353. Data-Driven Personalization for Online Retail - Jo Evershed and Gareth Harvey

Welcome to episode 353 of the Brainy Business Understanding the Psychology of why People Buy in today's episode, I'm excited to introduce you to Joe Evershed and Dr. Gareth Harvey.

Ready? Let's get started.

You are listening to the Brainy Business Podcast where we dig into the psychology of why people buy and help you incorporate behavioral economics into your business, making it more brain friendly. Now, here's your host, Melina Palmer.

Hello. Hello everyone.

My name is Melina Palmer and I want to welcome you to the Brainy Business Podcast. In today's conversation, I'm joined by Joe Evershed and Dr. Gareth Harvey. Joe is the founder and CEO of Gorilla Experiment Builder, a powerful, flexible and intuitive platform for running behavioral research online that allows researchers to go far beyond surveys. She also convenes and hosts the Behavioral Science Online Conference. An Innovate UK Women in Innovation Award winner, Jo is on a mission to provide behavioral scientists with tools to liberate their work from the lab and accelerate research initiatives that can be tested rigorously at scale. Joe leads a multidisciplinary team of software engineers and and psychologists focused on creating powerful and accessible experimental research infrastructure, and you'll hear a little bit about that through the test we worked on that we're talking about in today's episode. Gareth is the Director of Consumer Psychology For Decide, the UK's oldest independent creative agency. However, before joining Decide, Gareth was a professor of consumer psychology, having taught at universities in Geneva, Paris and North Wales. Despite having worked in universities for nearly 10 years, Gareth was a somewhat unconventional academic. Rather than focusing on publishing academic papers, which are typically only of interest to other academics, he focused on commercial research. As the Director of the Laboratory of Consumer Psychology, he ran projects for the likes of Unilever, Cadbury's and Aldi, helping them apply principles of psychology to improve their marketing. However, along the way he has also undertaken some more unusual projects, for example, working with counterterrorism units to help identify smuggling techniques in ports and airports, exploring how customers react to robots in supermarkets, and even investigating how visitors to art galleries react to interactive exhibitions. A decade on and he is now a Chartered Psychologist and and a member of the Global association of Applied Behavioral Scientists. As am I and many guests we've had on the show and I also interviewed the team at GAPS when that was launching. It'll be linked for you in the show notes if you want to learn more about it. In addition to his commercial work, Gareth has also worked as the course Director for the UK's first MSc in Consumer Psychology and has taught on a number of highly specialized master level modules including Consumer Psychology, Theater Theory, Nudges and Behavior Change, and Applied Consumer Psychology. Today's conversation is a fun one because we're officially kicking off the book launch for my third book, the Truth About Pricing, which comes out in just a couple of weeks, launching on January 9, 2024. In our conversation today, we're talking about an experiment we did together around the COVID for the book, which I can't wait to share with all of you. I know many of you participated in.

Some way or another in various versions.

Of testing, so thank you so much for that. And before we get into talking about that experiment, we have a really in depth and fascinating conversation about behavioral insights.

And research in the grocery store that I think you're going to like.

So we're going to jump right into that. Just really quickly. Before we get into the conversation, I want to be sure you know there are links in the show, notes for everything, including related past episodes, links to articles and books including the Truth About Pricing, so you can order yours and check out that final cover and so much more. It's all within the app you're listening to and at the brainy business.com353.

Now.

Let'S jump right in. Jo Gareth, welcome to the Brainy Business podcast.

Hi Melina, lovely to be here.

Hey, nice to see you.

Yes, I'm so excited. It's always fun when you do that introduction and there's more than one person because of the kind of awkward politeness of like who's do we talk at the same time? Who goes first? How do we make that happen?

It's always nice to have a little bit of confusion at the start of a call.

I love it. Yeah, we make it a little awkward before we jump in to have everybody kind of on edge. Well, with that in mind, I'm sure that there are people listening that know of you both and have heard your names, but potentially not everyone. So we can now kind of choose who wants to go first in introducing themselves and the work that they do.

I can go first. I'm Joe, Joe Evershed. I'm the founder of Gorilla Experiment Builder, which is a research platform for primarily experimental psychologists and behavioral scientists, both in academia and industry. And we're academics and an industry. They use it around the world to create, to design and host reliable online experiments so they can collect behavioral data. So whether you want to do surveys or reaction time tasks, or as we did for you, Melina, a shopping task, multiplayer tasks, then Gorilla is a Useful tool if that's your thing.

And I'm Gareth, I am the Director of Consumer Psychology for Decide. So if you're not familiar, Decide is the UK's oldest design consultancy and for over the last 70 years we've been working with some of the largest brands. So that's everyone from the likes of Fortnum and Mason, Procter and Gamble, Chicago Town, and we sort of develop product packaging, adverts and websites. So my background's a little bit weird. I did the traditional academic route, worked in universities, did a PhD in consumer psychology, ended up being a professor of consumer psychology teaching over in Geneva in Ashajay before leaving the academic world to go over to the industry side. So take what we know from academia and trying to apply it in the.

Commercial sector and it's probably worth knowing. Gareth and I met because I also host a conference called the Behavioural Science Online Conference, which is all about people doing amazing behavioral science online, collecting huge participant samples and getting really insightful data. And I had Gareth come and talk about various topics I think we've spoken about, including the psychology of pricing. I think at one point, yes.

So like I said, although I started off in academia rather than doing the traditional research, my research focused more on the industry side of it. So my PhD was looking at the designing of supermarket layouts. So that's everything from the music that's played, how wide the aisles are going to be, where the products are placed, the lighting, the color, so all of those sort of things. So although there's some lab based studies, most of my research takes place out in the wild where we have real people who do real things which often aren't quite what the textbook says or they're not in the nice controlled way that we can have in a university lab environment. So it's just trying to get that balance between what we know that works in a lab and taking it out into the real world.

