in a quite good way all the way through, it'd be the most boring novel
you could, you could imagine.
GH: Yeah. Uh, it's gives bit of a challenge for the brands because where do you
want it? You always want your brand to be on the really positive things. This
idea of that misattribution of arousal or the [00:18:00] emotional transfer there,
it's like Zillows is the idea that if it's in a really high point, those positive
experiences get transferred over to, uh, to that brand.
It's why I used to do a lot of work. Um, coming up with a playlist of the music
stores would play. We tried to make sure, because if you play music and it's
pretty much a global phenomenon, music's in a major key sounds more positive.
If you've got music in a minor key, it sounds more negative. We use this in lab
based studies to induce mood, induce moods we use in retailers to make sure we
don't have too many songs in a minor key when we play it.
I'll also think very carefully about the tempo of music selected. So, for example,
higher tempo music, more energy generally speaking, that gives us more
impulse purchases, lower tempo music. They'll walk slower, they see more, but

they're not going to make as many impulse purchases. So we're trying to get that
sweet spot there somewhere in the middle.
RS: I, I, I, I love that. Uh, I think that particularly the point about the minor and
the major key inducing positive or or negative [00:19:00] mood, uh, I've done
some work last year, maybe year before with um. Um, Chris Davies and, uh,
news uk and we showed a group of people, so 800 plus some promotional ads,
so it might be, you know, pizza for 10 pounds or a bottle of wine for a, for a
well-known brand for five pounds.
And then we asked 'em to rate how good value they thought the deal was. And
then later on we asked 'em whether, what, what mood they're in, positive,
negative, happy, sad, relaxed, stressful workup stuff. What we found is that
when people were in a negative mood, yep. 60% of them thought the offers and
promotions were good or great value.
Whereas when they were in a positive mood, that number went up to 76%. So
you get this 26% improvement in, um, valuation of exactly the same offer if
someone was in a, in a positive mood. So. Hadn't ever thought of music as being
one of the tools you could induce that positivity about. I think that's a lovely
idea and one that could be applied [00:20:00] by someone who runs a single
mom and dad store all the way up to a Walmart or a a Tesco's.
That that's absolutely lovely. The other thing I'd say
GH: special offers, promotions. They're probably one of my bug bear, shall we
say, supermarkets giveaway. It's hard to calculate. We set, we think it's
somewhere between 15 and 20% more margin than they need to for the simple
reason. Supermarkets generally tell brands that if you want to be e listed here,
you're going to have to put a special offer for us.
And it's a case of they don't really say what it's going to be because they know
they're not paying for it. It's being imposed on the brands. Brands sort of panic
and say, five will come up with something and yet they have no clue what sort
of offer works best in different situations. How do we use a buy one get one
free?
We'll use that because it sounds good. People know what that is. 50% off, 50%
more, buy two for five pounds. There are so many different mechanisms there,
and half the time when we've [00:21:00] analyzed the data, we can just sort of
just show that different offers will work in different situations. Some of the time

