352. Unlocking Consumer Behavior with Imotions: Neuroscience Insights for UX Designers (Refreshed Episode)

Welcome to episode 352 of the Brainy Business Understanding the Psychology of why People Buy. Today's episode is all about research and testing with eyemotions. 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 Melina Palmer hello.

Hello everyone. My name is Melina Palmer and I want to welcome you to the Brainy Business Podcast. Today's episode is a fun dive into research and an experiment I was honored to be able to do with imotions to support the launch of my first book, what yout Customer Wants and Can't Tell youl. It's hard to believe, but this episode is being refreshed today because of a new test that was done to support the launch of my first third book, the Truth About Pricing, which is on pre sale now and will launch in just a couple weeks now on January 9, 2024. The test you'll hear about today was around the website launch to support what your customer wants and can't tell you and some really cool behavioral testing we did with eye tracking to help optimize that website page. This upcoming Friday's episode is all about the COVID for the third book and the different rounds of testing that went into the final decision that was made. It has been such a fun ride and I'm so thankful to everyone who has supported me over the years from buying a book to having me out to speak to support on tests like these. It's all so awesome. And I'm so thankful to everyone who has supported me over the years from tests like these of course, but also to everyone who's ever bought a book or just listening to the show, sharing it with a friend or having me out to speak at an event, all of it comes in together. It's so awesome. I thank you all. It all means so much to me. Thank you, thank you, thank you. So you know there are links for everything including past episodes, books, articles and a full time stamp summary of the episode all waiting for you in the show. Notes which are found within the app you're listening to and@the brainybusiness.com 352 all right, let's talk about this website experiment and all the great stuff that iMotions does. Dr. Jessica Wilson, welcome to the Brainy Business podcast.

Thank you so much. I'm excited to be here.

Absolutely. Well for me we were doing in our little like Pre chat, you know, talking. That we've known each other now for two years, maybe, I think as we've talked. Has it really. It's been at least 18 months.

Time's flown.

I know. I mean, it feels like it's been. I know it's been at least 18 months since we were doing the first study, really, that we worked on. And so we had to have been talking and met before we actually ran that. So at least 18 months. Probably going on maybe not quite two years, but it's been a while. Yeah.

Yeah. I think.

I think.

I think with COVID you just ended up with like this blur in between where, you know, you don't really remember what happened, but stuff happened. So.

Right. Yeah, it was somewhere, I feel. Yeah. Like maybe we started talking right in that pandemic timeframe too. Right when that all started. So who knows?

But.

So I know you. Great to see you again. But can you tell the audience a little bit about who you are and what you do?

Yeah, absolutely. So I am the technical director at iMotions, their North American branch. So my background is in research and in neuroscience. A lifetime ago, I was actually studying Parkinson's disease, so doing a lot of muscle activity and working with participants, and I really enjoyed it. But I really thrived on understanding how these technologies can be applied to different verticals, different industries, which really led me to iMotions. So we're a software company that provides this platform that can integrate and synchronize different neuroscience tools. We have a study builder, platform analysis tools, and export options, and we sell equally to commercial and academic clients. So it's my delight to get to meet all sorts of people from all over the place, studying different things and see, okay, how can we sort of bridge the gap between what I know and what you know? And how can we use neuroscience tools and leverage neuroscience knowledge to advance your research and answer some really interesting questions? Absolutely.

And so I first got introduced to emotions because the Texas A and M Human Behavior Lab is powered by emotions for, you know, the vast majority, I believe, of what's going on there in the lab. And so the first time I went there, which, you know, that's been three years ago now for my first visit almost of trying out the eye tracking with the glasses and all the different software and stuff. And so we have talked a little bit about the types of studies that your software does with eye motions. But can you expand a little bit on kind of what types of testing imotions does from the perspective of the actual research, like doing the work, I guess, the scientists on your side, what the studies are, what sort of things you test and do.

Yeah, it's almost a really difficult question to answer because there's so many different applications and our client base is really all over. When I first joined the company, I was really interested in consumer neuroscience and consumer insights. So this can be everything from product testing, ad testing, image testing, and even sensory work. But what really struck me after staying at the company for as long as I have is the breadth of applications people are interested in. So automotive is one, gaming architecture, even design, military, defense applications. There's, you know, the list really goes on. So I guess I always like to say if there's any instance in which you're recording something in terms of behavior, you just want to see someone's reaction to stuff, then usually I. Motions has a place in that. But if we're talking about sort of specific research questions, a lot of what we see are video testing. So how do people respond to different kinds of creative and how do people respond physiologically to that creative. On the usability side, which we'll get into in a little bit, how do people interact with different websites or different devices? Where do people attend to, which is where a lot of the eye tracking stuff comes in. But then also, what is the nature of people's reactions? Can we see how confused someone is with an interface and does that relate to the subsequent behavior that we see whether people follow through on a purchase or manage to complete a task? Then on the academic side, we have clients working in communications, education, business and marketing. And that's really, I think, where a lot of the interesting questions come from. And I know working in the Texas A and M lab, you've seen your fair share of that too. Yeah.

