← The Brainy Business | Understanding the Psychology of Why People Buy | Behavioral Economics
The Brainy Business | Understanding the Psychology of Why People Buy | Behavioral Economics
540. The Art of Experimentation to Gain Consumer Insights
00:42:49
In this episode of The Brainy Business podcast, Melina Palmer welcomes Dr. Henry Stott, co-founder of Dectech, to discuss the nuances of business experimentation and how to better predict customer behavior. Henry shares insights from his extensive...
Loading summary
Transcript68 lines
- [00:00]
Melina Palmer
Welcome to episode 540 of the Brainy Business Understanding the Psychology of why People Buy. In today's episode, I'm excited to introduce you to Dr. Henry Stott. Ready? Let's get started.
- [00:18]
Podcast Announcer
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.
- [00:34]
Melina Palmer
Hello.
- [00:34]
Hello everyone. My name is Melina Palmer and I want to welcome you to the Brainy Business Podcast. When's the last time you made a business decision that felt obvious only to find it didn't land the way you expected? Maybe it was a pricing tweak, a revised email flow, or a new feature rollout. You tested it, maybe even surveyed your customers, but the results didn't line up. Why? In today's refreshed episode, I'm joined by Dr. Henry Stott, co founder of DEC Tech Tech, who brings a refreshingly rigorous yet pragmatic view to this kind of business experimentation. Henry and his team have spent decades helping companies design smarter trials and better predict how customers will actually behave when real money or effort is on the line. And spoiler alert, it often looks nothing like what we predict in abstract planning. What I loved about this conversation is how nuanced it gets. Henry doesn't just talk about building tests, he he talks about when to use different kinds of experiments, how they blend behavioral science with commercial goals, and how even well intentioned trials can be derailed if you aren't deeply considering the context and delivery of your experience. As you listen, I invite you to consider this. Where might your data or even your past experiments be misleading you because they don't reflect the actual pressures or environments your customers are in when making decisions. Now, really quickly, before we get into the episode, I want to be sure you know there are links in the show, notes for my top related past episodes, books, ways to get in touch, and more.
- [02:08]
It's all within the app you're listening.
- [02:10]
To and@the brainybusiness.com 540. Now let's jump right in. Dr. Henry Stott, welcome to the Brainy Business Podcast.
- [02:21]
Dr. Henry Stott
Thanks very much for having me.
- [02:23]
Melina Palmer
Yeah, so I am super excited to talk with you today about the work that you've been doing in applying behavioral economics and behavioral science through Behavior Lab and DEC Tech. But if you can give just a brief introduction, tell us about you. I know you've been in the field for quite some time, have a very exciting and interesting past. If you can tell us about it.
- [02:44]
Dr. Henry Stott
I'll try and live up to the billing, but founded DEC Tech about 20 years ago with my colleague Professor Nick Chaytor, who I met at Warwick University. And back then we were interested in how you applied behavioral quite early adopters in that sense of how you apply behavioral science, behavioral economics to commercial problems. My background before that was I was at a strategy consulting firm called Olafel Wyman, which I joined when it was 20 people in New York City. My first job, I only had two jobs, that one and this one. So yeah, we got into doing that and have been spending basically the last 20 years refining our view on how you harness those methods to more accurately predict how consumers and other people will behave in novel situations.
- [03:33]
Melina Palmer
Right. And so for novel situations, what do you mean? As far as I know, you've done some very interesting work in delivery options and for clients like banking and petrol, like all kind of across the board. So do you have some examples of types of problems that you've worked on and solved over the years?