though, it's just having a benchmark or a sign on something on the shelf to grab
people's attention to notice it.
That's what is the mechanism that's apply. The other thing is trying to work out
what's your objective. If you look at something and it's a price discount. A lot of
the time, people would just stock up the, so the brand switching isn't, or the
increase in sales is not caused by brand switching. Maybe 15% is, most of it's
people who would already buy that product.
They'll just buy it, stock it up, and they'll save it for, for future. So it doesn't
actually achieve anything. Particularly, uh, if you look at buy one, get one
freeze, they really don't work particularly well on things like cakes or treats,
because people look at it, they, oh, they'll notice it, but they may leave the
second one, but it's free.
But they're like. I really don't want, or I do want to have two chocolate bars in
the house, but I'm not going to be able to cope with it. I will eat them straight
away so they end up leaving it. So if you're selling bleach, buy one, get one free.
A great mechanism selling cakes and treats [00:22:00] less so. So there's so
much here we can play around with.
It's fascinating,
MAF: uh, and, and I, and applying thinking about the types of offers based on
the type of consumable that, that, uh, that people are buying makes, uh, makes
all the sense in the world. Have you, have you created some simple rules or
some simple applications? So the one that you started with here makes, makes
clear to me something that may.
Get consumed quickly or be more of indulgence. Getting an extra one free may
not be as exciting as being able to stock up on bleach. Any other rules like that
that in your mind, just a general rules of thumb
GH: that people can think about? The first thing to think about is it comes back
to the most basic, tend to behavioral science, prospect theory, losses loom larger
than gains.
We all know that. We tend, when it comes to creating a special offer, a
promotion, we get that wrong all the time. So for example, sign up to our club
card today and get 15% off. That's how it's always phrased. [00:23:00] Just flip
it on its head, don't sign up today. Your shopping's gonna cost us 15% more.
And it's like, oh, ah, yeah, I don't think I'm gonna be doing that one.

So it changes how you, how people think about it. Likewise, when it comes to
what works, is it 15% off? Uh, well, I'm saving money. Or 15% more. It's all
about framing it in terms of where prospect theory comes from. Uh, it's
something most marketers are aware of, but you just don't think about it. So just
bring it to mind whenever you're coming up with these stakes.
RS: And I guess if you, if you gotta mention losses, the other thing you could
think about would be being much more clear about when the offer expires and if
it's gonna expire in the next few days. Making sure people know, because there
might be an element of people can go well. I don't wanna go through the pain of
paying that for that stay.
I, I might get it next week. We've
GH: done something a little bit different with that. So yes. Say's, got set date,
but if it's say 10% off, don't have a hard deadline on it. So 10% off once it ends,
okay. [00:24:00] Maybe for the next week have 7% off and actually have this
tiered one. 'cause what you find out is when you have that hard deadline.
People who come in late and didn't realize it was an offer. They look at it like,
oh, for God's sake, they get really annoyed and then they, they, it's a variation of
reactance, so they avoid buying the product. Whereas if you have this tiered
ending, we actually find that sales end up increasing overall because at the end
of the offer, you've still got, people will continue buying it, whereas if you,
whether the hard offer people get, let's be polite, cheesed off that they missed it
and then they boycott the
RS: brand.
Yeah. So you might either want, as you say, tiered offer or. Make sure
whenever the offer expires, you expunge your mention of it from the store,
because I think that point of the irritation of paying, uh, more than others had
done previously would be, would
GH: be terrible. We tested that and even actually find, we found it was much
better to actually have it go.
I think we went from 15, 10, 7 and five and we, what we actually did, we
finished the offer slightly early, so it wasn't that the offer ran for longer, we just
gradually declined it from, I [00:25:00] think it was a week. Five days, two
days, and we went that way. And that was better than just getting rid because
other customers mentioned it to each other.

People remembered and things. So yeah, people talk.
MAF: And just a clarity point on that, Gareth, were, were do was the entire.
Offer plan made aware to the customers at the beginning? Nope. So they didn't
know it was gonna go down to 10, 7, 5. They just showed up and the number
was lower
GH: and lower over the course of the time.
So we, we tr we tried both ways, so yes, it was better when we did the 10, 5, 7,
1, uh, or, well, 10, sorry, fifteen, ten seven. That would, it was better when it
was made public in advance, but it was still better having it when they didn't
know than they're just having a higher end at 15 to zero.
RS: So those are some lovely examples in retail of applying psychology to
discounting and, and generation of positive mood.
Outside of retail, uh, outside of supermarkets. Have you, have you seen
[00:26:00] any particularly impressive examples of HA science in
GH: action? So the first one I was gonna talk about was actually kind of
sticking with that high-end feature because. Like I say, I designed supermarket
layouts. I love seeing technology in action and how things are changing.
So in Switzerland, if you're into your avocados, they now have machines that
am negro that tell you how rare your avocado is. Just so it goes from it. Not
particularly behavioral science-y, but if you're into your avocados, quite useful.
But the one I want to talk about is actually robot shelf stackers. So there's a
number of companies out there who have introduced these things.
So Bossanova, Simba are doing these sort of things. And the idea is. They are
intended to make sure that all the products are on the shelf when they're meant
to be. Or from a consumer point of view, the labels are facing forwards. The
behavioral science challenge for these brands is how do they interact with
consumers because their big, bulky things, um, supermarket shelf is about
somewhere between six foot, six foot five, two meters ish, [00:27:00] uh,
depending on where you are in the world.
These things are moving around and if a consumer sees them, they sort of like,
uh, I'm not sure what that is. They walk away from them, which ends up from
being a revenue saving approach for the supermarket, has a negative one. I'm