And so as far as the types of testing though, we've got, so we've got the eye tracking like you talked about. There's some facial recognition. So we've talked about if people are going to do a more basic, like, you know, you go out into Internet land and say, well, I'm going to look and have just some eye tracking and see where there's awareness on, you know, like people were looking at this for a really, really long time. So that means it's good. Like yay. Yes. People are looking at it, say, well, hold on. Yeah, not necessarily. If you don't have the facial in there, you know, so we don't. We talk about brow furrow.

Yes.

So if I'm looking at it but have this face like, huh, Then it's really bad that I'm looking at it for a long time because I'm probably confused. And so that's something that you would want to tweak. And so a heat map without facial stuff isn't going to give you as much information. What's the full ish gamut of study that could be done with a lab that is powered with EyeMotion's software of different types of tests?

Yeah, absolutely. I think your most basic setup would be eye tracking, facial expression analysis, which we've talked about, but then also skin conductance, also known as galvanic skin response or electrodermal activity. The three of those work really well together because you have complementary pieces of information. You have eye tracking, which gives you the attention aspect. Facial expression analysis will give you a sense of balance. What is the nature of that reaction? Is it positive or negative? Then skin conductance gives you a sense of physiological arousal. What is the intensity of that reaction as can be measured physiologically? All of that. Those three work really well together. I call them the holy Trinity when I'm talking about them. People who are unfamiliar with this stuff because they work really well together. And then also in terms of a learning curve, they're all very easy to get started with on the biometric journey. But once you decide to go further down the rabbit hole, then there are many more tools that we can discuss. So we can look at EEG electroencephalography, and that can get at some more subtle measures like engagement cognitive workload approach or avoidance motivation, which we see a lot in market research. You can get into heart rate and heart rate variability, which is another indicator of emotional content and stress. You can get into respiration even and many other different kinds of tools that are available. But I think the important thing though is the idea that the research question should dictate the technology and not the other way around. Often what I see is people get so excited about the technology and they want the shiny thing that they just want all of the technology available and they just want to have like a shotgun approach to their research question. And often I'll say, okay, you're really interested in eeg. What measures specifically are you interested in? And they'll say, everything. And I'll say, that's not an answer. There are a million ways in which you can slice and dice neural activity. Right? And so you really need to have some specific measures in mind. So I always like to emphasize these tools are really helpful in certain ways. But at the end of the day, the research question is king and That'll help dictate what tools you use. And part of what Imotions offers is that consultation to be able to sit down with you and say, okay, here are the tools that might best suit your question. Here's the kind of metrics you can get out of them. Here's what you should be looking for and to really make sure we have sort of the right plan going forward.

Yeah. And so we did actually two studies, tests together for my website at the Brainy Business, which was a lot of fun. And I know we're going to talk about that today. Kind of what we were looking at, what we found, why we did two different tests and how they were run and things like that. But if you were to give, you know, some examples of the. A typical type of a test, I know you're generally talking about, you know, someone comes with a problem and you help them kind of iterate and make sure it's a question that can be answered properly by a test. Because I know from working with businesses, like you said, that I'm excited about all of it. It's like I want to know everything I can possibly know. You say, well, if we don't go in with a specific question, you know, it's not that helpful. So how do you help with a business? You know, if they say we have a problem, like we need more customers, we need to see some lift in this product vertical, or we want to work on attention in the aisles or whatever it is, how do you work to turn that into a research question? What's like a type of a problem? And if you have a specific case study that you're able to share, that would be great just for people to kind of grasp what, what y' all do.

