Does male insecurity increase Ferrari sales? (And 6 more psych-myths)

0:00Today on Nudge, I'm interviewing someone who has spent her career studying people. So you recruited 1 ,200 participants and put them through the simulation. She shares examples of the weird and counterintuitive methods organisations have used to persuade. Islington Council needed residents to water new street trees. She knows an awful lot about human psychology. When they debrief the participants afterwards, half of the participants never consciously noticed the advert. But it changed what they bought anyway. Today on Nudge, she shares some advice that I think everyone could test. The winning group, their ask included just one additional sentence, which is most people donate around£70.

0:44And she even has an answer to a myth that has plagued car salesmen for decades. And what they found was that men who were told they were below average found the sports car significantly. Significantly. Well, to hear the answer of that, you'll have to keep listening to today's episode of Nudge.

1:09Most businesses have an overwhelming amount of data. It's typically scattered across the business in different tools with different teams. But because it's scattered, no one can use it. That is where HubSpot come in. HubSpot connects it all. every interaction, every support ticket, every conversation, and they connect it into one platform that every team can work from. Because when you know more, you grow more. Check out HubSpot.com, the agentic customer platform for growing businesses. Hello, everyone. You are listening to Nudge with me, Phil Agnew, and I'm delighted to welcome Jo Evershed onto the show.

1:51Hi, I'm Jo Evershed. I'm the CEO of Guerrilla Experiment Builder. Gorilla is an experimental research platform. So in a nutshell, participants go in and gorgeous cognitive and behavioural data comes out. Originally, Gorilla was built for psychology researchers so that they could run properly rigorous behavioural experiments online. Not just surveys, actually controlled experiments with near millisecond level timing and at a speed and sample size that no physical lab could match. Gorilla is used at half the UK universities to teach research methods and it is used by many businesses to conduct experiments that reveal a participant's behaviour, not just their answers to a survey.

2:35Today on Nudge we'll be covering some of those experiments starting with an optician's planned rebrand. Vision Express, which many Brits listening will know of. They are one of the UK's largest opticians and they wanted to rebrand. They wanted a new colour for their brand but didn't really know how to decide what colour to pick. Obviously you could ask people what colour they preferred but Vision Express decided to run a bit of a different study. This is one of those studies that for me was really thrilling because I've actually seen a real world change on the back of research done on my platform.

3:13Vision Express needed to pick a new brand color before a national rollout. And a folks group could only tell you what people say they like, not what actually gets through the door. So this was Tim Routledge over at CX Labs. He was working with Vision Express. And what he recreated was essentially like a high street experience or like a shopping mall experience inside Gorilla. So it felt a bit like walking through a Google Street View. And into this Street View, he could place Vision Express, boots which is one of the other high street opticians and spec savers the other high street opticians now boots has the blue color spec savers has the green color so vision express were trialing purple and orange or a yellow now color theory would tell you one thing focus groups tell you another and he was like well let's just put it to the behavioral test so he recruited 1200 participants all with corrected to normal vision so people who go shopping for glasses and put them through the simulation.

4:09There was actually a really clever mechanism he put in there to get twice as much data, which is that the first shop you clicked on, which is probably your brand loyalty preference, was always shut. So then you had to choose a second one instead. And to their surprise, it was the purple colour that won out. Vision Express tested three colours, orange, purple and yellow, and they were surprised when purple won. In focus groups, customers told them that purple made the brand seem luxury and expensive, But these behavioural experiments revealed something different. And then when you went into the shop, you were asked another question about value for money and other things.

4:48So purple is often associated with luxury brands. But actually, the participants, they chose it more. But they also associated it with value for money. Surprisingly, when the store was coloured purple, consumers predicted the glasses would cost less. Now, you shouldn't take this advice to mean that you should colour all of your stores purple. I'm not sure these findings spill over into other domains at all, but it shows that surveys and focus groups have their limitations. When asked, no customer said that purple would make them more likely to buy, but that's exactly what these behavioural tests revealed.

5:23All right, let's move on to the next study. Islington Council needed residents to water new street trees. So new trees that you've planted, they need about 80 litres of water a week, right? They weren't going to get round and water them all. They needed to find a way to get local residents to water them. If London isn't anything like Cambridge, then everyone's coming home with a bottle of water at the end of the day. And if you could just get them to empty it onto a tree, they were going to win. Very, very straightforward. Every marketer knows the first challenge is you've got to get somebody's attention.

