Top Hacks To Boost Your eCommerce Website’s Conversion Rates

Kurt Elster is the Co-founder and CEO of Ethercycle, an agency with a proven track record of creating web-optimization strategies for top brands like Jay Leno’s Garage and HOONIGAN. As a senior e-commerce consultant and Shopify expert, Kurt helps merchants boost website sales. Kurt is also the host of The Unofficial Shopify Podcast, which explores tales of e-commerce entrepreneurship with a focus on Shopify. The show has over two million plays and features industry experts like Shopify’s President, Harley Finkelstein.

Here’s a glimpse of what you’ll learn:

Kurt Elster shares Ethercycle’s backstory
Kurt’s journey from a WordPress developer to an e-commerce podcast host
What makes a quality website?
Kurt addresses the validity of web design best practices
The importance of split testing in e-commerce
Are personalization and segmentation still relevant?
Case studies of e-commerce sites with controversial designs
Kurt’s top tip for improved mental health as an entrepreneur: identify your cognitive biases

In this episode…

In today's online shopping landscape, a user-friendly website is an integral factor of an e-commerce business' chances of success. An appealing, well-optimized, and functional website attracts more visitors – and, ultimately, more sales. How can you improve your site’s user experience, navigation, and interactivity to boost traffic?

eCommerce expert Kurt Elster emphasizes the importance of an intuitive website design. Removing any roadblocks from your prospects’ paths through split testing, web optimization, and other advanced marketing tactics can maximize conversion rates.

In this episode of the Up Arrow Podcast, William Harris welcomes Kurt Elster, Founder of Ethercycle and host of The Unofficial Shopify Podcast, to explore trends in e-commerce website design. Kurt shares his journey from a WordPress developer to an e-commerce consultant and how he started Ethercycle. He also identifies common design mistakes and provides tips for website optimization.

Resources Mentioned in this episode

Sponsor for this episode...

This episode is brought to you by Elumynt. Elumynt is a performance-driven e-commerce marketing agency focused on finding the best opportunities for you to grow and scale your business.

Our paid search, social, and programmatic services have proven to increase traffic and ROAS, allowing you to make more money efficiently.

To learn more, visit www.elumynt.com

Episode Transcript

Intro 0:03

Welcome to the Up Arrow Podcast with William Harris, featuring top business leaders sharing strategies and resources to get to the next level. Now, let's get started with the Show.

William Harris 0:15

Hey everybody, it's William Harris here. I'm the founder and CEO of Elumynt and the host of this podcast where I feature experts in the DTC space sharing strategies on how to scale your business and achieve your goals. And you are in for an epic treat today because I have the one the only Kurt Elster here. That nasty guy still needs to teach me how to use my road caster Pro a little bit more. Kurt is one of the most highly regarded independent consultants in his industry. He's a senior e-commerce consultant who helps Shopify merchants like Jay Leno's Garage uncover hidden profits in their websites, through his e-commerce agency Ethercycle, with over 2 million downloads. Now Kurt is best known for hosting the Unofficial Shopify Podcast, which I've actually had the pleasure of being on. In fact, my episode about “The Facebook ROAS Death Spiral.” I don't know if you realize this as well. Kurt, it’s currently the second most watched YouTube video on your podcast, second Harley Finkelstein, the freaking President of Shopify. Tough to beat. It's truly an honor to have you here.

Kurt Elster 1:16

Thank you so much.

William Harris 1:19

How's that for an introduction? Do you remember how we introduced how we started off our episode on your show?

Kurt Elster 1:25

Yes, it was the only you know, I don't have a ton were like put the full video up on YouTube. Like when I do that means I was really invested. And so for years, it was like You busted in through the door. Yeah. Which we like three cameras set up. Oh my gosh, what am I like running a sitcom? And then there, there was like, an entire premise of the cake is a lie.

William Harris 1:50

Yeah, the cake. Cake is lie when it's fake, right? Um, so yeah, we staged the whole thing it was it was fun, though. So before we get into the official good stuff here, I do want to make sure that I announce our sponsors Elumynt. Elumynt is an e-commerce agency. We focus on driving profit for e-commerce businesses; we've actually helped 13 customers get acquired, the largest ones sold for 800 million. And Adweek recently ranked us as the 12th fastest-growing agency in the world. So that's pretty cool. right onto Thanks, Ben. Onto the good stuff here. I want to get into the backstory of starting Ethercycle. How do you end up with Jay Leno's Garage? 2 million downloads over 2 million downloads, you know, first podcast like tell me about like, what led up to where you are today.