- [03:56]
Dr. Henry Stott
There are in effect forecasting the future. There is sort of two classes of problem from my perspective. And the first class of problem is a precedented problem. So where you see a lot of data science, research and so forth, what you've got are people who are collecting data on the behavior of consumers or whatever they're collecting data on within a given environment and then perhaps they're trying to forecast what will this other, this person do in this situation. So if I change the price, who's going to churn out of my subscriber base or whatever it's going to be? And that is that sort of first class of problem. So it's sort of precedented problem and you're trying to forecast the future usually at a sort of micro level based on those, those precedents. And then the second class of problem are the sort of the harder ones and in many ways I think the more valuable ones, which are just there's never been anything like it before. And those class problems are sort of in the school of new product development or very large pricing architecture changes or new features or different acquisition journeys and so on. And those are the kinds of problems we've set out to try and understand. So I mean, we're good statisticians I'd like to think, and can do the data science stuff, but we don't really specialize in that. Our specialization is more trying to understand how will this person, consumer or indeed other kinds of, you know, sort of citizens in different situations, how will they react to this completely new decision environment.
- [05:32]
Melina Palmer
I know I got some information and it talked with some of your staff about the Deliveroo project, which had quite a few options I guess, as far as different execution pieces. Can you maybe walk through a little bit of what you did about on the work on Deliveroo? I like as well that where I was provided with some information about translating the challenge into the research questions that you were asking and kind of how that executed into the way that you created the study there.
- [06:07]
Dr. Henry Stott
The problem they faced was sort of fundamental questions about their subscription architecture or their pricing architecture. And the question was whether they were going to launch a subscription product or not. Mean in effect a sort of Amazon prime type strategy. But that didn't exist and there were various commercial levers they could pull in the execution of that which they wanted to understand prior to making a go no go decision on doing that kind of proposition. So they had that problem. That is not the kind of thing that they felt comfortable. I mean, I can talk later about other research methods, but it's not the kind of thing, for example you could a B test, you couldn't just launch it because it didn't exist yet. So we use our normal approach of an immersive randomized controlled trial in a sort of paid environment, which again I can elaborate on later, but we used that method to explore how consumers reaction changed as you changed the proposition. So in that sense the subscription product we were offering alongside the normal product, but the subscription product in the randomized controlled trial sense had different features attached to it. So it might add, might have a different spend requirement, it might have a different sort of special promotion associated with it. So you might be, restaurants might be offering things that are specifically just for a subscriber like that. There might be features to do with what the delivery specification was like how, you know, how tight is the delivery window and that kind of thing. So we put through a whole range of different variants of that design to see, you know, to answer the two exam questions. One is which is the best design? And the second one is what are the economics of launching the best design? And in the end they launched it, you get a 14 day free trial, then it costs about 11 pounds a month, minimum order of 10 pounds, et cetera. That's in the UK and it's called and they branded it Deliveroo Plus. And you know that all that design has come from the optimization within that sort of behavioral experiment that we undertook prior to that launch.
- [08:26]
Melina Palmer
So you were testing different, so all the features remaining the same. But say if you only call out one main feature versus four or five. And if there was sensitivity to price, where I'm seeing a version here, what I really liked about it is you didn't just go with the lowest price, which is where a lot of people feel like you have to go with the most affordable. That's the only way anyone's going to buy. And so you feel like you have to make it the most economical as possible. But you know, I'm seeing a mock up where it shows it at 799 pounds per month and it ended up being 11. And I'm sure you tested a wide range to see where that price sensitivity was. And it definitely goes back to, like I say, pricing is not about price. Everything in the way that you present it before you get to the price matters more than that actual number. And I like that that shows if you don't have it make it so complicated and make it a real easy buy option, then it's still a great value and people can actually see that more when you present the options in a better way.
- [09:32]
Dr. Henry Stott
Yeah, I absolutely agree. I think, I mean, yes, we did. We tested a number of features and we also tested variants of the features. So different delivery specifications, you know, does it arrive within an hour, within 30 minutes, within whatever, you know, et cetera. So, and I think the point you're making, I mean, I have a lot of clients who are being ground down, for want of a better word in a, in commoditizing industries where they are on some kind of increasingly tight margin, lower price point because they are in effect failing to innovate value added features. So it is important, I think in this kind of work to not just look at price elasticity to the point you're making, but to look at that in conjunction with what the proposition is and to understand that there are features within everybody's proposition which the customers want and features that they don't and also to some degree heterogeneity. So features that some people want and some people don't. So the designing of a proposition so that it fits snugly into what people want is the first objective. And I think a lot of people fail, a lot of people fail to do that and as a result they end up just fixating on the price and inevitably they have to drop that price because they haven't been adding value in the, the, the ergonomics and the, and the attractiveness of the actual core, the core, core proposition.