Yeah, absolutely. So let's take video testing as sort of our sort of a sample standard method here. Usually what people want to do is they're. They're interested in sort of what aspects of the video are most engaging that I can maybe leverage. Are there elements within the video that might be boring or would cause people to disengage, for example? Or another great method would be an AB test. So say you have two iterations of a commercial and you want to see which one will be more effective and which one you want to actually take to market, then Imotions works really well in being able to get those biosensor inputs for that AB test. So if it were me running a study like that, I would be using eye tracking, facial expression analysis, skin conductance, maybe eeg. I also wanted to get into approach avoidance, motivation, and sort of a willingness to take action on a product. But I would have people watch those two ads with all of the sensors attached. Maybe follow up with some surveys afterwards, such as willingness to pay. How much did you enjoy this commercial? Maybe some recall tasks as well. Did you actually recall what brand we were looking at? And what we can get is look at some key elements in the video. So if there's a joke, for example, that we want to see if the joke landed, we could look at facial expression analysis and skin conductance. To look at the effect of that on audiences, we could look at a decrease in skin conductance response as something that suggests disengagement or boredom. With an element of the commercial, I might be able to say, okay, for this particular prototype, we might want to rethink these 10 seconds here because it looks like some people are disengaging. We can subsequently follow up with the eye tracking and say, okay, did people actually notice the product that we're trying to show them? Are they noticing the logo when we put it in a particular place? How can we manipulate these things, these very specific elements, and then see how people respond physiologically to those elements? From that data, you can draw some general insights that can help guide your decision process moving forward. If you notice that people aren't noticing the logo, and you want that logo to be sort of a central focus of that scene, then you can explore with different placements and see whether people actually notice the logo and spend more time dwelling on it compared to distractors. If there's a joke that doesn't land or that people find confusing, then you can see that in the data, try to iterate a bit better, make a new design decision, and then follow through again and see if that improvement has helped. So being able to take that and sort of look at the specific details and see what decisions can we make to make this better? That's something that we can do pretty easily with video testing. And it's actually a method that we kind of recommend is to have sort of multiple versions of this testing through to the design process, from animatic, for example, up to finished prototype. That way, with the data that you're using, you can actually make sure you're making the best design decisions at the time when it counts, when you're actually designing the thing and not when you have a finished product. And then the biosensor data indicates that people don't like it. It's like, well, now you're back at square one, right? So that's One example of video testing. One example of a client of ours that's done video testing is actually the University of South Florida. This is one of my favorite stories. They actually have a sales and marketing lab in the MUMA School of. And they do a whole bunch of different kinds of testing, including video testing, other studies related to marketing, and they really have this resource that's open to the entire research community. One of the studies that they recently came out public with is a collaboration with Baycare Health Systems to actually look at the impact of COVID 19 infomercials and information on consumers. So Baker actually provided the health communications and creative that they wanted for testing. And USF ran these video tests to see, okay, what parts in this very timely, very important video are people attending to most? Are there instances where they might look confused, they look like they might not be paying attention? And are there ways then based off of this process and based off of this method of assessment, that we can actually improve our videos and make them more effective methods of communication for a topic that's extremely important for the world today. So that's one example of being able to take biosensor research and use that to iterate on something that can then be improved on and provided for consumers. That's just video testing, though. That's just one example. Usability is one example that we'll definitely get into with different methods. We've seen people do, you know, sensory testing as well. So if I have different, you know, food samples, for example, and I want to see which ones people react most favorably to and which ones might turn people off, we can use biosensory research to assess that. I think sort of the common theme in all of these different methods is that the most effective ones have a distinct research question and distinct variables. And usually we are comparing one thing to another. So we're looking at. If we're looking at a website, we're comparing prototype A to prototype B. If we're looking at packages or videos, usually it's an AB test. But there are other ways in which we can explore, do more qualitative exploration of where people are looking. Say in a shelf setting, for example, in a supermarket, being able to go up to a shelf and say, okay, where do people tend to look most on average? Then there's no sort of method of comparing there. But it's an easy way to get a lot of information about a population's behavior in a particular setting and then again, use that information to manipulate certain factors, like where a certain product might be placed on the shelf or how you position it next to promotional collateral or even other competitors, for example. Those are just a couple of things that we can, we can discuss.

Yeah, well, and I know from when we did the work for my site, we had the option and started out with just some version of, you know, free scrolling, right? Just sort of to see where people go, what they click on. There's a little bit of. If you give somebody, you know, two or three minutes to just sort of poke around and you can see are they clicking on, Is everybody going to one tab? Are people getting stuck somewhere? Do they actually watch a video that exists or does it just sort of sit there? Are you getting the brow furrow? What are they doing? Just to have an idea. So we have that knowledge, but we also are testing something very specific. But, you know, before we get in, we just want to see is there fact finding on whatever their task is going to be. Just kind of let them do that and learn from what happens versus having any real hypothesis about that kind of free scroll. We just want to see what happens.

Yeah, yeah, absolutely. For the study that we did, that first one, we kind of had a little bit of both. So we had that sort of qualitative first blush of when people are presented with your website, where do they naturally gravitate to? Are there certain things that draw attention more than others? Are there certain things that get ignored? What pages do people tend to navigate to? And then what do they focus on afterwards? And then after that initial. Because that has to do with sort of the design and the aesthetics of it, right? What's most eye catching. But for that second phase of the study, we adopted more of a task based sort of approach. And we were comparing, you know, two different prototypes, right, the new website versus the old website. And we had specific tasks to see do people actually behave the way that we would like them to in terms of getting oriented to products that best fit them or directing them to the information that you want to showcase the most. In that case, having a very specific question and having, you know, comparative metrics across two different prototypes that made it a lot simpler. And especially when you're looking at websites because, you know, only in certain circumstances are you really just perusing for the sake of perusing. More often than not, if you're going for a website, you're looking for something specific. So especially in usability studies, you have to keep that in mind that people are going there for a certain purpose and you have to incorporate that task into the kinds of design that you're looking at, you know, unless it's like social media where people are just passively scrolling through things. People are interested in the brainy business and they're coming there to learn something and to learn about you and the products that you have. So it's important to ask questions that are ecologically valid and that apply to, you know, how people are actually going to be using the product you're evaluating.