Well, I know we have lots of people that listen to the show that work in those sorts of industries in the fast moving consumer goods space. We've had some guests. So someone from Mars and Kind was on and we were talking about a project that we did that had to do with the, like at the checkout, you know, chocolate at checkout in, in UK and some other areas. I've had someone talking about packaging and doing different testing based on color and contrast. We've talked about smell and like the sense of touch when it comes to products and I know that those are always really interesting. Do you have any thing that you can share of interesting work you've done.

Some of the interesting things that we like to play around with, it's the idea of just trying to explore and actually finding out some of the things where you need to get the intervention in the right place. So take the idea of using smell. If you're going to put just chocolate in the supermarket or the smell of chocolate, everyone always thinks that you want to put the smell in the chocolate aisle. And that's generally a complete waste of time because by the time you're down that aisle, people are already seeing the products. You've got a visual stimuli that is so much stronger than anything you can do. Whereas if you put the smell of, say, chocolate at the entrance to the store, we can typically get an uplift of sales of chocolate by about 12%. It's just going to be acting as a retrieval cue. So it's all about trying to think of what is the right product or right stimuli and just trying to make sure that we're putting it in the right place rather than just thinking about only that small, narrow space that you've got to work with.

Yeah. It's so important to take the step back. The thing that feels obvious like you're saying, like, oh, well, get them. They're in the aisle. We want them thinking about chocolate and not the, you know, sour candy or whatever it is. But if they never make it to the aisle, then whiskey.

Right.

You got nothing. Right. So. But being able to get more people there where they're more likely to then be driving the purchase, that they might not have done so otherwise, makes for a really good investment. And people that weren't going to get it doesn't really impact them either way. You know, if they're not enticed by that scent of chocolate, I don't know who those people are. I've heard they exist. But.

Gareth, was it you that told me that really interesting fact that because the rumor on the street is that they put the smell of bread through the store, isn't it, to entice you in? And then you said to me, it's.

Never bread because bread has such a low profit margin. So especially in the uk, there's always a competition to which supermarket sells the lowest bread. It's the sort of thing the newspapers always talk about. So the price of bread is really quite controlled because they're always competing. So if they put the smell of fresh bread in the store, it does prime you to buy more bread, but it also primes you to buy more baked goods. These are Kind of big items in your trolley. So like cakes, bread, muffins, whatever. So when you look down at your trolley, you've ended up buying quite a lot of big bulky items with a really low profit margin. So it makes absolutely no sense for us to actually pump the smell of fresh bread into the store. If they've got a bakery in store, like a lot of the big brands do, they'll pump the smell outside to make you hungry, because when you're hungry, you buy more. But once we get you inside that store, we really don't want you to be focusing on the smell of bread because it's going to be counterintuitive. You buy, you buy those big bulky items, you look at your trolley, think, oh dear, I bought a lot, and you cut your trip short. So it's something we always try and avoid.

No, but that's, you know, you think though, for a. Just a bakery, right? So if there was, you know, they want that sense of bread because it is so enticing. And you think about too, it's like, you know, going to. For an ice cream shop, right. The smell that we have. Even though, like, even if you're not getting the waffle cone, like you smell waffle cones and it makes you think of ice cream.

So good. The waffle cone, right.

And again, if you take somewhere like Subway, they only sell one product where actually you can use aromas. If we're talking about a supermarket or any big retailer, it's so much more complicated because every product had its own smell. And if you're putting one smell, which is product specific, we can drive sales of that, but it's going to have a negative impact in some other part of the store. So actually aromas in store, we use them, but you've got to be very, very careful. It's so much easier if you're a bakery or a store that only sells one type of product.

Right. Well, that is interesting though. Like, I think just for any of it, I'm always talking about that thoughtfulness aspect and getting into the mindset and knowing what it is that you're wanting to achieve. And so taking this, again, getting someone in the store in the same way we were talking about, like getting them to the chocolate aisle that they might not have gone by, that is likely to drive conversions. People are going to get more. They might have those impulse buys and then like, again, for the grocery store to get you in matters. And so that smell of the bread or whatever, you go, oh, you know, I do need to pick up a Couple things, and you may remember bread, but we've all been in the grocery store. Then there's so many things, and you get distracted by cookies or, like, you know, the Oreos or chips or the other fruit and vegetables or whatever it is that you're buying while you're there.

I don't think many people are distracted by vegetables, but the rest of the list I can go with.

I know. Oh, man, I'm so excited for. I can't believe asparagus is on sale. Which, you know, that's. Who knows?

But what you're saying about product location, it's so key because most people have the assumption that we walk around the store, we go up and down every aisle, or at least most of them. We spend a lot of time. We put little microchips in the shopping trolleys or in your baskets so we can see exactly which way you walk. It's a lot easier now because we now have digital CCTV so we can produce heat maps. And the way that most people shop is you walk around the perimeter. You walk around the perimeter and you dive down the aisle and you grab what you need and then you go back to the safety of your aisle. Sorry, the perimeter. So when it comes to putting products on the shelves, those key items we know most people are buying, we've tried to put them about a third of the way down the aisle. If we put them in the middle, you can't see them from the end. So people don't go down it. If we put them about 20%, 25% of the way down, you walk down, you grab it, and then you run back to where you were safe, about a third of the way down the aisle, that sort of sweet spot where it's far enough. Yeah, far enough that you're in. You're like, well, I might as well continue walking down the rest of the aisle. So, yeah, it's really key that you.

See the next aisle. You're far enough in that you see the next one and go, oh, well, I did need that.

Well, I guess that's the clever thing with the third, isn't it, that a third in, you can see another third in, which is the third from the other end, which is where the other good stuff is.

You feel like you've walked down it far enough, so you just continue going. So that's the key thing.

Oh, well, I get more exercise, though, so that's good.

I love it. Well, thank you for sharing. Like I said, I know that there are people listening from both sides. One in the Consumer side understanding the way that we shop and things like that we see, you know, we've heard the you know those staples often are at the back right Milk things that you know that you have to. And I've heard then it's someone like possibly debunked and said oh no, that's not why it's back there. But I think it's why it's generally.