Kurt Elster 2:37

Alright, so I have been, I've been engaged in e-commerce my entire adult life, I started trying to locally find Furbies in the 90s when eBay was like a brand new thing. They didn't be used Pay Pal yet. It was like selling Furbies on eBay, retail arbitrage was a thing I started with when I was a teenager. I did that through college. And I ended up being that's like how I got my first proper post-college gig. I ended up being a channel manager for a drop shipper. And so I was like, I'm really into this e-commerce stuff, this is fabulous. This like speaks to, my interests. And so I then foolishly not knowing what I didn't know said, I'll build my own e-commerce platform. And it'll just be me and a friend who was on unemployment. And that's how we'll do it. And that's actually that's really, really hard. not successful. But that's where the name The name Ethercycle came from was it was our target market was bike shops. And it was a platform for bike shops. So the brand name made way more sense in that context. But it didn't work. And I needed to pay the rent, because I'd also jumped into an office because I thought that's what you needed to do. And so I started, we started selling WordPress sites to local businesses. I just like wrote a letter introducing myself slipped it under their doors. And we got Carlos away. And I said, you know, if it works with local businesses, what happens if I reach out to the because we're in the Chicago suburbs. So what happens if I reach out to these big ad agencies in Chicago, these marketing firms, and it turns out a few of them said, hey, we'd be happy to outsource web development work to you. So we're able to wrap it up that way. And then at the same time, I had a friend who owned a bike shop the the Chicago's finest recumbent bike shop, Ambling cycle, he goes, I hate my website is supposed to sell online, I just wanted to be easy. You know what to do tell would just help. And I said, I heard of this thing. This is like 2011. I said, I heard this thing called Shopify. I think their whole thing is they're easy. We'll just build a custom theme on that, again, like not knowing what I'm getting ensure only we were successful at it. And so they said, we got this thing called experts program, you should join our deal. And then it like, five a few years after that. I said, Man, I gotta get clients in here. I should start a podcast. I didn't really know what I was doing. Right? I had no idea. I just had a $30 microphone a really echoey room like just awful audio quality didn't stop me just kept Public. Shame. Years later, you know, 400 500 episodes, suddenly you got 2 million downloads of the President of Shopify is like taking you seriously is quite incredible. And no one can stop you. It's self publishing. It's great. And we get better at it over time, of course. And around that time is when we like, I put together like, wait a second, why are we doing anything other than Shopify here? So those two things went together. And then we just stayed the course. You know, you keep doing the same thing. You get better at it. And at some point, you're just last man standing.

William Harris 5:30

Yeah, you know what, though? I mean, I like, you know, there's a lot about what you just said was one of this is just that dedication to sticking through something, anything, for the most part almost inevitably leaves you in a better spot, because everybody else ends up getting out of it. At some point in time, everybody else doesn't have the tenacity to stick with it. And I think that's huge. You have a lot of talent, though, too. And so, you know, that can't be looked over. But I appreciate what you said about you went into it without knowing all the reasons why you shouldn't. And I feel like that's actually one of the biggest keys for success. And anything that somebody does is you can overanalyze all the reasons why you shouldn't do something. And I think that prevents a lot of people from actually just taking action and making something happen.

Kurt Elster 6:13

There's there's a lot of advantage in not knowing what you don't know, like just being ignorant and jumping into it. Because you don't realize what you're getting yourself into. Like today. I often wonder that for years, I've wondered, man, if I know if I knew what I knew now, what I what I talked myself out of it, then. And that's totally plausible.

William Harris 6:34

Yeah, absolutely. Oh, in fact, I think it's very highly likely that you would write like, I think that about a lot of the stuff that I've done that, more often than not, if I would have known what I know, today, I would have said, Don't do it. And sometimes I have to catch myself with that, where you know, somebody else wants to do something that maybe I've accomplished. And I look at him like you don't want to do it. Like there's all this this this like don't even bother like it's really not worth it. But but the reality is it is and you just have to like don't don't Jade that perspective on some of those things, too. And it's like let them stay naive a little bit. They need that naivety in that excitement at that stage. Because it is it is tough. There's a lot of hard things you're gonna encounter when you run a business.

Kurt Elster 7:12

Yeah, absolutely.