5:52What the students did is they tested three messages on Guerrilla. So there's a social media simulation in Guerrilla. So they could create a randomized control trial with these three different types of messages. They could pour participants into it. Participants would see a simulated social media feed of 16 posts and one of these posts was the water the trees message. The team tested three variations of this message. They ran three messages on Gorilla, plain instructions, please water the tree, a photo, like engaging, creating that visual appeal or this meme. And then they measured after exposure to this advert, the intention to water.

6:32So participants simply scrolled as they would on any afternoon when they're scrolling through social media. Some saw a factual information message about watering trees. Others saw the same factual information message, but this time with a photo. But the final group saw a funny meme. And that meme encouraged people to water plants in a sort of jokey, banterous way. Which do you reckon worked best? What they found was there was much greater intention to water in the meme situation. And in fact, there was no difference between the plain instructions and the photo. If you haven't caught somebody's attention, you're just not going to land your message at all.

7:13This doesn't mean that you should turn all of your ad messages into memes. But it does show that occasionally humour works better than information. It's common for councils to only use bland, factual statements. but the study shows that that's not always the best approach. I'll hand back to Jo for the next study. This time, she helped a pharmaceutical company create an advert that actually encouraged people to buy. So there's a global healthcare company. They were launching a new cold and flu product and they wanted to know what packaging and messaging was actually going to drive purchases and in particular whether the same message would work in every single market.

7:52Just close your eyes a minute. Imagine a pharmacy. right they are bonkers shopping environments they're totally mad there are thousands of products thousands and like they're all in little boxes this big 90 of them white with tiny writing right so if you had a focus group here and you go do you like this packaging they go oh yes what do you think of this font oh yeah that's a really interesting font it's like okay would you pick it off a shelf? Oh, I suppose so. That's not really going to tell you whether somebody is going to pick it off a shelf. So instead, what we created in Guerrilla was a bit like old school.

8:32We had these still photos of pharmacies. And then there were buttons that they could click around to move it, move around the pharmacy. So again, a bit like a Google Street View, but inside a shop so that you could click and see that you could walk down an aisle, click onto an aisle, walk up to the counter, talk to the pharmacist. You could take each of these interactions and you could have a little chat as well. And then around the store, there was an advert placed, a large advert placed for the product that they were selling with the large packaging and the colors and the messaging. And then when you went to the relevant shelf, you were told you were buying a cold and flu medicine.

9:07So when you went to the cold and flu medicine shelf and you clicked on it, you were given a close-up of the shelf. I think there were three products there, the target product that they were selling and then two close competitors. And they were trying to see that was the forced choice. do you go for our one or one of the other ones? Participants in the study moved through a virtual environment, browsing shelves, interacting with products and even encountering a digital pharmacist. Nobody was told that they were evaluating marketing messaging. Instead, they were asked to simply buy a cold and flu product as they moved through the store.

9:43The question is, did any of the massive in-store adverts for the cold and flu product actually work. The best messaging and packaging outsold the worst messaging and packaging of the range that they were considering by around 10%. For a fast-moving consumer goods company, a 10 % swing between the worst messaging and the best messaging is a lot of money. The right ad made participants 10 % more likely to buy. But that's not even the most interesting finding from this study. When they debriefed the participants afterwards, half of the participants never consciously noticed the advert but it changed what they bought anyway and so we take that as actually evidence that the simulation is working really well because quite a lot of our decisions are made at that subconscious level there's plenty of things happening and happening to us in our environment the whole time which we know change our behavior but they're happening below the threshold of conscious awareness Half the participants did not notice the ad, yet it still shifted purchase decisions by 10%.

10:46It sounds counterintuitive. Surely for an ad to work, everyone needs to notice it. But it's clearly not the case. Even if we're not consciously aware, an ad can alter what we buy. Now before the break, I want to cover one more counterintuitive finding. This is an exercise that, this is Malstuft at UCL, runs with his undergraduate students. I think these are first year undergraduate students. They've been studying psychology for less than a year. And he's been running it for several years. I'm going to say around six years, right? And the challenge is, as a group, they pick a charity that they want to raise money for.

11:25And then Miles sets up the control condition, which is just a vanilla ask for money. And then every group goes away and tries to leverage all of the social psychology they've been learning across that year to try and nudge behavior and beat each other out. to raise more money for this charity. So there are two measures. One, there's a measure in the simulation, which is how much do you intend to give to charity? And then at the end of the experiment, every participant is landed on a JustGiving page and then actually asked for a real ask to actually give money. I love this. What a brilliant competition.