- [11:07]
Melina Palmer
Right. And that's if you just are looking at what your competitors are doing or how you're going to be able to compare with them. That's, you know, say an Uber Eats, which I don't. It sounds like the monthly subscription for Deliveroo came before anything that UberEats might have been doing.
- [11:26]
Dr. Henry Stott
Yep.
- [11:26]
Melina Palmer
So then maybe if Uber Eats was looking and saying, oh well, they're charging £12 or £11 so we have to do 10 and that you feel you just have to. But maybe you don't, you know, maybe your offer, your offer is different and if you don't dig in and really look at and test, you have to compete on price because you just haven't taken the time to look at your, like you said, your value proposition, what it's about and how you can appeal to a different subset of customers or maybe even the same people. But it doesn't have to just be a feature or a price thing. If, if somebody's offering a feature that would be really detrimental to your business model, it doesn't. Just because they offer it and showcase it doesn't mean it's what people actually care about.
- [12:07]
Dr. Henry Stott
I think that is true. I think a lot of the design of features and propositions, et cetera, is driven by an excessive fixation with competitors and a excessive self confidence on the part of management that they know what the customers want. And usually both those things are useful but probably overweighted. We worked for a long time with Tesco in the UK and their main competitor internally was deemed to be Asda. And so Tesco spent a lot of time worrying about where they sat against ASDA on pricing across. A typical supermarket store like that would have 40,000 lines in it. So there was an amazing amount of activity to manage price against Asda. And what that then led to was a strange parlor game whereby ASDA would price up by a few pennies, a few cents and then drop them and Tesco would chase them up and chase them down, et cetera. And the punchline to this is that when you delved into consumer perception of the price proposition at Tesco, you discover that the chasing of ASDA was deeply undermining of that price perception, because the majority of people shopping in Tesco were not shopping at asda. The majority of people shopping in Tesco were shopping in Tesco. And the corollary of that is that the price that they were referencing when they were judging the price that they were being offered this week was the price they paid last week. And this strange parlor game of the price going up and down to chase Asda was just creating a is undermining the credibility of the prices. And of course behavioral effects like loss aversion, which many of your listeners will be familiar with, mean that when you put your price up, people are more annoyed than they are happy when you drop it again. So the whole cycle of like up 10p this week, down 10p this week, up 10p this week, et cetera is a net negative pump on customers perceptions of your price. So then I think that's, that's then a sort of really important point. Looking to competitors the whole time is, is a mistake.
- [14:43]
Melina Palmer
Right. Well and for a lot of people, like you said it for grocery, you go, you have your store that you always go to unless they randomly don't have that thing that you really care about. But you know, that's the place where you go for your shopping. And so I'm not looking at the other. Some people do a lot of that price competitive shopping. But in general you've got a brand loyalty that comes up which might just be based on proximity to your house. So when you move it's going to change, you know, something like that. But definitely I'm, I'm not watching the prices in the competitors catalog actively enough that well, and even you know, going beyond grocery, so many businesses do that where we think people care about us and have looked at and studied with depth everything that we've ever put out into the world. And so it feels like you need to reference something you did 10 years ago because someone's going to remember that it happened. But people just don't. And like you said, they barely even remember what the price was last time that they came in. Except if it's constantly doing this up, down, up, down, then you really break outside of that natural habit cycle where my habit is to come in. I always go to Tesco, I always buy bread, milk, whatever. And if it's basically the same price I just sort of go and no worries. But when it's bouncing up, down, up, down, up, down, then I'm going to notice and think about if I should go somewhere else.