Yeah. And so we'll go ahead and jump in, we'll talk a little bit more. I think we've, we've alluded to at least some of the tests and what we were doing, but for the background story and, and you know, going into this, I was doing a redesign of the website and knew that it was going to be happening in conjunction with my upcoming book. And so we had talked about, you know, it would be fun to integrate. I really liked the idea of being able to do a study with eye motions in the process that could be included in the book and talked about at least a little bit. And then we could have, you know, the new website that, you know, you can go see it after the fact, you know that it exists when the book comes out. And so we did that research. So we had a first phase. So we did two different tests.

Yep.

The first one being, like you said, it was mock ups of what we thought a new site would look like. So I had had my site at that point for a couple of years. And the way that I, when I started the brainy business and switched everything over and started the podcast, I didn't know I was going to be doing courses. I didn't know I was going to end up teaching at Texas A and M. I didn't know. You know, I hoped a book might come out, but didn't know. And my consulting changed over time and so it kind of became what I lovingly referred to as Frankenstein, like adding so many extra, like we'll just add a tab, we'll do this. Like, I just need to be. It's good enough for now. But I knew that this redesign was going to be coming sooner than later. And, you know, book launch was a perfect time to change all that out. So in that process we had, you know, looking at the old site like you said, and then some mock ups of a new site. And those were done more like an in lab test. And I would love for you to talk about the type of testing that was done, but wanting to see, you know, are people finding the courses that we're encouraging them to go to. And we Change things up for the second test. But what can you say? What can you share about on kind of on your side of what the. The study was built on. Like, for me was like, I want. I have some theories about what I think is going to make a new site better and so incorporating the behavioral science stuff that I knew into an updated design. So I had my designer come up with a mock up and then we tested it against the existing site to see if our guesses were on point and if anything was lacking, that we could adjust it before we actually built a whole new site, like you said. And then you get to the other side and say, well, that's disappointing, it doesn't work. So we could figure it out before we got too far down the line.

Yeah, absolutely. So this was a ton of fun to do. So as, as you said, the first thing that we had people do was just a free one minute navigation. Because this is in part aesthetics and design in addition to the actual functionality that we wanted the site to have. So we wanted to see, you know, were there elements on each homepage that people tend, you know, attended to more than others or are there certain elements that got ignored or were people just scattered all over the place in their attention? If we're just looking at the 1 minute free navigation, both sites did pretty well. I mean, at least one minute is a good chunk of time. So people were able to attend to the pictures, the buttons, the different features involved. A lot of people navigated away from the homepage and we actually a lot of them clicked on the about page. So that indicated that your credentials and your background were really important to new visitors to your site. And a lot of time was spent reading the text based on the eye tracking. So people were really interested in getting to know you, Melina, as a researcher, your background and how you built this business and sort of the kind of services you're trying to offer. Now where the difference begins to happen is when we start introducing the task components. And there were two different scenarios that we explored, each being addressed to a separate Persona. So, Melina, with your business, you were interested in two different demographics. The first being sort of the more entrepreneurial types that might be interested in specific courses or, you know, advancing their learning on things. So learning about pricing or learning about mindset and just digging a little bit more to like the personal training side of things. But you also had this Persona for, for corporate members, so people that might be interested for more specific consulting and information on more custom projects. Is that correct? Did you want to Add anything more to that?

Yeah, no, definitely. I, over time really evolved to have of these two areas. And like you said, knowing that while the courses, you know, anyone can take a course, but in general for the like small business entrepreneurs, there's that diy, you know, jump in and do something. And for corporate projects, that's more of a, you know, a larger investment time with me, things like that and making it so it was easy for people to see where they should go and what products or services from the brainy business apply to them versus again in, in Frankenstein. It all just sort of was a big blob, I think, of stuff. And it wasn't easy to know, oh, I go here. So we were trying to solve that problem in the redesign.