11:54Students use all the nudges and behavioural science tactics they've learned at uni to persuade participants to donate as much money as possible for charity. If you were a participant in the study, you'd be first told that you won£100 in a lottery, and then you were shown one of the student's charity appeals. Let's hear which appeal worked best. The winning group, their ask included just one additional sentence, which is most people donate around£70. So the way they did this is they had a version of the advert with a slider in the middle, and when you landed on the page, the slider was set on the£70 saying most people donate around£70, Whereas in the other group, they showed the same slider moving, so it was just moving along, but never actually landing and settling on£70 and without that message.

12:47The participants who saw the anchor gave, well, intended to give, about£47 on average. And those who didn't see that anchoring effect, that social proof effect, they gave on average£27, nearly£20 less from the same pot, for the same cause, for the same appeal. The only difference between the winning group and the control group was a number and a slider that stopped moving. The single line, most people donate around£70, shifted generosity by almost 70%. It's a reminder of how effective anchoring and social proof really is. Where I think that's useful for, if you've got any charities listening to this, where I think that's really useful for them is that adding an anchor to your messaging costs nothing.

13:34It's a cost-free change that dramatically shifts the amount people donate. OK, so far we've shown how changing the colour of an optician's made people more likely to buy, how memes made residents water local plants, that ads which are missed by 50 % of buyers still have an effect, and how anchoring almost doubled donations. After the break, we'll cover if AI could replace police investigators and if insecurity about a man's sexual reproductive organs could make that man more likely to buy a sports car. All of that, coming up. This episode is sponsored by Riverside, which is what I use to record this podcast on.

14:17And this episode, of course, is not just an episode. By the time you hear it, I will make it into a newsletter. I often turn my episodes into blog posts, into LinkedIn posts, and this is something Riverside can really help with because it can translate the episodes I create into these other types of media and this is the bit that marketers should steal. I do one recording and get so much content off the back of it. Go to riverside.com, use code philagnew and get a month of the Riverside Grow plan on us. The Ross Simmons Show is brought to you by the HubSpot Podcast Network, the audio destination for business professionals.

14:55I've been following Ross for a while. His podcast is great and an episode I'd highly recommend is called Mastering Time, Prioritisation Over Productivity. He gives some really well thought out tips on how to prioritise your time, including something I care a lot about, which is how a small percentage of your efforts can drive the majority of your results. So go and listen to The Ross Simmons Show wherever you get your podcasts. Hello, welcome back. You're listening to Nudge with me, Phil Agnew, and the fantastic Joe Evershed. Now, there's no doubt that AI is threatening to take over a large number of jobs, from customer service reps to taxi drivers.

15:34But could you imagine an AI police officer? Could AI be better than a police officer at interviewing a witness? That's what students at UCL set out to test. I'm quite keen on behavioral researchers getting involved with testing AI. AI is something that's new to the market. It's coming very quickly. Lots of people are using it. We don't know how it's impacting people, their attention, their memory, their creativity. So one of the things I'm really keen on is that researchers start to use it and start to see how it impacts people so that we can use AI as a tool for good. So this particular study was, and I'm sure your listeners know about the eyewitness testimony effect and how you can reshape somebody's recollection by misasking questions or by asking biased questions.

16:27So, despite training police to be impartial, they are humans, right, with emotions and feelings. And so, the premise here is can an AI that has no emotions and feelings and therefore be more reliably unbiased, could it help with gathering eyewitness testimony? Would this be useful? So this is how the study unfolded. As a participant, you'd land in the study. You'd end up watching a video of a crime. And then after that, you're interviewed by, and so this is a two by two design where you've got a human or an AI, both of which have been trained to ask impartial questions. And then the other part of the design was that sometimes the human or AI would first try and build a relationship with the interviewee.

17:14So creating this sense of rapport. So to recap, 105 participants are brought into the lab. They watch a video of a bike theft. In the video, there are two thieves working in tandem, one stealing the bike while the other kept watch. Then they are interviewed by either an AI chatbot or a trained police officer. Surely the police officer is better than the AI. Now, the results are surprising. There was no difference in recall accuracy depending on the human or AI interviewer. Actually, that didn't have an effect. However, rapport building was material to eyewitness confidence. So not accuracy, but their confidence in the answer that they gave.