loving seeing how the design is to actually look at these things. So one of the
things we're kind of looking over, this is actually the shape of them.
So in the supermarket, the big thing they're always concerned about is
dominance. So if you look at them, the, the height of the ceiling in a
supermarket is generally quite high to make it feel not very claustrophobic. The
width of the aisles is really wide, again, to make it feel more, like, more likely
for you to explore.
These robots are being designed so they narrow in, so it kind of gives the
impression that they're narrower than they are. We look at the, uh, the
affordances, so it kind of has curves on them to look where they go. They've
tried some really silly things like putting goggling eyes on them to humanize
them.
There have been examples where they've made them humanoids. They're just
[00:28:00] less functional, so okay, we'll make them appear more friendly. So
it's all about trying to look at where they go for them. We're playing around
with ideas about what noises they make, because if something, this machine
moves next to you and you're like, I wasn't expecting it, it's off-putting.
So is it, is it like a high pitch post sort of beep, beep, beep. They've tried around
with more low pitch ones, which again, people get really offput 'cause it's that
negative noise of be, be, be that lurry reversing to kind of a back sound. So
they're really playing around with some of the things they're looking at, those
sort of, uh, examples when they're playing around and even the speed that they
move at.
It sounds a really obvious behavioral science question, but how quickly does it
move? It's got to be predictable. They have to move in this logical way that it
works with our human brain. Brain's a prediction machine. I need to know what
this thing is going to do. So they're really looking at how they're gonna be
moving in this logical way.
And the final question they've been talked about is, how do we frame or sell this
to consumers? Because we're [00:29:00] seeing more and more automation, it's
taking away human jobs. So what do we feel when we see this? And they think,
so some of them have got signs on them saying, I sports bills, I help keep you
safe.
They're trying to give that positive benefit to them. Others of them are, they're
just telling you them. It's with messages such as, and make your shopping

easier. There's never going to be a product that's sold out. So it's all about how
do you communicate this because it saves the supermarkets a lot of money.
But if, if shops don't react positively, positively to them, we're gonna be in a bit
of a negative head.
MAF: Fascinating. Uh, what a fascinating, uh, set of challenges when you are
trying to advise the supermarkets on how to make this technology more
amenable to consumers. Where do you put your, where's, where's, what's the top
two or three things that you are concerned about as a behavioral
GH: scientist, a consumer psychologist.
So we started to work on this probably about 10 years [00:30:00] ago for these
brands. And the first thing we did was the shape of them because we had one of
the concept ones delivered to us and it was like, oh God, that's huge. Because
they designed, it was designed by an engineer. It was lots of hard edges. It was
very, it was basically a two meter by, I don't know, something like 50
centimeters straight up.
So the first thing we tried to do is, as soon as I see that, I can't see anything past
it. It just blocks the viewer down the rest of the aisle. Now, when we design
supermarket shelves, we often place the target items, the things the shoppers
most want to buy about a third of the way down the aisle, because that means
you can see it from the end and you can see it down the end.
It encourages you to walk down. If you walk down a third of the way, you're
probably gonna keep walking down the rest of the aisle, but put it in the middle.
You may not notice it from the end, so you don't go down. If I put it a quarter of
the way down, you walk a quarter of the way down and you go back to the edge
of the aisle, the, uh, the power aisle.
So it defeats the whole point. If I had this big [00:31:00] giant robot shell
stacker halfway down, I can't see anything beyond it. So it was a case of can we
smooth out the curves? How do we make it so the bottom needs to be that big
'cause the center of gravity has to be there. That's where all the motors are. At
the top, there's just a lot of sensors, so let's make it smoother, make it more
human.
Then let's see about trying to make it leaner to a certain angle so it actually feels
less claustrophobic. So that was the first thing we ended up playing around
with, and then we played with the noise and the eyes. We wanted to make it feel