It'S not, it's not that way. There is actually more practical things. There might be some other convenient things so which actually forces you. But within the UK stores that's kind of changing a lot of the time. It's a case of it's generally not a good idea to antagonize shoppers. And especially now we're getting almost two types of shoppers into supermarkets. We're getting the grab and go shoppers who it's quickly running in. So we almost have a store within a store. So where there's some quick essentials you can get and then if people are doing that big stock up store shop then they can go down having more one. So we're actually sort of thinking about designing a store slightly differently now.

Interesting. I have a follow up question that I think is going to bring in some of Joe's expertise as well in a question that I have gotten a ton you know since the beginning of the pandemic being questions of the online like having to shift the experience into online and how stores can be looking at how do you think about it in a world where you don't have end caps anymore and we don't like the ads but like there's the value of having things next to each other so we don't forget. But if I'm ordering to be picking up and like we lose the impulse buys but that's actually a big way that we buy products. You don't know what's on sale and you don't know products you haven't had. There's so many problems with it. Do you have any insights for people in that aspect of like using an app or online shopping for grocery.

Well I think with, with apps they still have the ability to serve you adverts midway through your shop and place them to go for those impulsive purchases. And you can also when you get there's usually offers along the lines of ah, last time you bought this thing you also bought this thing. Would you like to add that as well? Oh yeah, that would be helpful. Or oh we noticed that you haven't yet bought this, this and this. You often buy these too. So that's one way that, like true supermarket apps can replicate a similar experience, in fact, a more personalized experience, because it can be based on what you buy rather than what the average software buys on their way out. And I would far rather see personalized adverts than average adverts because I'm more likely to get the thing that I want within sort of our testing supermarket. So our research supermarket, the online one, we have the capacity to do things called swaps, which is quite good, fun. Epidemiologists like these a lot. So if somebody adds Frosties, let's say, to their supermarket trolley, we can make an offer of a product that's on the same shelf. So probably an acceptable category, but just slightly healthier. So Cocoa Pops, just slightly healthier than Frosties, because chocolate is a natural product. It's got some micronutrients in it, whereas Frosty's really just is pure sugar. Right? Or crunching up cornflakes would also be an alternative. That's just a little bit healthier because it's got nuts in it, which is, you know, if you can eat them, they're good. And so that's one of the ways we. One of the. One of the protocols that's been run by epidemiologists is can we get people to just make a. Literally a little nudge in the same category, a little bit healthier? Can we make acceptable swaps that people will go for? So that's a different way. It can be adapted.

The other thing I think is really interesting is when people design websites and they're writing their product descriptions, they spend a lot of time making sure they've got the right photos. They might have four or five different photos of the product. They'll have lots of information about the product, they'll write a long blurb, but they forget that, especially if it's something we buy routinely, like groceries. Most of the time, the shopper doesn't actually click on the product page. All that information that they've written gets completely ignored. So I did a project for a big alcohol brand, and we're looking at what they bought on an online supermarket, and I think 93% of the purchases were just made from the main page. They didn't actually click on the page to read the descriptions about the product because, well, they're probably familiar with it. So we've really got to think slightly carefully about actually how important the homepage is, what those first impressions are, because that's all that 93% of the shoppers are going to see. So actually, really, we need to think a little bit more carefully because actually that real estate is more important. The bit where we spend most of our time on. Actually only 7% of shoppers are looking at it, which is maybe we need to think about actually how we use our attention.

Now, I read this really interesting study that looked at changing those hero images. So rather than, you know, in a supermarket, normally you just see the picture of the product on a white background, but they don't put any hero text or benefits on top of it to sort of sell the product in a more advertising way. But this paper had changed all those images so that it did do that. There were like sensorial cues, there were hero images. But Gareth, didn't you tell me supermarkets don't do that for some reason?

Yeah, so there's that combination of things going on. We've got some best practice and standards that have been set by the industry and there's some of the things which are done to actually help sell. So in terms of what our best practice is, when we design product packaging for work in a supermarket, we've got lots of real estate. If you take, say, breakfast cereal, it's about a foot tall or 30cm tall, lots of information we can put on it. If we put exactly the same product online, you're looking at a space that's maybe an inch square, two and a half centimeters, and all that information gets lost. So there's new standard guidelines which are designed for product images for mobile optimization. And that actually means the photos you see on an online shop aren't actually what the product looks like. We've actually had to redesign it, so it will generally tell you what the brand name is, serving size, how many units you get in it, and that's more or less it. So we have to redesign the product so it works so you can see it on a mobile. Because when we design web shops, it's far easier to design the web shop or any website to work on a mobile and scale up, rather than trying to design a web shop that works on a desktop and scale down. So actually the images are very different. The other point, if you're designing another one and we're not talking about groceries, it's far more effective if we can put in cues like just to tell you more about the product. So, for example, rather than having a table that's floating in midair, just a white background, as photographers love to do, put it in a room, it makes it, the perceptual fluency, so much easier we can just visualize it, we can imagine, we can work out how it's going to work. And it's so much easier for you to be involved with a product. But best practice, maybe not best practice guidelines set by industry mainly sort of Amazon's photo guides, they encourage you not to do that. So it kind of, you've got this kind of contradictory, what we know sells better, but what the online platforms are accepting. So we've got a bit of a difference going on there.

Yeah, it is interesting. I'll link to this episode in the show notes. But when Christian Meshbia was on talking about his book look and talking about how we perceive information and we think that we look at a room or something and we take all the different parts and then we put it together and say what it is. But really we get the holistic sense of what we're looking at and then the parts only become parts because based on the experience of the entire thing in front of us. So you can see, okay, the context of this is dinner. And so this is a plate because people are going to be eating dinner and it's a family coming together. And here's the. Like the, the chair only matters in the context of the bigger instance of, of what dinner is. And even if there's a weird like blobby thing next to a table, I can intuit that that's probably a chair where I wouldn't necessarily think that when it's out by itself and doesn't have the greater like experience as part of it. And so in this way you think about the opportunity where you have something like on Instagram and you're scrolling through and you see the picture of a really delicious meal or you see someone in an outfit or something and it has all the little. Then like the dots that pop up. Like if you want to buy this, you can click here to get the information to buy the stuff. If you. Where I don't then have to go do the work to say, I wonder what kind of bag that is. Or like, how did they do this? So showcasing the really delicious looking meal and you could say, like, I want all those ingredients, put them in my cart.