17:57And I should add, the human police officers were better at building that rapport. So the AI chatbot is as good at getting accurate statements, but not as good at building confidence. Now, time for the final and arguably most ridiculous study of this episode. If a man feels insecure about his manhood, or more specifically the size of his manhood, if he feels that way, would he actually be more likely to buy a sports car? Let's find out. So this is like super boffin at UCL, Daniel Richardson. He's absolutely extraordinary. And it all started with this book about the super wealthy. Chuck Thompson was interviewing people at Ferrari and Lamborghini dealerships, trying to understand what motivates somebody who already has four Lamborghinis to buy a fifth.

18:49And for most of us, that's just like, I just don't understand. Why do you need a fifth one? And every single dealership he went into, the one thing, you know, their answers changed. But the one thing that stayed constant was a joke about male anatomy, a joke we've all heard. And, you know, you know, it happens once, it happens twice. You're like, OK, that's just frivolous. But once you've heard it again and again, you're like, this is a thing. Like, is this a real thing? Like, could we actually get an answer to this question? So he got in touch with various scientists to go, like, do you think there's a way of finding out?

19:22To which, I mean, as you can imagine, the answer is no, you can't ask a man about that and expect an honest answer. It's not going to work. And then, of course, he managed to get in touch with Dan Richardson. And Dan, I mean, he's an absolute genius, was like, yeah, I think I can find a way of doing that. So he came up with the most ingenious method. he told the participants they were taking part in a multitasking experiment where they were interleaving general knowledge with shopping decisions so you'd get a question and then you made a shopping decision and then a general knowledge question then a shopping decision so on and so forth so hidden in the study was this false fact about the average size of this part of man's anatomy.

20:11And this was immediately followed by a question about how much do you want to own this sports car with a picture of the sports car. That was the core bit of it. In two of the conditions, they weren't asked about the sports car. They were asked about a Rolex and a bottle of champagne. So that was to identify whether it was luxury goods in general or just sports cars in particular. And in the two conditions, in one condition, they were told an underestimate of average male size. And in the other condition, they were told an overestimate of average male size. And what they found was that men who were told they were below average found the sports car significantly more desirable.

20:56One of the bits that I find amusing here is that there's absolutely no effect on men under 30. They all want the sports car. It doesn't matter they just want it but men over 30 this is where this is where making them feel vulnerable changes behavior there you have it men who feel inadequate are more likely to buy ferraris assuming they're over the age of 30 chatbots do a fairly good job of interrogating witnesses anchoring made donors give away 70 more to charity ads in a pharmacy can improve sales by 10 even if your customers don't consciously notice. Memes make residents more likely to water plants.

21:34And purple made an optician's more competitive. And if you're over 30 and you have a sports car, why don't you pop a message in the comments of today's show to let us all know? I'm sure we won't judge.

21:49Massive, massive thank you to Jo for coming on today's show. If you want to get in touch with her or to do some work or hear more about Gorilla, here's where to reach her. Absolutely. If you're interested in using Gorilla within your agency, running any behavioral testing, then if you go to www.gorilla.sc, SC being Sugar Charlie for science, then you will find our website and you can contact us there. You can find me on LinkedIn as well. I'm Jo Evershed, I work at Gorilla. I'd love to connect with anybody, reach out to me there, can set you up with a meeting. Love to hear from you. You'll find links to both Gorilla and Joe's LinkedIn in the show notes.

22:29Now on today's show we've covered seven insights about human psychology but of course there are many many more insights that you can use to improve your marketing. I have spent the past eight years reading studies, books and peer-reviewed papers to find these insights and I've documented them all in my new product, the Nudge Vaults. The Nudge Vaults contains over 600 insights into the science behind great marketing. Here's Martin, a behavioural science practitioner talking about why he uses the nudge vaults. The main thing that I really like about the vaults is that Phil has done all the hard work for me.

23:05Research and applied examples just used to be in my head, which is unreliable, or buried in documents somewhere. So having everything in one place not only saves me a lot of time, but allows me to pull out the relevant examples for my training clients. And because Phil's continually adding to the vaults, there's always something new to learn. So thanks very much, Phil. I'd heartily recommend the vaults to anyone. If you'd like to get access to the Nudge Vaults yourself or just learn a bit more, then go to nudgepodcast.com forward slash vaults. That's nudgepodcast.com forward slash vaults, or just click the link in the show notes of today's episode.

23:43Once you click that link, you can even preview your first 50 insights for free. So you can try it out for free and see if it is for you. So please do go and check out the Nudge Vaults. That is all from me today. Thanks again to Joe for joining me on Nudge. I'll be back next Monday for another episode. Cheers.