- [16:14]
Dr. Henry Stott
Yeah, no, I think that's right. I think the movement of prices definitely orientates people to pay attention to it. I mean they call them known value items and it is the case, I think that across a 40,000 range you might be regularly purchasing 100, 200 items. So you're not going to remember all those prices but there's a couple you'll probably know and you may well clock those prices when you're in other places. And that will then give you a feeling of the price competitiveness of that supplier, and you may then disseminate that impression. So there'll be some kind of social proof coming out of your price positioning like that. But I mean, at least half of that judgment is made up of movement of prices within a store, as opposed to contrasting across stores, as you say. And so that's been really important. I mean, the other flavor here that's sort of also relevant is people's sensitivity to the order of pricing rather than the magnitude of differences. So if you go back to the delivery example, in the Uber eats example, UberEat may be at a price point and then Deliveroo might be at a higher price point, but the sensitivity of customers to the size of that discrepancy is, I suggest, relatively weak. I mean, we tend to find that people are much more sensitive to the order of things than they are to the size of the differences. So if they're £2 dearer or £3 dearer, you know, that doesn't make such a big difference. And there's a diminishing sensitivity to that. So it's just the really big call there is, are you going to be below match or above? Then? If you're above, then how much you're above is kind of not. Actually, people are not going to notice that much. They're just going to notice that you're above. And you can be above if you have better features. If you have features that I want.
- [18:18]
Melina Palmer
Right. Maybe that it arrives quicker or that's where we've been having to use. UberEats is what's near us here in. In Washington State. And so we've been experimenting with that. But all the restaurants are way on the other side of town. And so we do get cold fries and, you know, whatever when things get delivered to us here. And so if there was some sort of a guaranteed quick delivery, it would be worth paying more for when we live way over on the side of. Of town.
- [18:47]
Dr. Henry Stott
Well, that's right. Yeah. I think that there's all kinds of features that people will pay for, like the company I was reading about the other day who are designing pizza delivery trucks that cook the pizza on the way over.
- [18:59]
Melina Palmer
Ooh.
- [19:00]
Yeah.
- [19:00]
Dr. Henry Stott
It literally arrives out of the oven using all kinds of clever automated driving and so on. So, yeah, I would pay more for that pizza. And so that was a great role for innovation.
- [19:13]
Melina Palmer
Yeah, absolutely. I was in a behavioral science book club thing on LinkedIn. Rory Sutherland joined for one of our book club events and was talking about. And he mentioned that the Thing about where people don't want to be waiting. The problem with the delay is not necessarily the length of time, but it's in that, like, boredom where. Where there's nothing going on and you become impatient. That's the pain point, really. And so even things like, I think it was Domino's, where they made their pizza tracker, where you're able to look and see what's happening at each step, or where you can see that, you know, Amazon says that they're loading onto the truck now, or something's happening, which is maybe a little bit different than what's actually happening in that exact moment. But it makes you feel like you're part of the process and you can see, see those updates. So it's not as painful of a wait.
- [20:04]
Dr. Henry Stott
Rory is citing some, well, precedented results there. I remember Even, you know, 30 years ago, discussions about how it made sense to put mirrors in elevators so you could check your tie or whatever, and it would sort of pass some of the time as you move from floor to floor. But rather more seriously, the Design Council in the UK had a great case study like that, which dates back about 10 years, I think, where they were looking at violence in accidents and emergency, whose American name is just escaping me. But anyway, in hospital, where you go in, you're going in injured and violence there was in part being mediated, they felt, by people not knowing where they stood in the process, and the process taking a long time. And so they did a randomized control trial where they introduced more signage in some hospital A and E units that gave people a better sense as to where they sat, how long they were going to be around, where they were in the process, what the next steps were coming on and so on. And they saw statistically significant decrease in violent incidents in those hospitals. So it's an excellent point.
- [21:17]
Melina Palmer
Right, well, and that's our perception of time changes so much. Where people would wait for hours and hours and hours for, you know, tickets to a concert they were excited about or for the new iPhone when it was going to come out. But if you were at the bank and having to wait for five minutes, it's just excruciatingly long, unless there's some version of, you know, distraction that's able to help you get through those moments. So knowing whether somebody's excited about what they're getting or just feeling like it's a mundane thing that they have to deal with. I have a lot of financial institutions that I work with and know that that is a big perception issue and just something where even where you can say, you know, we went down from 10 minutes to 5, everyone should be so happy. But that's just not how it works. And if you're not asking the right question, if you're not looking at the real problem, it's still going to be an issue. Even if you shave, you know, 30 seconds off, it's. It's just not going to fix the issue. That's, you know, for, you know, Disneyland, where the waiting in line for hours for the ride, they have a lot of distractions so that you feel like it's part of an experience before you're able to actually go on the ride itself.