less intimidating because no one wants to come up. I dunno, Robocop down the
aisle.
It's not the image you're going for gully eyes, it's childish. Children loved it.
Suddenly it was a case of, oh, it becomes fun. We change it, change the one we
are working with from a metallic gray to a nice, fun, vibrant color. Because
certainly everyone's like, whoa, what's that? And again, it's all about I wonder,
whoa, what's not that?
Rather than a, uh, I'm not sure what's going on here. So it is a really interesting
one to change what's going on, [00:32:00] how the people react to it.
MAF: And
GH: outta
MAF: interest. What was the, uh, what were the KPIs or the analytics you were
looking at to tell how successful your tests were being? Was it interviews with
people after they walked by or if there's something more quantitative
GH: you were using?
So as a psychologist, I don't generally like self-report measures because people
have this annoying habit of lying to me. So it was a case of trying to work out.
When we had them, we were measuring through CCTV and cameras, how many
people were going down the aisles. Actually, we were comparing before and
after to suddenly notice is there going to be a decline.
We were actually looking at straightforward sales data. If we had something
here, were we seeing a decline in people buying things, not just next to the
robot, but let's say about four or five meters, either side of it. So we're really
interested in looking at those sales data. Yeah. I love
RS: that as a 'cause It's very easy to say, oh, well we don't like self-report, but
we're just gonna have to do it.
'cause it's very hard to get metrics. But thinking about the problem [00:33:00]
creatively, that idea of, well, how often do people go down the aisle when the
robot's there? That's a brilliant bit of, uh, of, of observed data. That's a, that's a
lovely, and it's,
GH: it's also worth looking at what data you're collecting already.

So the, the stores already had all the data to be able to track sales. All, most
major retailers will have some form of CCTV that is now digital, so we can
create heat maps from it. So we can create heat, heat maps, we can create flow
maps so we know where people are moving. We can certainly see, actually,
we're now seeing people are skipping this out, they're walking down to the next
one.
We have the existing data, so it made my life easier working with that data
source rather than having to install my own cameras.
RS: Um, so we've talked a lot about supermarket, but there, are there any, um,
areas outside of the supermarket we've seen behavioral sites and action that you,
you, you, you think other others could learn from?
GH: So, one of the more interesting ones, we, we did some work recently and it
was with Behringer Ingelheim. Now you have probably never heard of them.
Um, I will be honest, I had never heard of them [00:34:00] either. Yet they are
the largest independent pharmaceutical coke company in the world. And they
are huge, huge amounts of work on animal animal care.
So pets, but also tumors. But I was really interested in working with the, uh, vet
team there, and they've got some really interesting challenges, partly because in
the EU prescription drugs, you cannot advertise them direct to consumers. So all
their marketing is going. To their, uh, the trade press. So we've got a B2B
marketer, so it's a pharmaceutical company, marketing to vets, and they were
trying to work out, well, what challenges have they got?
And the problem they had was that experts are working out what's wrong with a
dog. In some ways, you can say they're more skillful than a doctor because the
dog will come in and it can't say By stomach's hurting, my heart's hurting,
whatever. All they're doing is just saying it's a dog. But all their training has
been in working out what's wrong with it.
Certainly, they have to then convince the owner that you need to pay for
[00:35:00] this drug, or more importantly, they need to convince the owner that
they need to pay for a test. Now, if you're taking something like mitral valve
disease, a really common disease in dogs, often this is detected when an owner
brings in the dog for, let's say, vaccinations or something routine.
As far as the owner's concerned, the dog is completely healthy. But suddenly
they're like, oh actually we need to do an ultrasound now an ultrasound is going
to be costing a couple of hundred pounds and they have to convince an owner to