Boop.

Like that functionality good. And a benefit that we don't necessarily have. And you could like scroll through and say, okay, well I already have cinnamon, I don't need more salt. Like you could remove some things that you already have and then just add everything else. And we'll give you the, we'll send you the recipe right away because you said you wanted it and so like, ta da. Right. Like that's a big win that you get online that you don't necessarily have in store.

Don't think we have that in the UK yet. But I'd love that if my online grocery store allowed me to save recipes, then I could just be like, I'm going to be making bolognese and the shepherd's pie and the this. Thank you very much.

Yeah, and just fill up my cart, give me all the things.

But the other thing I would really love is if the supermarkets gave me the option to like change some of the settings. It's like, like, if I could put in my motivation, it's like, I would like you to serve up products that you can charge me a higher margin, I don't mind, but that are healthier by a metric of lower salt or lower calorie content or lower calorie density. And it changed the salt order. I think, I think supermarkets who have become particularly online supermarkets, could be become a massive force for good in terms of nudging consumer behavior and being responsible citizens, making sure that we're filling up our trolleys with healthy, nutritious and delicious food. But that would take a leap.

It's a really interesting idea because since we've had loyalty cards, I think in the UK they were introduced in 92, they've got that sort of level of data in terms of knowing what we buy. And all this special offers that you get in the post or by email now, they're kind of tailored to you. So the example is they notice that if you give Heinz tomato ketchup, you get through it every six weeks, you might suddenly find five weeks after you last bought it, you get a voucher for 20 pence off because they know that you're going to be ready for buying it. So they've got that data. And if you could sort of personalize it to be able to say, like, oh, hey, I'm on a health kick. Because they already sort of do those sort of things because they know at the start of the year most of us are deluding ourselves with New Year's resolutions. So they kind of give us more healthy options in there. But if you could just say, right, I don't know, I'm going on holiday, I want to be doing healthier for the month before it. It'd be great for you to sign up, but as of yet, I'm not aware of anyone anywhere in the world that's doing a variation of that.

Well, that's also, if you think on the Fresh start effect piece. So it's of course not just the new year, but if we have a birthday coming up, you know, so which a lot of those loyalty cards have information knowing when my birthday is. And so if you assume that in the month to two months before my birthday I might be looking for healthier options, like that's something that you know, could be coming up or you know, before I know people have, it's like Mother's Day graduations. Like there are other points where people are looking at that and I think, you know, like you were saying, Joe, being able to put in a setting or some information that's really important to you that the stores don't necessarily ask. I've made this and I'm going to like say complaint in like quotes, right? But because I love Starbucks and everyone who listens to the show potentially knows that. And I've got my, I've got my, like my Starbucks. Like, you know, I'm in Seattle, love Starbucks.

And you have a favorite syrup.

Oh, so I do, I do.

Chai tea lattes.

That's my. Yes, I love, I know I love chai tea. And I know and I always then make the disclaimer of. I know that everyone who is outside of like the US and hears me say chai tea thinks I'm really dumb and I'm saying it wrong. But Starbucks calls it that. That is the name of the drink. Even though I know I'm saying tt, I know it, but it's, it's accurate in that way. So anyway, with this though. So I have celiac disease, I cannot eat things with gluten. Right. And I want nothing that's even in this like cross contaminated world. And I know they have a lot of gluten free stuff, but it's really hard to find options sometimes. So Starbucks will then do that. Well, if you buy your regular drink and add in a pastry, you'll get however many bonus stars. And there's no way for me to say like, hey, I'm never going to do that. Like you can promote it with a thousand bonus stars if you want. I'm not going to. I mean in that case maybe I would and I would give it away or something but like I'm not going to get it because I can't. It'll make me really sick. Right. And so same with the grocery store. Like I was looking, I was just in the store yesterday, I was going through store's app and it was combining like if you buy this particular brand of. It was Nilla wafers that if you buy them you get 50 cents off bananas. Which is funny because I don't know that anyone even still makes that cake with the like pudding and stuff anymore that they do around. But my grandpa used to make that. But like an association, I get it. Like never going to, never going to buy it. We don't buy gluten stuff for the house and people that are vegan or vegetarian or whatever. To know that the ads and the things that, that you would present are just, that's never going to be on my homepage is I think would be a really valuable thing if I can input. But there's nowhere for me to give that. But it would actually be really valuable both for the store to know and you know, for me as a consumer. So I think there's a lot of win wins there.

Imagine you logged into your online supermarket and you could go in and say exclusion free household. Just don't show me any of those products. Like save yourself the time. Save me the time. You get some really valuable data because now they can make sense of your data sets. It's like, oh, get it of course.

But especially if it's opted because with so many things you've got to be really careful about, shoppers just don't like the idea of having data on it. Now they may not realize or they may not think that the club card collects a huge amount of data but once they're aware of it and if you're telling people about opting in, they get, or they potentially can get very nervous. So again, as long as it's an opt in and it's a case of you can choose whether to go for that, you're okay. But again, especially with the heightened data regulations we've got coming in, there are a few more tricky logistical and legislative hurdles we need to see.