- [22:26]
Dr. Henry Stott
Yes, no, that's right. I mean, duration neglect is the term of art. I think you're right. I mean, any amount of time spent in a queue in the bank is part of your life you're never going to get back. And so at some level, we are all aiming to get that down to zero. And the difference between 5 and 10 minutes is perhaps not that great. And the, I mean, there is a curve. The curve climbs steeply and then starts to flatten off. But on the flip side, you know, there are other goods where time delay can be a good thing. You know, there is a psychology literature about savoring.
- [22:59]
Melina Palmer
Right.
- [22:59]
Dr. Henry Stott
So these are, these are features that you can manage and then test.
- [23:03]
Melina Palmer
Right.
- [23:03]
Well, and that's the. Why it's important to test to see which ones work for you. And just because you think that this behavior is going to be one that's important, maybe, maybe it's not as important as something else. You know, as we're talking about banking, it seems like it makes sense to talk about. I really love this case study from Lloyd's of what, what you had done for their online buying experience. I know for financial institutions, not buying in the same way, but conversion, I guess, online conversions and the really great results that they had in going through that customer journey optimization process. Can you talk a little bit about that project?
- [23:41]
Dr. Henry Stott
That's interesting study in that it's sort of similar methodology. So it's an immersive randomized control trial approach where you recreate a facsimile of the journey that is potentially offered to a customer. You get paid participants to come through and experience that journey. But it's a randomized controlled trial. So different people experience different versions of the journey and you see which journey is most effective across a, a variety of metrics of interest. So I mean, metrics of interest might be conversion or in effect, revenue measures. But I mean, it can also be to do with how satisfied people are with experience and the degree to which they actually understand what they've purchased at the end of it from a regulatory perspective. So sort of like the outcomes of interest can be quite wide ranging in these kinds of studies. So the particular one you're talking about with Lloyds bank, we were looking to design a home insurance renewal process that worked best for the customers, but also worked best for the bank. And there's a variety of levers you can pull. I mean, those that have studied their behavioral science will know that there's a variety of different effects that you might call upon in once you've studied this journey. And maybe we should not make this too complicated and maybe we should put a prompt here and we should also fill this, and we should, etc. Etc. And so you can take the kind of paradigm I'm talking about and have these ideas about how you could change the design fairly profoundly in a way that you wouldn't, you wouldn't want to do those changes until you were sure they were worth doing, is, I guess, my point.
- [25:22]
Melina Palmer
Right. You don't want to do that with a live group. Yeah. So if you put that out and it's a bomb, then you.
- [25:30]
Dr. Henry Stott
Live with that.
- [25:31]
Melina Palmer
Yeah. You created a bad experience for a bunch of people and you can't get that back.
- [25:35]
Dr. Henry Stott
Or indeed it's just very expensive to do or, you know, the actual building of feature. So, for example, one of the things we tested was the name, your price condition, which was not quite as it sounds in the sense that you put in a target price for how much you wanted to pay and then the bank would then provide you with the specification for the product with, you know, this size of excess and this amount of, you know, whether it's on direct debit or not and that kind of thing, and how much, how much content coverage you got. And it could tailor that proposition to hit the budget that you'd asked for. Now that that kind of infrastructure is, would be difficult to deliver. And so you would want to know that people really wanted that before you started trying to build it into your systems. So it's not an a B test problem. It's, and it turns out actually wasn't that, it wasn't that popular. So we save them having to do it. But I think these kinds of facsimile environment like that, it's like a B testing in a lab, in a lab environment. And so you create these environments, pump people through them, lots of different variants, and you don't have to AB test. You can A, B, C, D, E, F, G, H test, right. And cover a lot of ground. And then. And then on the back of that redesign your experience with confidence of knowing that it's optimized relative to all the myriad of different things you could have done.