Yeah, absolutely. What I really liked was that for each sort of Persona type that you had, you didn't make it too confusing. You had two Personas and you had one product that you really wanted to cater to each Persona. So for the entrepreneurial small business types, you wanted to direct them most of the courses because that's where you thought the most value was for them. Whereas for the more commercial or corporate enterprise types, you wanted to send them to custom consulting and to book a session with you on your calendar. Right. So these were the two tasks with the two scenarios that we then had for the study. And people had exposure to both tasks in a randomized order. So each task was preceded with an instruction about sort of the Persona that you were. So imagine that you are a new entrepreneur trying to start a business and you're really interested in the brainy business. For learning something about how you can conduct your business, go to a product that you're interested in, click on it, put in your information, and when you're finished with the task, close the browser. Similarly, for the corporate task, it said, you are working for a large marketing department in a conglomerate. You are interested in the brainy business. For help with a particular project, go to the page, click on the product that best applies to you, put in your information, and then close the browser when you're done. So what we did was we looked at eye tracking, first of all to to see where are people attending in the layout and where do they actually look at the right things that direct them to the right product, or are there things that could be distracting them, that could be dividing their attention? And then we also looked at facial expression analysis as an indicator of frustration or negative valence with a particular design. So if someone is having a lot of trouble with a website, you'll See a lot of brow furrow, which indicates sort of they're not having a good time, they're frustrated or confused or trying to read things. So we often look at that in usability. It's just an indicator of negative valence or confusion. And then we also looked at skin conductance is just a sense of general engagement with the page and with the experience. So what you did in the new design was that you had two very clear buttons that said entrepreneur versus corporate client. And that really made the difference.

Right.

If you actually look at the old page, there was a big blob of attention data where the offerings are listed, but there's a lot of confusion, and people often click the wrong ones trying to actually complete the task. But on the new page, you have this really clear delineation. Are you a commercial enterprise or are you an entrepreneur? And we can see for each Persona that our participants looked at, the amount of time it took for them to actually navigate to the button that they should be clicking on and then following through to subsequent courses.

Yeah, well. And one of the things I do with clients, and I think it's so important in testing, which is why one of the best things about running a test like this is you're really forced to narrow down to this. Like, if someone can only do one thing, I only want them to do one thing. If you are this type of person, what is that? What's the number one thing that's most important? Like we said for a corporate client, I need to have a conversation with them. I need them to go send me an email or book a discovery call so we can talk for a half an hour and see if we're a fit for each other. And again, for the entrepreneurial thing, I want to make it really easy for someone that's interested in a course to go get a course. And I actually, in a lot of cases, you know, I don't want to have to talk to every single person that wants a course that's not a good use of their time or mine and should just be able to go pick it out and get it yourself. And that's why it's there. And, you know, over time, then is the book. You know, it's like, I want people to be able to go find and get the book if that's a fit for them. And so when you go through this, you have to narrow down that if there's only one thing. So that is a really good task that you have to do. And then you can see, you know, is it working? Or not? Are people getting stuck somewhere or not? But I think it's also really important to note that the task in the way that they were created, like if someone's doing this themselves, I've seen far too much where someone will say, we're really curious if you can find the pricing course. Can you tell me, are you able to find the pricing course which is like, yeah, sure, okay, I found the pricing course, but that doesn't indicate whether someone is actually interested in it. And if they're coming from the perspective of a potential customer that might want that, and you say, hey, you're interested in something that might solve your problem as a, a small business owner, go look and see what exists. Because it's not like everybody who comes to your site is going to know exactly what you offer and exactly why it's going to matter and what it's called, and then go find it. And so to get into that mindset and asking a more vague question of this is sort of who you are, where do you end up can help you find if people would get stuck somewhere so you can fix something versus if you sort of like inadvertently prime them by telling them them exactly what you want them to do, which I think a lot of people could get into if they're not running a more formal test like this?