That's how I feel about my data is I want someone to use my data to help me turn into the best version of myself. That I think is completely wonderful. To use my data to turn me into the worst version of myself really isn't wonderful. But if I can go onto my supermarket and say please give me all the nudges that mean I buy less sugary stuff and, and more nutritious and delicious stuff is firmly it's something I would opt into most of the time I might turn it off. You know, in the run up to my birthday just because it's a treat day. Today I was thinking about a lovely case study by Dimitri Koutakidis over at the primary Care Health Sciences over at the University of Oxford. So he did a study where they changed the. When people added drinks to their supermarket trolley. And I think in this case it was. It was Coca Cola. They offered them alternative Coca Cola products. So there was the control condition where there was no information, no additional information, but then there was a little label just said, oh, this one has fewer calories, or this one has 208 fewer calories, or the worst one, this would take 46 minutes to walk off, or 46 minutes less to walk off. If you go for the lower one and you know, and he's got the data, he knows which one is more persuasive. It's the one that tells you that you're going to have to walk 46 minutes longer if you drink the full sugar bottle of Coca Cola. So, you know, if we can empower citizens to be able to nudge themselves to bring out their best selves, I think that's a really interesting new avenue for behavioural data to be used and shopping data to be reused.

It's really exciting. It's just, again, where do we put those things on the journey? Because like I say, I can't. I remember the paper, I can't remember where they put it because I think they're on the individual product pages. And it's a case of, if people aren't seeing those sort of things, how do we do it without making the web page too cluttered? And then we get into accessibility issues.

Yeah, well, so because this was done in our. In our shop builder tool, so it was done as part of a swap. And of course, no supermarkets offer swaps. Right? They. That's not functionality built into supermarkets because the brand owner of. I don't know, I'll go back to Kellogg's Frosties doesn't want to lose the sale to Kellogg's Cocoa Pops. Right? They. They're in their category, they want to make that sale. They don't want an alternate healthier alternative to be offered by the supermarket. But maybe there's a conversation that could be happened here because it is at least within the brand category. And then you could be a brand who supports the people who shop from your brand to be the best version of yourselves and that can become a brand value.

Yeah, I love it. I think lots of interesting options for people, you know, to be considering. Just looking at, again, where I would liken this to, you know, you think about something like, people hate standing in lines. A line is terrible. It will never be anything good. How do we make this A shorter, a little bit longer, worse or a little bit better, a little bit less bad. Right. Is sort of the approach people had to lines and many people still do. And then you look at like Disney turning the line into an experience and you know, universal, like, you know, people stand and go through the whole line of stuff at Harry Potter to be able to experience and don't even go on the ride. Like, the, the line is a big value add. And so I think, you know, instead, anytime we can look at a problem and instead of saying like, how do we make this less bad? We can say like, what are the unique opportunities we have here? And where we're talking about, like, yeah, I would probably nudge into the thing that's going to opt. I would opt into the thing that's going to nudge me. See, I said that all backwards. And then I like, what did I say? Okay, I would opt into the thing that will nudge me to align with what I've said I want to do. Right. And so soda being an example, like, I've cut out soda several times in like I'm, I'm currently now again on a time where I don't drink soda and I've gone years and then you accidentally have one and like fall off the wagon very quickly. I've always joked that I fall off the wagon, fill it with soda and then drink it. All right. Like that's, I go really hardcore when I start drinking soda again.

Wow, you love it.

I do. I know. I love Dr. Pepper, it's my favorite. I. But like I said, I haven't had soda for a long time now. But if I was going to put, if I did put it in the cart and it, and I was able to tell my supermarket, like, I'm cutting out soda, I don't want it. If you see that I add soda, like, yes, you should ask me the question of if I really want it. And maybe I'm buying it for the kids and like, yes, I want it. They're having a sleepover and I, I've made the decision, but at least it's asking. It's like, Molina, you told us to ask you if you really want soda.

Right.

Like that's a preference I could put in. Yeah.

But from the retailer's perspective, we could probably take it and make it slightly more subtle. So rather than putting in a cooling off mechanism effect, just think about the default options. So we know online the anchoring effect is far stronger than in a retail store. So whatever you first see sets your expectations. So we could just simply check, because they've already got the health data about all their products. It's already supplied to them. Just think about what's the default option. So it could be whether you wanted less sugar, lower calories. We can change the order. Certain things are presented. So you've got a balance in there. And again, giving the shopper the option of how do they want their default option? So somewhere halfway between the two.

Right. And it's not to say, like, do you not. Like, again, you know, the retailer is looking to make money. I get that. So then it's like, do you want. Do you actually want Perrier? Do you want the sparkling water? Is that what you meant when you pick this or whatever. Right.

When you click Dr. Pepper by accident.

Yeah. You must have meant Perrier. We put it in your cart.

Or was it champagne? We're not really sure.

Hey, you know I'm not against a white wine, right? That could have been put in there. Well, I have so enjoyed this conversation and I need to remember now that we're 30 minutes in, we're here to talk about something completely different, which is still a shopping experience, still really cool stuff. And I think a really great hard transition into what we're talking about, but I'm okay with it, is that we're talking about how we had done some work together for my new book, the Truth About Pricing, which in, you know, spoiler alert, the truth about Pricing is that often price and whether people buy or not is not about the price. There's all this other stuff that happens that influences whether someone's going to buy or not, which is a lot of what we've already been talking about here in this conversation, where, like, Joe, you were saying, I, I'm okay paying more if it's going to do this thing. And I'm. I'm okay with this. And people are more likely to buy because of the scent or, like, what they see than something else. Right. So the book has many of these things. And knowing that while we're not supposed to judge a book by its cover, we definitely do. And so I had done just, like, taking this, like, step way back. So this was in the summer. And I have been working on the Truth about Pricing for a long time. And my. I. The one thing I had said when I've done my other books is, you know, like, we talk about color of them. And I've said, as long as it's not green, because the pricing book is going to be green, is what I've said from day One that I knew that this was what this was going to be. And then my publisher sent me some mock ups of potential covers for the book. And they had a red one, a yellow one, and then two green options. And I saw the red one and went, huh, like, this is actually very interesting because it drew my eye differently than the other covers did. And I thought it just was very appealing in a way of making you stop and pay attention. And then I start kind of spinning on all the ways that red could be an interesting cover, cover, color option. So I asked my publisher to do a green mock up just so I could see it in what I've always said I wanted. And then I had them side by side. I couldn't decide. So I put it out on LinkedIn and Twitter and. Or X these days. But then it was Twitter and you know, all across my socials and ask people, what do you think? You know, which one would you pick? And we'll get into some of the data I got from that. But then Joe, we had recently had a conversation about something else in advance of that. So then you'd said, hey, we could maybe run a test, like, we could see, put it in it, we could build a shop test and like, let's see what actually happens. And so we got to put that all together to test red versus green. And some of the feedback I had gotten and I will just share so in the results across social media. And like, I know there's a reason my first two books have the title being you know what your customer wants or employees need and can't tell you.