William Harris 7:14

So WordPress was an interesting start to this is just completely tangential for anything that's really worthwhile. But I actually sold some WordPress sites as well, a long time ago. I think you got into it before I did. But my first client was my pediatrician. And I could just remember going to my client and my my doctor just being like, Hey, I I feel like you need a website he had like music or whatever that he made. I won't say his name just cuz I don't know if he wants me to. But he had music that he made and everything in his own CD like kids music. And so I was like, I'd love to make a website for you. So that was my first client in the WordPress did

Kurt Elster 7:46

did a play music.

William Harris 7:48

I don't remember it. Probably. The soundtrack is probably, I mean, this was probably the MySpace days. So yeah, it probably had at least some kind of like small music player. The so one of the things we wanted to talk to you about. And I think the thing that you and I, when we first met you remember when we first met in person we talked, you know, online for a long time, the first time we met was actually in Minnesota

Kurt Elster 8:18

at work, and you say, hey, Your fly's down. And I really, really cool and subtle about it. And I was about to go give a talk. And so it was always since then I've been like, that's a good dude. And I'm very grateful. That was really cool about that, and saved me from talking to like, 40 people with my fly open.

William Harris 8:39

That's hilarious. I don't remember that part. I was gonna say just Yeah, I remember you getting up there that we were here. And I think it was Charlie Joachim over at Cecil, and Todd frosted there. And you did a talk basically, where you just tore down a bunch of people's websites, right? And you're just saying, Hey, here's a bunch of ways that we can make this better. And it was so helpful for people to just see on the fly the way that your brain thinks. And it leads towards just trends and best practices. And CRO in general, which I know that you've made a really big name for yourself on and what do you see right now as some of the trends that we need to be watching? Or how do you go about just approaching? What's a good website? Well,

Kurt Elster 9:21

there's a few things like I like the the squint test, if I can, like really just kind of like squint at it blur my eyes. Can I still work this website or even figure out what it's about? If yes, all right. That's a really good sign like that is my two second litmus test. That's cool. Okay. So a lot of that's about like, you know, a did you resist the urge to be subtle that's like such a trap, and has been so for like, since as long as design has been a thing. It's like subtle, sophisticated. So let's have 10 point, Bray fonts on a dark gray background. And then, you know, like, let's be, let's be really cagey about what, who we do and what we sell because then we'll look really sophisticated. In mysterious, please don't absolutely don't do that. If you do that I'm just like clicking the Back button and leaving and going back to scrolling through the endless gazing pool of, you know, Instagram or Tiktok, or whatever the heck. You I just, like, just right off the bat, is this thing easy to read? And is it dead obvious as to what's going on here? Because the other thing you hear from merchants, nobody reads? Nobody reads but that like you go bury all the info that the person needs and not make it obvious. Maybe there's a reason they're not reading.

William Harris 10:30

Yeah, that's fair. I feel like this reminds me of my early days in SAS. So before e-commerce, I was in SAS, and I remember at the time, there are two real big websites that I appreciated for just just validating our testing whether or not our site was good, there was the user is my mom, and the user is drunk. Do you remember these websites? I don't know if there was

Kurt Elster 10:51

a testing? Yeah. Richard Littauer, I believe was the man's name.

William Harris 10:55

That sounds familiar. Yeah,

Kurt Elster 10:57

I had him on my podcast is like, what a minute. Guests. That's how I remember the day when I was I thought that was like a big gift at the time, I was very excited. He really, you would pay him, he would get drunk with me and test a series of websites. And he was like, I don't know how sustainable this is, you know, buddy?

William Harris 11:17

No, I know. But it was. The reality is, it's the kind of the idea of the squint test that you're talking about where it's like, you know, alright, let's, let's see if you can see what this is, you know, saying when you're squinting, if you can tell what's going on while you're drunk. Or if your mom can tell what's going on on this website. Those are good signs that you've got a website that's clear enough.

Kurt Elster 11:34

The user is a crazy lazy drunk, like assume that they're a narcissist. If your website's just talking about you, alright, that's not going to interest me as a narcissist. So you got to, like, make it the whole website should be you focused language. And if like, I'm bleary eyed, drunk, can I still navigate this thing? Right? Sure. If it can't, there can't be road bumps, I'm lazy. It can't make me work to try to get to the outcome that you're trying to push on me, which is buy something, right? And so if you can pass those three sniff tests, all right, you're in really good shape. Now, when I think about all of this, I'm thinking about it entirely in the context of mobile, I think as web professionals, we all spend a lot of time on desktop. But our users don't, especially in e- commerce, you know, if you told me your website was 75%, mobile, that would be low, you know, 80 to 90 is much more typical.