Yeah, yeah. There's two great points that I want to emphasize. And one, that whole sentiment of you just want to test one thing, I agree with that so strongly. In fact, a lot of the advice I give to clients is, you know, after having done research as a scientist for several years, you know, I often tell people the best research is boring. You're looking at one thing and one thing only. You're making one manipulation, maybe two if you want to be ambitious, but you don't want to go too crazy with it and have too many different variables at once because you're just not going to get the granularity of data that you're looking for and you're not going to get really strong insights out of it. The best research looks at one or two things, really isolates those, avoids any other sort of distracting elements or variables that could confound things and gives you a clear answer. Is this the right way that you're going yes or no? Do you want to address another question? Well, that's going to be another study. That's what entire research programs are based off of, is just this continuous questioning. And it's easy for people to fall into the trap of, oh, I can just do one Study and I get so much data out of it and I can get so many insights out of it. But often trying to rush it before you're really ready for something like that, it really robs you of a lot of the insights you can really get out of it. So I wholeheartedly agree with that. The aspect about the tasks and priming, I think that was a really good idea because we actually saw some really interesting things if we look at the entrepreneurial Persona. So I believe the actual hypothetical was you are a small business owner, you're an entrepreneur, and you're really interested in studying pricing. So go on the brainy business site, try to find a product that works best for you, and so on and so forth. A lot of people immediately oriented to the big button that said, are you an entrepreneur? Yes or no. And then once you actually get onto the course page afterwards, then it's really easy. You have two other options that are smack dab in the middle of the page with high contrast, sharp shapes like big rectangular buttons. You have the pricing course and you have the mindset course. And people attended really strongly to each of those buttons, but attended more to the pricing course, obviously. And then everyone subsequently followed through and clicked on it. You had a very high success rate with the new website design compared to the old website. Actually, yeah, I'm looking at the stats right now. The old website, you had about a 12% success rate of actually clicking on the pricing course, which is where you wanted to take them for that Persona. And the success rate went up to 70% for the new website design. So having those really salient design elements, those big buttons, here are your choices that made it really easy for people. And you see that in the eye tracking data and you see that in the subsequent behavioral metrics. So that turned out great. The second task with the corporate Persona got really interesting. So we actually had. So that first choice we had entrepreneurial or corporate, which Persona do you want? That was very clear. Everyone clicked on the right one. But once they actually got to the corporate landing page, people started getting confused. And we actually saw a very distributed pattern of eye tracking data across all the different elements of the page. There were a lot of figures, text boxes, there was a video, and people appeared to have some difficulty trying to figure out, okay, am I supposed to try and find a product here or what am I trying to do? So you can see that in the eye tracking data. That page especially had a lot of brow furrow. Actually, if you were to look at each of the new website versus the old website. The new website actually had more brow furrow, which is now what we expected, considering it rated higher on ease of use and usability and like, ability to design. All of the brow furrow happened on this one page, on the corporate landing page. And that's because people were exhibiting confusion over, okay, I'm this Persona, I want to get more information, what exactly am I trying to, what should I go for? And it just wasn't as clear to them. And we also saw this reflected in the behavioral metrics. So for the corporate task, we actually saw people did worse on the new website compared to the old one just because the design elements just weren't as clear. So that led to some really interesting directives that we could take for this corporate page. This is the one weakness in the new design that we can really develop some new ways of trying to adapt to really highlight the consulting aspect, highlight booking an appointment with you so we can do a better job of funneling that demographic into an action that we want them to take. Right.

And I think it's important too. So one is the reminder. We ran a test to see what would work or not. There were some elements when I was working with the designer. So, you know, there were a couple ways that they were designed. And some of the things was like, I don't like how this page looks, but we're running the test, so let's see what happens. Right. But you know, some of the calls to action we had, there were potentially too many on the page with different buttons. But testing them to see, you know, is the picture drawing them in here. And like you said it, we were able to find that there was, you know, too, it was giving too much information about this. Instead of just being able to say, you found the brainy business, you're excited to be here, have a conversation with Melina, like, really clear what your next step was. And so being able to find that, you know, potentially whether it's like paradox of choice or just cognitive load being there's like too much here and you don't know as clearly of like pricing course, boop. Like, that's what I want. But making that cleaner when coming up with the final design on the website. And so we only tested these two paths essentially, but I was able to use insights from those two tests really to be able to see when we did the full redesign and all these other pages, you know, those big buttons with large contrast really worked and making it very clear, are you A or B, what are you doing? You know, we were able to integrate that into other areas of the site as well. And in some cases where we had too much information, you know, how do we cut that back to make it very clear of, you know, contact here, learn more there, you can kind of self discover. And so as we, I think could talk about this first test all day, there was so much interest there. But we then did a second test.

We did.

So I worked with my team. After we got these insights, we found some things that worked, some things that didn't, some was confirming, some was surprising. And that's what the best, best tests are. Right? We're learning things on both sides. And then we go back, we made a new design for the site and then when it went live, we had two different versions of one page that was the apply it, you know, page that has to do with the book. And I mentioned in the book. So again, like before the site was ready, I said that this was going to be a test that people were going to do. And like all of this that we got to do before the book came out. And then it's like after that it goes live. It was kind of fun. But we had the page go live, selling the book. And then we were testing two different versions of the same page and really only changed one thing, but wanted to see how social proof would impact in Testimonials in the two different areas. So I had like 50 people that were nice enough to do very lovely endorsements for my book in it coming out. And so when we were in pre sale, then we were doing this test and so some of them that would come back, you know, it's like a paragraph of lovely things that are said. And I've told clients for years, you don't have to use the whole testimonial. You know, these little blips of the kind of movie trailer, like amazing, wonderful, the best business book to come out in years. Like these little things you can grab that can get buried in a paragraph, you can call those out, are going to be much more impactful than doing the whole thing. And so we tested both versions and what did we find?