- [26:56]
Melina Palmer
And so I know that you do a lot of different tests for different types of companies and different problems. And so there is no set. You know, it's not like somebody comes in, you say, well, this is the box solution that you get for, for your issue. But do you have a typical. Like you said, it's not an AB test. So it's not. You're only testing two conditions. But do you typically test, you know, five or 50 or, you know, sort of. How does that even work? Let's talk about the example of a test. So if somebody comes to you with a problem and then you know how, how that kind of gets worked out in determining like average number of variations and things that you can test kind of how that would work.
- [27:40]
Dr. Henry Stott
People come with different levels of problem in the sense that some people might arrive and say, I'm losing 10,000 accounts a week. What do I do about that? And therefore there's a sort of wide range of levers that one might pull to solve that kind of problem through to we're having an argument about what we should name this product and who's going to buy it. Then that's. Then you might describe that as a very tight brief. And then across these two levels of problem, you would also have things that might vary continuously, like price. So you might want to test a variety of different price points through things that are more 01. So you do or don't have this feature. You do or don't offer people to name their price. But typically, one of these experiments, we'd be running 20 variants and some of those variations would be continually varying things that might vary across different levels. I mean, the nearest analog, actually, which some of your listeners will be familiar with, is conjoint or max diff testing. But I take issue with that approach because there's lots of evidence that context is extremely important in decision making.
- [29:05]
Melina Palmer
Right.
- [29:06]
Dr. Henry Stott
As you'll know, and Conjoint and maxdiff are methods of elicit, eliciting preferences that are so divorced from any kind of decision environment that someone would naturally encounter. That, from my perspective, and this is borne out by our research, really, is that the results you get in terms of people's preferences out the other side of those kinds of paradigms is heavily Distorted. So, you know, from our perspective, the best way to get to these things is to immerse customers in an experience, the decision making environment, which is as close as possible to the environment that they would naturally encounter. Because only in that way are you going to get something that resonates with the way they're going to then behave in the real world. So it's sort of like Conjoint because you're testing this, that and the other level. But it's not like Conjoint because it has this realism dimension. And actually, while I'm on the subject, the other reason it's not like Conjoint is it's more than just preferences. Because when you immerse people in these trials, you can capture everything about them. And going back to the Lloyd's example, you could figure out not just what your conversion rate is, but you can see where people are dropping out. And you can see whether people have remembered what they purchased. And you can see whether at the end of it they said that was really painful process or they said that was a really easy process. And I understood everything was going on and I felt entirely happy with what was going, what happened. And you just can't do those sorts of things, which is why I think this sort of immersive lab approach is better.
- [30:46]
Melina Palmer
Right. And so can you explain a little bit about what it means to be in the immersive lab? So are you doing, I know that doing work with the Human Behavior Lab at Texas A and M University. And they have the eye tracking and the skin response and the EEG and whatnot. Do you use those sorts of techniques or do you have other things that you're using in your lab?
- [31:06]
Dr. Henry Stott
We certainly use those kinds of things and I like them a lot. I think though, when you're testing 20, 30 variants like this, you need to do it at scale. So I mean, typical study by US would run 5,000 people through a study. So at that point everything's got to be automatic and no one's traveling into an office, into a laboratory to do it. So what we would tend to do is create a online decision environment, recruit 5,000 people, paid participants to come along and start that experiment. And then they would, you know, each person would have, in a randomized, controlled trial sense, have a different experience as they go into that environment. They'd be offered different kinds of things at different price points, perhaps given different information earlier on about the market, context, exposed to different kinds of advertising, editorial, told things about competitive pricing or not, etc. Just anything you Want really, you can put whatever you want, but our experience is that you can get quite close to replicating the kind of mindset people will be in using this kind of approach. And so you then get things that replicate extremely well. Our clients tend to go on and then do the things that we advise them. And it's always very gratifying to see how closely their experience aligns when they launch something for real with the advice that they've got from doing these kinds of trial. But, yeah, these trials have to kind of be run entirely virtually just. Just because of the industrial numbers of volumes. You need to. To explore all the different permutations as well.