Right.

So knowing even when I ask people what they want, it doesn't mean they're going to buy it when, you know, push comes to shove when the moment is there. And even if they say they don't like red, it doesn't mean that they really don't or that it didn't draw their eye or whatnot. But across social, it was like between 70 and 80% in favor of the green cover. And there were a couple factors that we learned in this. Partially the red was. There were some concerns about the level of contrast between the colors, that it might be an issue for accessibility. So but that's like something that could be fixed even if the overarching red is an okay concept. And I had someone then submit a different. I had so many people messaging me with like, oh, I came up with, I made it. I mocked up another cover for you. I think you might want to do this color. I did this and like people Just sending me stuff, which thank you to everyone. That was a lot of fun. So we did test one of the contrasted ones that was then like a dark purple with the green to be able to check that out and see. See how it would compete when we put it into the shop. So do you want to talk a little bit about the experiment?

Yeah. Well, it was such an exciting opportunity there. There I was seeing all of this, like, focus group ish data come in going, I like green, I like green, I like green. But this occasional detenting voice going, no, red, red, red, team red over here. You know, my brand is a red. Working with red is extremely hard. But, you know, I do like. But so I could. I could see the. The difficulties that people were facing. It's like, oh, green. Yeah, money, definitely. But, oh, it could disappear. Contrast all the rest of it. And so I thought it was this exciting opportunity to chat to Molina, who obviously amazing behavioral scientist and understands why what people say and what people do is not necessarily the same thing. It's like, we should put this to the test. We should put it to the test. We should get some real data and then we could talk about it because that would be really good fun. So that's what we did. And I sort of. I asked Molina to send me through a list of 24 books and their book jackets, I think. And then someone in my team, Megan, put together the shop in like an hour or so, uploaded the books, their descriptions, the prices, laid them out in a store, made it look nice, set up the experiment, you know, asking some demographic information, some consent. And then there was a bit at the end, which we'll come back to afterwards. And we had our A4 way between subject experiments. So we had a control condition. I just wanted to get a few participants. We turned this off part way through the experiment, but we wanted a control condition to get a baseline of. With none of your books in it.

I think the control had my other two books, but it didn't have the pricing book.

But that's right. Yeah, yeah.

Didn't have the new book, but it did have my first two books.

Yeah. And that was partly because we knew we were going to try and get participants from your audience and your followers and people who follow your podcast. And for all of you who took part, thank you so much. That was a lot of fun seeing that data come through. And then we tested the three book jacket colors. So those were the three other conditions. They had the green book, the red book, and the purple, the purple book. And we put it to the test. And should I say what happened?

Yeah, yeah.

It was quite funny.

Oh my gosh. I know. I got the first email and like I said, I was so. I was already feeling like I'm not really sure. But then all the feedback came in from social that, like I said, was like 70, 80 green. And because of the timing on these two and I know I told you, it's like I have to tell the publisher to like move forward with something because they're, they're having to go. So we're gonna say go with green and we'll see if the test supports what everyone said online. But like, this is what's already out. This is, you know, what we had. So we'll say it's move forward as if it will continue to be green. And then, and then.

Well, wait, before we get to the. And then just to drag it out a little bit longer, Gareth, because this is where you joined the conversation, wasn't it? Because I tagged you in the LinkedIn thing. We said, why would the behavioral data be different from self report data? Like, why do people decide differently when something is on a shelf versus when they just think about it in isolation? And Gareth, you said so.

Typically in a supermarket, the way you make a decision and it's the same online is you're going with what you quickly see. Visual saliency is probably one of the best. Well, it is the best predictor of what someone's going to buy. So that could be whichever product is the biggest, whichever one stands out the most. That's what grabs our attention. As soon as you ask someone like which one of these two products you like, you stop and think and you're going to be considering it. And it's. That's just not what happens. Like when you buy something online, we can look at the data, we can see that people are spending maybe two to three seconds and they're having to review maybe 10, 12 products in that time. It's what grabs their attention. So you're making two fundamentally different decisions here. So it's all about visual saliency first and then what associations comes next. The associations doesn't matter if you don't notice it.

Yeah.

So this is what we got into. This is one why we wanted to run this test. And then we were bad scientists here, weren't we, Melina? We just have to fess up. We were bad scientists, yes, but we had like 50 participants worth of data. And so we downloaded it and we looked at it and it was like the red book.

Like way ridiculously in favor of red. And you go, wait, what? Like.

And they were like, no, no, no. There might have been an early rush towards red and it's just moving gradually. And that's what was happening, is that each time more participants came in, it was. Because you can imagine in an election when somebody passed the very first ballot, right, at some point, it's 100% for candidate A and 0 for candidate B. So when you've only got a small amount of data, you get these weird swings. So it was all for the red book. And then as the data came in, the gap narrowed and narrowed and narrowed and we got a chunk of data from your audience. But partway through, we felt we didn't have enough power. And this is one of the things that's really obviously important with online experiments, is you need to be able to get your participants so you can source participants who will just take part for fun, as we did, but you don't get the volume in the time you want. Like, it's a labour of love to get people to take part for fun and to find a group of people who are behavioural scientists who might be interested in a book about pricing. It's pretty niche. So what. What we also did is we got in touch with our friends over at Prolific, which is a platform for recruiting participants to take part in science studies online, and they offered to join in our little research project. And they. They helped us collect some data from some small business owners because they have a category of small business owners and small business owners are always going to be likely to be interested in sort of business E books, psychology books, pricing ebooks, that. That's in their niche. So then we did manage to get data from another 300 participants overnight, which was great. And once we had all that data in the green book, one.