William Harris 12:29

Yeah, well, and I think that that's one of the things that's interesting is, I remember when this shifted significantly, you know, over the last few years to where mobile was truly mobile-first, and Google started evaluating even the effectiveness of your website accordingly. But you're right as as professionals, we use our desktop extensively, and not just one monitor. But we've got several monitors, I've got, you know, a couple extra monitors right here. They're big. And so I can see everything that I want and in full detail. And that's not the reality that most people are experiencing on your website. And so, to your point, it's like you should be able to see what this is very clearly on mobile, don't bury it make it easy to read. What else? I mean, like these are your best practices, but like, how do you feel about best practices in general? Yeah, I,

Kurt Elster 13:13

I like best practices, but I also take it with a grain of salt, you know? Sure. Essentially, when we talk about best practices, it's really like, well, what is Bae Martin say we should do pay Mark usability Institute is a fascist resource. I love them. But their approach is they it'll be like, Oh, hey, we looked at the 2000 websites, we did user testing across 200 of them. So it very much becomes like, well, this is we're pushing toward the average. Right? And so there are always going to be exceptions and edge cases. And like perfectly legitimate reasons to not do things exactly the way Mark told us, or situations where it just isn't practical to engage in those best practices. I think the risk is that we've gotten into is a lot of people, you could just say, well, baymard says it's best practice. And then like, that's the end of the discussion there. Because there's no more argument this is what you should be looking at that is like, Okay, well, that that means that's a good starting point. That means like, that's a safe place to begin. And then maybe we test that later, like what you trust, but verify, I think is the way we should approach best practices. But at the same time, if your website if you don't have a huge quantity of traffic, it can be difficult to split test things. And certainly you can split test one thing at a time, if you've got if you don't have enough traffic, and so it's not always practical to verify these things. But you know, I think the one thing everybody could do is like the screen recordings, oh my gosh, my clarity is free, install that and really could view screen recordings of people actually using the site. And the way I do it now is like Alright, show me only mobile show me only rage click Next, I got effect for whoever I could do this better. Yeah, sure, I want the person who's on their website to just start clicking like 25 times, and then it'll flag that. And you can watch like a recording of the person losing their mind, just like mashing on this color button that doesn't work on one device, but no one ever noticed. Because that person's not going to complain, then they're not going to buy, they're just going to quit the site and never gonna hear from them again. But Microsoft clarity saw that it was spying on them, it watched it. So you watch, I'll sit for like 30 minutes an hour, and I'll watch those one and a half x speed, you know, move through quickly. And I'll have, it'll skip an activity. So if they stop, it just moves past it. And, you know, doing that you could watch a few dozen of these. And very quickly, you'll recognize where like, oh, there's an issue here like I can on this, this huge apparel site. And we discovered, it took too long for the color variant switcher to work. And so people would rage click on that, and it took there was no breadcrumbs that would let you go for product collection. And so that was frustrating for folks. And you couldn't actually zoom the image on mobile, but it didn't stop them from trying in a very angry fashion. So like, if I could, if you can uncover those issues and then address them, the whole thing performs better. But unless you're looking for them, I promise you're never going to hear about it, you're never going to notice.

William Harris 16:25

Now that's that's really good. I gotta go back here real quick to for those of you who are only listening and not watching the video, perfectly lip synced that random complaint that was pretty impressive. But so you know, the idea here of like you said with with Bay Martins, some of these best practices, it's the idea of like, Don't reinvent the wheel. Circle is probably a really good safe shape to start with, if you're making a wheel, you can play around with other shapes, then after that, right, you know, test some stuff, but it's like, hey, circles probably been proven, go and start there. There are some best practices that are worth starting at, and then recording things and seeing where people are getting blocked and work on that. And I think that's just brilliant advice. Because there are edge cases, like you said, everybody does use things a little bit differently. And so if you take just the average of every single person, you know, that's fine. But what about your specific users, your users, you might, you know, tend towards a very specific, you know, group or demo or whatever. And they react a little bit differently. And let's just even say for extreme examples, it's like how my mom would use a website versus how my daughter would use a website, they likely have very different ways that they're going to approach us and you can start with a general best practice. But then, if I'm over-indexing towards one or the other, I need to look at well, what is my, my particular demo? How do they interactive? How are they where are they getting stumped to check and work on that,