Yeah, this one was super fun too. So we had this one page and this was more like we didn't have a specific task for people. We just said, hey, we're going to show you a page promoting Alina's new book. Go through the page, see if there's anything that's interesting. When you're done looking at the page, just close the browser. And what we wanted to see was just general search behavior. So a bit of a departure from, like an actual task. We weren't asking people to buy the book or anything, but just have a look at the page and see what sticks out. And as you said, sort of the difference between these two prototypes was in how we had framed the testimonials. So on one page, which I called the long condition, we had the entire paragraphs of the testimonials. So big blocks of text, because we figure, hey, more good testimonials are good. Right? And then on the other iteration of the page, we had taken just the highlights of those testimonials. Testimonials. So smaller bubbles, fewer text, but still getting the point across. And we wanted to see were there differences in visual attention across either of the websites. And then did that also translate to facial expression analysis content as indicated by brow furrow against suggesting frustration, negativity, and so on. So I was looking at the heat maps, and there is a huge, stark difference in how visual attention accumulated between the two pages. Even though the content's really similar, like the design elements are all the same, but except for this one thing that we changed. And with the short testimonials, you had this really nice diverse pattern of visual attention across the page. You know, a lot of it was collected up top, but a lot of people scrolled down through the entire thing, paused on some of the text, looked at some of the images. Really nice, you know, dispersal of visual attention throughout. So people got to see a little bit of everything, which is really nice. If you look at the long testimonial, all the visual attention is aggregated up in the very top of the page. There is very little actually going through, through the rest of the page. It's almost as if people scrolled down the page, looked at all the text and went, hey, yeah, no, thank you, and then went back up to the top of the page. It's almost. It almost looks like that's what happened. And you see this very characteristic sort of F shape of visual attention that you often see in webpages, especially long ones. People will start at the very top of the page and you'll see a lot of aggregation from the left to the right, on the top and on the left side going down. And then it slowly fades over time as people start skipping past things. They're like, oh, it's too much stuff. And they just start skipping through things and not paying as much attention through the rest of the page. So you really see that difference between the two. If we actually look at, you know, draw some areas of interest for the testimonial blocks for each thing. So when you're looking at eye tracking data, you can obviously look at the heat maps, which are very pretty, very visually striking, and they give you a sense of where visual attention is allocated on a page. But what you can also do is draw areas of interest around sections that are interesting to you. So in this case, looking at the testimonials, since that was our main dependent variable here. But you know, if you're looking at usability, you can look at particular features on a page, you can look at ad place placement on a page, there's many things you can use AOIs for. But we drew areas of interest around the testimonials themselves and looked at the aggregation of eye tracking behavior there and we found that in terms of hit time. So how quickly it took people to find the testimonials, people found the longer ones first. I mean, they're huge, they're eye catching. So it's natural that attention will naturally be drawn to them quicker than other elements on the page. But that's about where the good things end. For the long version of the site, all other measures I looked at. So the number of times people gazed within the testimonial area, people looked a lot more often at the short testimonials compared to the long. The amount of time that was spent within the testimonials was nearly double for the short testimonials compared to the long. The number of revisits, the number of times someone looked in an area and then away at another element and then back at the testimonials again, that was also higher for the short condition. So across most of the different measures of eye tracking, it looked like less is more. In this case, just taking the highlights from the testimonials goes a long way in being able to direct attention to that area. And people will spend more time looking there, probably because it's not nearly so overwhelming. People get turned off from large bricks of text, right then we saw similar measures with the facial expression data. If we just look at the amount of brow furrow people showed between each prototype, there was almost five times as much brow furrow in the page with the long testimonials compared to the short. Not only did people spend less time looking at them, their experience in general just wasn't as good. They're like, what's with all this text? This is confusing. I don't like it. I'm going to spend all this time looking at the top instead, because I don't want to look at anything else. There were very clear differences. And I think as someone, if I were consulting for you in this case, having done the study, I would say based off of the information we've collected, I would go with the shorter testimonials because people will tend to focus more on those. They have an easier time with your site and they're going to have a better experience overall.

Yeah. And being able to gather more of that social proof and see, wow, a lot of really amazing people said really fantastic things about this. You're able to absorb all the, the goodness of those and then make your choice if you want to buy versus like, you know, that's a lot to say. And what did they say? What's that again? And you just kind of get exhausted and are ready to bail pretty quickly. So I love again, in this case proved what I expected and have known, you know, for a long time. But being able to see, like you said, five times more brow furrow, less time than that's spent overall learning about these important points and the things people are saying about the book, those highlights we pulled, we want people to be able to then be primed with and excited. It builds up some anticipation of the book that they're going to buy and receive and get excited about it. So I loved having that information to be able to go off of. And obviously a big thanks to imotions for doing this, this multi part test. It was a lot of fun. I know we're gonna, I will definitely be including pictures and things on that apply it page so everybody can go see those there and there's other information about the test and so we'll be able to showcase so you can, you know, see some of these now that we're officially releasing. I know a lot of people. So the second one we did through webcams. So I talked about it on LinkedIn and with students versus as being, you know, having to go into a lab for the test. So some of you listeners potentially were a part of that test and now you know what we were looking at because we obviously didn't tell anybody.