- [32:46]
Melina Palmer
Right. Well, and I think clearly that is something that I'm sure has worked out very well for you in a pandemic world where there are a lot of labs that are finding themselves with difficulty and problem right now when they can only do things in the lab. So those that I've talked to that have a virtual or online experience, and you are able to basically business as usual at a time where people are looking to test more because they're realizing they need to make changes. You know, it's a one of those, you know, hindsight bias, right, where we're looking back and say, man, that was a great decision that we did this this way. We must have known.
- [33:21]
Dr. Henry Stott
The steady march towards digitization is, of course, sort of in our favor in this regard because more and more commerce is taking place online, and therefore virtual testing laboratory formats like this will be increasingly relevant. So in that sense, I think the pandemic has accelerated that trend, and that's got to be good. But obviously the pandemic in other ways has been spectacularly disruptive. So Net Net, I bet I'd be happier without it.
- [33:55]
Melina Palmer
Yes, yes, for sure. But going back to the way, as far as, like you said, having customers implement the recommended changes, and so looking at you tested, like you said, 20 different variants or more for this Lloyds bank example and ended up with, I'm looking at, you know, an example of what they implemented, showing this, you know, a personalized journey, having a different, you know, framing of a 40% introductory discount. And then we have default choices and making it simple and having choice architecture and a different tiering or relativity set up in there and saying, you know, you tested it and saw these are the things that are going to work, and that their acquisitions went up from 45% to 75%. Yeah, pretty great.
- [34:43]
Dr. Henry Stott
Pretty punchy. Yes. Yeah.
- [34:45]
Melina Palmer
So we did the tests, we looked at all these things, and if you can say across the board, we're converting another 30% of people.
- [34:54]
Dr. Henry Stott
Yeah, just.
- [34:55]
Melina Palmer
There you go. Right.
- [34:58]
Dr. Henry Stott
Why would you look twice? Yeah, that's right. No, I think, I think, I mean, I mean this is a point you've made on previously on this podcast, is that remarkably innocuous small changes can have profound effects on the economics of your business. And so that's great because it means there's all this opportunity out there. But that's painful because you have to kiss a lot of frogs where you find that subtle change.
- [35:29]
Melina Palmer
Right.
- [35:29]
But if you can do it in mass quickly via system like this, then.
- [35:34]
Dr. Henry Stott
Yes, and to the earlier point, I mean, there is a lot of behavioral science out there and a lot of great behavioral researchers that have gone before us who have outlined the kinds of effects you can look for.
- [35:49]
Melina Palmer
Right.
- [35:49]
Dr. Henry Stott
So it's, I mean, you know, often a lot of innovation is about generate test cycles. And you know, I've been describing what I regard as the higher fidelity way of filtering and testing and getting the, getting the nugget changes the best strategies out the other side. But of course the usefulness of that is limited also by the quality of the generate process. And you know, often find a lot of my clients lack in that department in that they are often quite timid or they want to test things that have been around for a while and they're just refusing to give up on for some reason, etc. Whereas I think a better version of this process, and one that we tend to try and pursue too, is before you start testing, you think of the best ideas and you, you, you try and innovate. And the behavioral science literature is extremely helpful in that regard.
- [36:46]
Melina Palmer
Right. A lot of people I've been speaking with recently, as well as I've been interviewing a lot of people in behavioral science, behavioral economics, as we're doing this applied work is often the best thing about testing is the thing that you kind of throw in where you say this crazy thing we came up with that probably wouldn't work, but let's just see if we do a, we put in a physical gold coin or if we have this, you know, something with an atm. I was talking with Steve Wendell recently who wrote Sign for Behavior Change, talking about they had an ATM that's eating people's cash to help them not want to pay for an ATM fees. And it just those things that get thrown in that you think, let's just see, it probably won't work, but it performs exceptionally well above what you think should work, which I think is just a testament to the importance of testing and throwing in things that seem really counterintuitive, that can really do something amazing. Have you had any experience like that?