pay for something when they can't see anything wrong with it. So we've got
cognitive diShottonnance going on and the vets, they have no idea how to train
them so they suddenly baring Ingelheim worked out.
We need to educate the vets and some of the first things they had to do was
change the language they spoke and vets are constantly, will use the language
they're familiar with as we all do. Certainly they will be using these really
technical language, and it was a case of the owner has just switches off.
They don't understand it, they get intimidated by it. The other thing they said
worked out, they had to [00:36:00] make sure they didn't go too far because
sometimes vets would use language, which they thought would be easier for
people to understand, but people got the wrong impression. So they might say,
oh, well we detected your dog has a big heart.
I'm like, oh, yeah, yeah, yeah, he does. It's really nice. And it's like, oh, okay.
That's not right. So. We simplified it. It's like, okay, your dog has an enlarged
heart, okay. And it's trying to work out well, how do we go around and change
these sort of things? Working about going back to things like, well, we need to
show them this.
People don't quite understand it. So we would show them things like, um, the
CT scan and no one has a clue what the CT shows, but it's a case of, it's
reassuring. It's providing them the evidence. And it's also talking about working
out which order they've had these conversations. So when you're trying to
persuade people, one of the worst things you can come across is something
called the inoculation effect.
So the inoculation effect is if you lead with a bad argument, people have this
tendency to think that whatever comes next is also a bad argument. It's kind of a
[00:37:00] self-defense mechanism. And the problem you kind of had is
sometimes the vets would start off and they were. In fact they were selling their
health plans.
'cause they'd be saying, oh this medication, because it's a chronic condition, it's
gonna be expensive. They'd say, well if you have this health plan or if you do
this plan, whatever, we'll be able to help with your medication. And then they'd
lead in with a big thing. People were just switching off. 'cause they're like, okay,
here comes the sales pitch.

They're going with this. And even, and then they went with this big information
that came next. They weren't really paying attention to it. So just getting them to
think about the order they communicated was so important. That's, that's
RS: fascinating. Um, I did a study a while ago, uh, where we created this
fictitious brand.
Yeah. So we mocked up, looked beautiful, and it's called Black Sheep Vodka.
And half the people we said it was award-winning, refreshing, satisfactory,
weak, and vinegar. Okay. And the other half we said it was RY week,
satisfactory, refreshing, and award-winning. [00:38:00] And the group that
heard the positive words first, they rated the beauty, the design, their perceived
taste.
So about 10 or 15% better than the people who heard the negative. Words first.
GH: Yeah. So we've got that primacy of recency going around. Yes. But the
other, the other thing you've gotta be a careful here is you're talking about
something, their pride and joy people. Sometimes it feels like people love their
pets more than they do their children.
When you tell them the bad news, something really emotionally salient, do you
get into this attentional, bleak territory where it's a case of they switch off
everything that comes next for the next couple of seconds. It's so emotionally
overwhelming. They weren't remembering things as well. So if you told them
really bad news and it was a case of the vets were thinking that, I've told them
this, I've told them what they have to do, they just wouldn't remember that.
So certainly thinking about, okay, how do we structure this information? If
you're giving this really big emotional news, just know that whatever comes
next for the next minute or two, they're probably not [00:39:00] going to
remember. So it's thinking about how we structure that information there. That's
fascinating.
What, what did you call that? Did you say emotional blink? Attentional, blink.
So it come, it comes from the, it's a visual saliency phenomenon from Jane
Raymond originally, uh, Jane Raymond and Kim Shapiro. And originally they
were doing it with flushing simul. So you'd flush lots of words on the screen and
you'll be told to look out for, let's say, I don't know, red, the word red.
And you'd see it, you'd feel quite smug, but you would forget or you wouldn't
notice any words that came asterisk in the next three or four words that came