Hooray.

Which worked out another 300, it will change again. But I think it's always nice when the focus group data and the behavioral data align, because when they're completely different, then you've got to. Then you need to go and find out why. Right, right.

And it's interesting to see with this too, was having this group from Prolific, of which. Thank you so much, to Prolific, we appreciate you that it's not just like people that know Molina, going and looking at books from Molina, because that's also. If you know that I sent it and you see, and you know, I'm working on a book about pricing and you happen to see it and you're in the red condition, like, you know it's like it could be just like my loyalists that ended up in that one and there were some other people that were in the other condition. So it can get a little bit muddled in that way when it was a little bit more transparent of what we were doing and what we were looking at kind of and why people didn't necessarily know it was about the COVID But also in this case I was very thoughtful and I think we all were in knowing based on a conversation I had about which color cover do you like turned into a study about book covers. Oh, ah, I wonder what we're looking at. Right. It was pretty obvious that, you know, we, it wasn't the, the purest of samples. So adding in that data from Prolific really I think helped to balance that, that out to see how the book performs against other books. When someone doesn't have any background necessarily of me or the work and they're not, you know, ready for that to help show that it wasn't about the exact, that color of the COVID wasn't necessarily the, the most important factor.

The other thing I think was really cool was the fact that you were naughty and peeked at the data. And it just shows you the importance of power analysis and funnel analysis. You often get headlines in the UK where the newspaper say, oh, let's say Birmingham has such a cancer rate and then a small city is either much higher or much lower. And all it is is a regression to the mean. It's a simple case of if you've got a large sample, you tend to get the mean. A small sample, you're going to get something way up or way down, 50 people. And especially when you've got a lot of variation, you're going to get messy data. Getting those extra people from Prolific allowed you to get close to that mean. And it's kind of really cool just to suddenly see these sort of differences.

Really cool. But Melina, tell me, how did it make you feel having this behavioral data? Like, what was that like for you?

It's funny, I love doing these sorts of tests. So I am. And it's been real. I am so fortunate to have met so many amazing people through the work that I do that often, you know. So with my first book, Emotions, we did a study on the website that we were doing for selling the book on my website of we did different versions to be testing, you know, that we were able to share and talk about. And then here being able to do like looking at the covers of the books, you know, being able to try and test some of the work that we do to show, like, you know, practicing what you preach to see what you can find out is really important. And so I, I enjoy it for that aspect and that I also dislike the uncertainty that you feel where it's like, in this case, it's like I'm already asking all of the questions and I can logic any answer, right? And I know this. So trying to just let it, let it guide. And I was ready. You know, it's like, okay, I guess if we have to go red, we're gonna go red. Like, I, you know, I was willing to go with whatever the data said. So that is one thing have to be. Be ready for. And ultimately the decision was up to my publisher as well. If they said, we don't care, we want to go with green, like, say, okay, right, Like, I'm kind of stuck in that particular case. But, you know, it did kind of line out that green did win, and it was. It won by a margin that if we had already started down a red path, I think we would have been okay. But what was really interesting in the data that I think was super important, important for me is knowing that while the red or, sorry, the green book of my books got higher sales than the other color covers, it didn't do the best in books about pricing for businesses, which is what category it's in. And it's also very nice that I have very generous people that I've made good relationships with over the years that will potentially follow up with me and let me know, like, I want to tell you about my thought process when I did the survey for. For this or whatever. So I did have a very kind student of mine that reached out and said, hey, I did the. I did the thing and I didn't buy your book, and I want to tell you why I didn't pick your book. Which is, again, we know their logicing why they did the thing. They thought about a lot more than they might have when they're doing the test or whatever. But, you know, in that case, they said it had to do with the description. And I know when we set up the shop, I had told you, Joe, that I didn't, I didn't. I didn't like the description. I didn't think it was right. And we just were moving forward with it being about the COVID and seeing that in this case, the COVID matters. And that description is going to be very important in a way that potentially, like, we were talking about the shops and that, you know, that 93% of people in the supermarket, don't get past that first page. I think with when you're looking for a book to see if it's going to be helping your business and you're going to invest in your shopping for a specific book, I think that's a little bit different. You know, when you're determining what you want, it's not, not everyone, of course, but making sure that that description is really spot on and is going to catch their attention is really critical to make sure that the book then stands out and it reinforces like, yes, I want this when they get to that page. So that's some insight that really came from this about how important that page is going to be.

One really cool thing though is the dependent variable is although we often look at which books they bought, you can actually start dig down into the data a lot more. So if you're talking about the description, we can see did they click on it, how long did they spend reading it? So if we know that someone clicked on the page and within 2, 3 seconds they clicked off, it was like, oh, clearly we did something wrong in that first sentence in that product description. If they suddenly realized they read all of it and they still didn't buy it, okay, then maybe there's just something in the whole thing that's wrong. So when we run online shops, it's not just what they bought. We know exactly what they clicked, what order they clicked on, we know how long they spent reading it. We've got so much data, it's just a case of working out, okay, what are we going to do with this? We could even start looking, seeing like regional differences. Like in the us, green is the color of money. Uk, that means nothing to us in that link. So we could start to dig down into those differences. So we have so much data, it's just working out how much you're going to tease it and just making sure we're not digging for something. That's if we look enough, we'll find something. Because of statistics.

Sitting with data and waiting for it to start talking to you is always one of those really exciting bits in the project. We actually did pull out two interesting different ways of looking at your data, Melina. So one, your the green condition one in terms of absolute sales of your book in that condition compared to the other two. But also the green condition, your book ranked higher in the list of books that were bought overall, so it ends up being in the top three. And I'm always reminded that, you know, and particularly with books where there are A lot of books within a category, being in the top three of a category, that's fantastic. You don't need to be top one because when people are reading around a subject they're going to buy multiple books. And that was the bit of feedback I got from one of my followers. She emailed me very kindly afterwards. I love taking part in your experiment, but I couldn't check out because I had six books in my basket. And after that we've just got to go over to Amazon and buy them. But yeah, so I think that's the other thing too is that we had these two, we've got a lot of other data which we haven't dug into, but these two, these two different angles, the absolute sales, but also the ranking in the sales.