- [37:45]
Dr. Henry Stott
I totally agree with this point in that I think that a lot of people, particularly in retail environments, are very bored a lot of the time.
- [37:54]
Melina Palmer
Right.
- [37:55]
Dr. Henry Stott
I could cite a number of studies we've done where one of the outcomes was to recommend that you shouldn't just be doing this. What you should be doing is actually something different the whole time. That the change in its own right is the good thing. Thinking of, say, a retail example, a good sales lever. Are these kind of featured weeks where you said, this week I'm doing this, this week I'm doing that to the point earlier, it's pretty bad to have a failed mission because some basic thing is not in stock or you have some basic problem or you've messed around with the pricing or that kind of stuff. But on the other hand, people are also looking for something interesting and enlivening about their retail experience. And so something novel, something fun is quite often one of the outcomes of the experiment. We were looking at a lottery provider here in the UK and one of the things we tested that worked quite well was not picking numbers, but picking fascia to those numbers. Those could be emojis or those could be animals or celebrities or whatever. And you know, potentially you could skin the lottery ticket that way. People, when we tested it, people did use that feature and they did rotate from week to week as to which skin they used and all the rest of it. People like a bit of novelty.
- [39:17]
Melina Palmer
Absolutely, yes. Novelty and story have a huge impact on the brain that I think is. And it makes our jobs more fun. Right. When you get to test these fun little things. So we talked about so many amazing things. You obviously work on some fantastic projects and have had wonderful results for wide range of clients. So for anyone who's interested, or for all those who are interested in talking with you, what's the best way to get in contact with you or the team at DEC Tech and the Behavior.
- [39:47]
Dr. Henry Stott
Lab, Dectex on LinkedIn. So if you followed us on there, we publish briefs about twice a year reports twice a year and those cover some of the things we've been discussing here. The next one is on the role of social influencers and the previous one was on pandemic behaviors and what's going to happen post lockdown to the retail landscape. You can stay in touch with us that way. And there's a website and I'm on there and as is my email, so quite easy to get in touch with.
- [40:22]
Melina Palmer
Perfect. We'll make sure to link to both the LinkedIn as well as that website in the show notes, so it's very easy for anyone to find you and the great projects that you're working on. Well, thanks again for joining me today Henry. It was wonderful to chat with you and look forward to seeing more from Behavior Lab and Deck Tech in future.
- [40:42]
Dr. Henry Stott
Great. Thanks very much.
- [40:45]
Melina Palmer
So what got your brain buzzing as you learned from Dr. Henry Stott today? For me, what really stood out was the conversation around the risk of using sanitized test environments that unintentionally remove friction or complexity, making it too easy for someone to say yes. That kind of clean data may look good in a report, but it won't predict real behavior under pressure. I also appreciated Henry's pushed to think more critically about when and where behavioral nudges are applied and that sometimes simplifying the journey means removing a nudge that felt clever but adds noise. Whether you're testing a new offer, onboarding flow or subscription model, the question isn't just does this work? It's does this still work under real world conditions? So as we close out the show, I invite you to consider this. What assumptions are you making about your customers experience that might not be true? How could you test and validate those before making your next big move? Come share it with me on social media. You'll find me as the brainy biz pretty much everywhere and as Melina Palmer on LinkedIn. There are links in the show notes to make it easy as well as links for my top related past episodes and books, ways to get in touch, and more. It's all waiting for you and in the app you're listening to and@the brainybusiness.com 540. And just like that, episode 540 on optimizing customer journeys with Dr. Henry Stott is done. Join me Thursday for another brainy episode of the Brainy Business Podcast. It's going to be a lot of fun. You don't want to miss it. Until then, thanks again for listening and learning with me and remember to be thoughtful.
- [42:32]
Podcast Announcer
Thank you for listening to the Brainy Business Podcast. Molina offers virtual strategy sessions, workshops and other services to help businesses be more brain friendly. For more free resources, visit thebrainybusiness.com.