Yeah, that was, that was a ton of fun. And I think, you know, that sort of leads to an interesting point of sort of how the tech has developed since then because, you know, as, you know, in the two years we've known each other, a lot's changed in the world and, and while primarily we do studies on a desktop platform, we bring people into a lab, we put all the equipment on them, it gets very involved. Well, all of that changed pretty drastically once the pandemic hit. And suddenly people weren't collecting data anymore because you can't bring people into a physical lab space. It just wasn't possible. Between the pandemic, there was also this general movement in the research landscape towards collecting data from more naturalistic environments. So not working within the controlled confines of a lab, but you know, can we record data from people in the comfort of their own homes? Can we record data as people are moving around in an actual space? So looking more at wearables and long term recordings and all that sort of thing. So that's sort of where our online data collection platform evolved. So we pivoted pretty heavily in the pandemic looking at remote options. And now we have an entirely different product that allows you to record facial expression analysis like you would with a regular webcam, but then also webcam based eye tracking data, which now removes the need from having to actually invest in eye tracking hardware for one. But what the module allows you to do is design a study like a simple AB test for video or for images or website testing. And once you create that study, you can actually distribute it online so you don't have to bring people into a lab. You're actually recording people from the comfort of their own home in a naturalistic setting where people are relaxed. What you get is research unrestricted by the bounds of geography. We're able to send this URL out to students at Texas A and M to again, post it on LinkedIn and get your listeners to come in and try the study. We got a wide demographic of people, which is a lot easier to get than having to, to line people up outside of a lab space, bring them in, deal with the logistics of all of that. So we got some really interesting results and the medium is completely different. But it's interesting how the technology is evolving and there's now more possibilities than ever for how we can collect this data.

Yeah, absolutely. As we wrap up, we've talked about a couple of labs that exist, including that human behavior lab at Tech Texas A and M. And for people that are looking to do tests, you know, we, and you know, Jessica, you and I talked about this. Going in is, you know, if somebody wants to conduct a test similar to the one that I did or a whole host of other things. You gave the example from University of South Florida, lots of cool stuff at A and M, those types of tests. If you are a company, you're looking to run a test and want some support, reach out to me, melina, @the brainybusiness.com and we could talk about doing something through the Human Behavior Lab, which is very exciting. If you are looking to build a lab to be your own new version of, you know, whether you're in consulting or an academic space, then what should they do?

Yeah, you can reach out to marketingemotions.com and we'll be happy to learn a little bit more about the kind of work you want to do and see where we can fit in and collaborate. So building labs is our bread and butter and we want to see you succeed. So if this is something that you're interested in pursuing, let us know. We'll be all too happy to help.

Awesome. Well, thank you again, Jessica for joining me on the show. It was fun to see you, as always, and exciting to finally share this research we've been working on for such a long time.

I know it's been so long in the making, so this is great.

So what got your brain buzzing as you learned about this website experiment and all the amazing experiments that you can do today? For me, I love a good experiment. It's one thing to say I bet this will work, or this would be compelling to me. But as you know from listening to this show, testing is always so key when it comes to understanding and predicting human behavior. Context is a a big piece of the puzzle, and what works in one scenario doesn't always work in the next. I was pretty sure that shorter testimonials would work better, but you never really know until you test. And you don't have to test everything, of course you could, sure. But that isn't a really good strategy for any business. It would be completely inefficient and ineffective. As I talk about and what your customer wants and can't tell you, you my three tips for doing your own experiments are to keep it small, to be thoughtful, and to test often. You need to know what's the most important thing to get right and test that. Be thoughtful about what's important. Don't test 800 different things at once, but do little things after you've taken the time to be thoughtful about the approach and the work and what you want someone to do. Do know what you're testing and why before you design and run the test. And if you liked hearing about this experiment with imotions today, you're gonna love the conversation in this upcoming Friday's episode with Joe Evershed and Dr. Gareth Harvey talking about the test we did together for the COVID of my third book, the Truth About Pricing, which is launching in just a couple weeks on January 9th 9th, 2024 and it is already on pre sale wherever you buy your books. So please do pick one up and check out the COVID in advance of that episode. We also talk a lot about really interesting research and tests that Gareth has done with customers in supermarkets and grocery stores. I think you're really going to like that conversation too. It's a really fun and interesting episode. As we close out the show today, don't forget about the show notes. They have links to related past episodes, books, a timestamp summary of this episode and more. It's all waiting for you in the app you're listening to and at the brainy business.com 352 and just like that, episode 352 on my experiment with imotions on the website launch page for what your customer wants and can't tell you is is done. Join me Friday for a brand new episode with Joe Evershed and Dr. Gareth Harvey to discuss a brand new experiment for the COVID of the Truth About Pricing. 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.

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.