Oh yeah. Well, I'm excited to learn more. We haven't had a chance to fully debrief on all those other things which I'm very excited to learn more about and maybe we'll do a follow up episode or something. But I think the important thing for everyone is to know that like yes, we behavioral science people do like to test things. And there's, we're not just telling you you should test, we like to run tests as well. And you know, it's important to be open to what could happen. Like I said I was ready to go yellow or contrast or whatever if that came up higher. And so being able to understand that you want to like let the results show what they're going to show and, and learn from it in a way that's really interesting and then being able to find what those insights can be to, to shape the experience that we're going to have. The great thing about, you know, what we did learn here being that the Amazon description and the title and the keywords and stuff that's in there is going to be really, really important one. My wonderful publisher Mango really specializes in that and so I know they're constantly testing anyway and so they will help to find the best version of this. But that is something that you can change very easily without any sort of issue. Whereas the COVID is pretty fixed. So it's good to have opportunity to keep testing on that and to keep learning. And I thank you to everyone who participated. I know there were just hundreds and hundreds across the questions I asked on social media and people that actually did the survey. So thank you to everyone who participated. It's been a lot of fun and hopefully it's good to hear all about it. So as we wrap up our conversation today, knowing that There will be have to be many more to come. For everyone that is excited to learn more about both of you and to connect and to hear what you're doing and maybe look at doing projects of their own, you know, what is their next best path to do. So for each of you and whomever wants to go first, you know, gets to.

Well, you can find me on LinkedIn. I'm Jo Evershed. I also have a freebie for anybody who wants it today. So I wrote a paper earlier. This is, you know, us researchers, we like to write papers. I wrote a paper earlier this year about what data from 1 million participants tells us about online behavioral studies. So because, because our, with our platform, we collect a lot of data across the world. We have metadata on participants, how long study are, are what they're doing, what the attrition rates are. So I have a paper that you can access, well, a sort of a more fun short PDF about that. So you can get that at gorilla sc1 million hyphen participants, if that interests you. But if you just want to follow along, see what we're doing, then, yeah, follow me on LinkedIn or follow me on Twitter or X or Blue Sky. I'm usually some combination of Evershed Joe or Joe Evershed. And I think I'm the only one in the world. So, you know, fairly straightforward to find.

And if you want to find me, I'm also on a range of social media platforms. Twitter, X, LinkedIn, BlueSky. I'm Gareth J. Harvey and if you want to find out more about the work I do, you can go to www.consumersite.co.uk and there is a newsletter you can sign up to. And the next edition is all about running online experiments and how to optimize web shops.

Oh, wow.

Yeah. So into this one.

Perfect. Well, I love it. So we will go ahead and link, of course, to all of that in the show notes to make it easy for everyone. And thank you again, both of you so much for working with me on this. I appreciate it so much. It's been a lot of fun and also then for coming on the, on the show to talk about it.

Well, thank you so much for having us. It's been such a delight working with you and so much fun setting up this experiment and helping you find that certainty that the green cover was the right way to go. And it's just been really delightful and a lot of fun.

Yeah, it's always good fun looking at data and actually seeing what people do because we always think we know what they're going to do, but it's nice actually being shown that we're sometimes right.

Yeah, for sure. Awesome. Well, thanks again.

Thank you again to Joe Evershed and Dr. Gareth Harvey for joining me on the show today. What got your brain buzzing in today's conversation? For me, this is an extra special.

Episode because as you know, it's really.

Kicking off the launch for my new book, the Truth About Pricing. I'm going to talk about it more in the coming weeks episodes leading up to the release, but this one has been in the works for a long, long time. Episode 5 of the podcast was also called the Truth About Pricing. I'm going to be refreshing that one soon and I have been talking, teaching and consulting on pricing strategy for years now. The first book was actually almost the Truth About Pricing, but I am glad.

It worked out this way. I think it is the way it.

Was meant to be to have what.

Your customer wants and can't tell you, what your employees need and can't tell you. And then now the Truth About Pricing. You're going to hear all about how.

Those books kind of like come together and how I created the book in an upcoming episode.

Just know that if you've ever struggled.

With prices or had questions about what.

The best price is for your business.

Or how to be more confident when selling, or had interest in the psychology behind prices, this book is for you. I wrote it with small and mid sized business owners in mind and it's.

The closest I could get to a.

Step by step guide to finally master pricing and move on to confident selling in your business. As I said, the book is already on pre order. You can go check it out wherever you buy books and of course on the Brainy Business website atthebrainybusiness.com Please share it around if there's anything you would like to do to help support the book launch.

Whether it's a podcast I should be.

A guest on, whether it's yours, or just something that should be on my.

Radar or a place to write a.

Post or to have me come speak or people that might be interested in the book. Anything. Please email me melina the brainy business.com and of course tag me in any posts on social media. You'll find me as the Brainy Biz pretty much everywhere and as Melina Palmer on LinkedIn. As we close out the show, don't forget about those show notes which include links to all those related past episodes and books including the Truth About Pricing as well as ways to connect with Joe and Gareth and learn how you could do your own tests, work or experiments with them, as well as a link to the Prolific website in case you need participants in those experiments. Thank you again, Prolific. You'll also find a full timestamp summary of the episode and more. It's all waiting for you in the app you're listening to and atthebrainybusiness.com 34 and thank you again to Jo Evershed and Dr. Gareth Harvey for joining me on the show today. It was a delight to chat with and learn from you. Join me Tuesday for another Brainy episode of the Brainy Business Podcast. It's going to be a lot of fun, you won't want to miss it. Until then, thanks again for listening and learning with me and remember to be thoughtful.

Thank you for listening to the Brainy Business Podcast. Melina offers virtual strategy sessions, workshops and other services to help businesses be more brain friendly. For more free resources visit thebrainybusiness.com.