after it. The reason it's called the attentional blink is if the next words that came
after it was a really emotionally salient, say your name, it's strong enough to
jump outta this phenomenon known as the attentional blink, and emotional
saliency things cause this blind spot in our memory.
And when you're delivering bad news, like your dog might have a chronic
condition, that's a real problem here. It's a case of trying to make sure that we
know there's going to be this sort of [00:40:00] pause in what we remember.
RS: So with the inoculation effect, the, like, the recommendation, the, that is
quite clear.
Think about the merits of your arguments and make sure the strongest goes first
with the attentional blink, what, what would you recommend that vets did?
GH: Differently. A couple of things. One, first of all, you know you're gonna be
delivering this big emotional things. Just know that they're not gonna remember
that sort of thing.
You've got that two minutes where actually this is the point where we're gonna
be telling them information, which is we're being reassuring. We're being
comforting in many ways. We're just trying to be liked at this point because if
they like us going back to Sini, they're more likely to actually, hopefully pay
attention to what comes next.
After those sort of things, we're gonna reassure them. We had to make sure that
information and the factual information we gave them, we gave them multiple
times. We would make sure it was done there. We'd also make sure the nurses
or whoever was on the reception desk also gave them the information to get.
So we had [00:41:00] that repetition was just so key. Um, the certain predictable
things like mitral valve disease, we had pieces of the paper so they could take
home and remember those sort of things. Because having those instructions
when you are a little bit calmer, it made life so much easier. So helpful.
MAF: So helpful.
You know, part of what we do, Gareth, is uh, we create show notes that, uh,
folks can read more and, uh, do more research on the topics we talk about in the
show. Maybe we can ask Richard and I will collect from you after just a little
bit more follow up, Richard, that they can go and read you reference some of
the original work here.

Maybe we can get that and share it along so people can learn more.
GH: Yeah. I'll tell you one of the other things I can do. When we talked about
special offers earlier, I've got a set of 15, 20 rules on how to run special offers
under which circumstances they work when they don't. All linked back to,
should we say, they're all linked to uh, academic research, but we've tried to
make sure these are studies we've replicated so we can say, okay, we've got
some data behind it that [00:42:00] shows it makes a difference.
So I can give you a link to that and we can try and share that.
MAF: That'd be lovely. I know we have lots of people that are always looking
for things like this. That'd be lovely. Thank you very much for that. Thank you.
No problems. Richard, should we ask before we come to an end, uh, you, to
give us a little summary of some of the key points of what, uh, we spoke about
today?
We had a wide ranging discussion.
RS: Yeah. Very wide ranging, but I, I think three, well, lots of interesting
things, but if I had to pick three, I think the initial area we talked about was
fascinating. So this is the idea that there is often a slippage between, um, results
you find in an experimental lab setting.
Then what works in, uh, the real world? So I think Gareth was positioned this
brilliantly to not to just reject academic findings, but if you are a brand, take
some academic findings as hypotheses and then make sure you are testing them
in real world settings to make sure that there isn't, uh, a slippage in what, what
we're looking to apply.
I thought that was fascinating, this [00:43:00] mantra. Make sure you are the
testing in the, in the real world. The second bit, and maybe for selfish reasons,
this is what I, I focused in when we were talking about tactics in, in retail
environments, was about the power of positive moods. So I think it, Gareth
mentioned there's lots and lots of evidence that people are in a good mood.
They're less price sensitive, they're more likely to spend, and then he's, uh,
suggestion for brands that tap into that. Retailers tap into that about playing
music that's in the, uh, major key rather. The mind I think is a lovely example
because it's so easy to apply. Most people are playing music. Make sure you use
the one that puts in a good mood.

GH: The nice thing about that is it's a study which has been shown to replicate
across cultures. So it's not one that's just tied to North America, Europe, it's
been shown across Southeast Asia. It's been shown across, uh, Africa as well.
Because so many of our studies are weird, Western educated, so forth, that one
appears not to be.
RS: That's a real, uh, one with lots of mixed support. And then I thought the
third and final thing, [00:44:00] well, I probably, you know, the, the, the two
bits about the vet I thought was super interesting was this inoculation effect.
The idea that if you open with a poor argument, it isn't just that argument that
will be dismissed by the listener, it will negatively impact what follows.
That was fascinating. And then that relates area of the attentional blink, which is
completely new to me, that if you deliver some bad news, uh, don't expect
people to remember what follows. So use the next minute or so for reassurance
and general emotion comforting, rather than continue to convey facts.
Make sure that key points are repeated rather than just delivered once. So I
think, yeah, lots and
MAF: Gareth. Before we come to a close, we always have a final question for
our guests that we'd love to hear this. Can you're welcome to extend [00:45:00]
beyond consumer psychology and behavioral science. What are you reading?
What are you watching that's got your attention right now, that's enjoyable to
you right now? What's an indulgence in your, uh, that that's really been
enjoyable for you?
GH: It is gonna sound really sad, but the outside of behavioral science, which is
far too much on my reading, which is quite, we're gonna skip over, is the
Economist.
So The Economist I find is probably one of the best things because I can't keep
up with the news. Try to read a newspaper every day is something I just
wouldn't be able to cope with. The Economist, whether it's the podcast or
actually just reading the magazine newspaper, it comes out once a week. It's got
the right balance of humor, the intellectual rigor, and I just like their outlook on
the world.
So it allows me to. Pretend that I know what's going on in the world at least. So
the podcast and the magazine.

MAF: Yeah. Richard and I are avid readers in The Economist, so it's funny that
you mentioned that. Well, that we're a good company here,
RS: so, yeah. Um, unfortunately, yeah, the, I think the [00:46:00] superiority of
the economists, uh, journalism is becoming more and more pronounced as
major national newspapers are, are losing online advertising revenue, so losing
advertising revenue to online providers.
I think you can start to see the, um, the problems occurring. The journalism
looks like it's being knocked out very quickly, whereas the Economist, because
it has such a deep pocket because of the subscription revenue, that the quality is
as good as it was 10, 20, 30 years ago.
GH: I think if you've ever read the The Economist style Guide as well, it's not
just the quality of the research that goes into it, but it's, it's the quality of the
communication.
So actually, if we talk about the behavioral science, it is written with one voice,
the, at least in the magazine itself, they're all anonymous. There's no by light.
And actually you've got that one tone, that level of humor. And I really think
that sets it apart from reading so many other newspapers.
MAF: Never considered that.
Totally agreed. It, uh, it is known in America as, uh, one of the only
publications that does [00:47:00] that, and it's, uh, it's, it really makes a big
statement. It's very cool. We wanna say thank you to Garrett for all that you
shared with us today. It was an amazing discussion. Uh, as we always do, we're
going to wrap up all of our, uh, conversation on our website, the consumer
behavior lab.com, show notes, links to our, the YouTube video, and all of your
favorite streaming channels are there for our listeners.
Until next time, Garrett, thank you for everything. Thanks for being with us. I'm
Michael Aaron Flicker. And I'm Richard Shutter. Thanks for tuning in.
GH: No worries. It's good fun.
Auto: Behavioral science for brands is brought to you by Method One. Method
One is a team of modern marketers that practices the art and science of behavior
change to fuel growth for indulgence brands. We do this by [00:48:00] building
interconnected marketing ecosystems that place the human experience at the
center of brand building strategies across owned, earned, and paid media.

To learn how to leverage behavioral science in your marketing or advertising,
visit us@www.methodone.com.