Kurt Elster 17:52

100%? Yeah, the? Well, and I think, I think the other risk with best practices is to go, where's the line? Right? If they say, if you start listening to well, this is every best practice. So does that become a to do list, right? Should I implement every single thing that suggested? Absolutely not, right? And less is more when you can eliminate elements that people don't interact with? And that's where heat maps are very useful, where you can see like, okay, no one is actually using this, this element that get rid of it. If I can get rid of it. It simplifies the site that makes it easier to use, it makes it more likely to figure out what's the next step in every page should have like one primary thing trying to get me to do and only one primary thing? Yeah, yes. Yeah. When I say primary, I mean, you can't have like 10 goals, right? Like you get to that one. And the moment you add a secondary goal, all you've done is dilute the first one, right? It's like with sliders. You're like a carousel slider with five images in it. Oh, my website's slow and my conversion rates are down. Low. Okay, well, maybe we shouldn't be loading like 10 one megabyte JPEGs in a slider when nine out of 10 people only ever see that first one. But sliders, settled debates, because now as a business owner, I don't have to make a decision. Right? Yeah, well, my content in there. I hit the slider, it says my point.

William Harris 19:12

Well, there's something about not making a decision that makes us feel good. Which is really funny. But you're right. A lot of people were just like, hey, let's just appease everybody. And that almost kind of leads into let's just say, one-to-one personalization. This was such a hot-button topic I remember at IRC years ago, and so you're leading towards okay, no best practices and we can optimize towards your individual users. How much do you see this as still being a thing where people are still talking about, like one-to-one personalization? Is that kind of like a fad that's been done or there are still things people are doing to you know, let's just say like micro segment people.

Kurt Elster 19:49

You cut out there one more time?

William Harris 19:51

No, I apologize. So like, is this a fad for still doing like one to one personalization, or is there still some validity to segmenting you know on a very granular level,

Kurt Elster 20:02

if you could figure out how to do it, right, absolutely go for it. There's also that you get into the issue of diminished returns. But the idea is sound, you know, get the right message to the right person at the right time, and then get rid of all the other stuff. And so if you can figure out a way, where you have an audience signal that you can like, correctly, identify something about that person, and then show them the right thing. Absolutely. And especially if it's like new versus returning visitors. Okay, on your site, maybe you have a recently viewed Items section is one of I love those recently viewed items sections. However, if you're a new visitor, the site, it is meaningless, it's useless. So for new visitors, maybe I just hide that section throughout the site. And for a Returning Visitor now, like they're gonna have that recently viewed Items section, it becomes useful. That's a simple example. But that's more personalization than most sites do. So I would look at it that way. It's like, can I use? That's probably the the easiest, safest way to look at personalization is, can I use it to just hide the irrelevant items from people?

William Harris 21:11

Yeah. So are you okay? If we look at a couple of examples of this and just kind of talk about what we like or dislike and stuff. The one that I think you and I talked about last week was the Glossier, Shopify just redid the Glossier website. And I'm gonna pull up here I can share my screen. Let's see. Make sure we select the right window

and we should be good. Now, let me know if you can see that by showing the right one here.

Kurt Elster 21:53

So can you

William Harris 21:54

see the glossary website? This is the last year website here.

Kurt Elster 21:57

I see a tab about leather pants. Why is my butt so sweaty? When I wear leather pants? See, that's like a question ChatGPT talk to you about it.

William Harris 22:06

Yeah, that's embarrassing here. Let's actually get the right one up here. Then there we go. Here's the Glossier one.

Kurt Elster 22:11

Alright, so here's a Harley Effa Twitter says the brown glossy sites live and powered by Shopify. And he has like glossy ads Supreme, the streetwear brand moved to Shopify around the same time, this is an exciting win for them. However, this got

William Harris 22:25

this got hit though. Like a lot of people hated this website at the same time. A lot of people loved it. A lot of people hated it. Why?

Kurt Elster 22:32

It's isn't that strange? I think if you like a site that's that polarizing. Hey, congrats on all the free publicity. I'm sure that their their conversion rate just tanked when they launched the site, because a whole bunch of unqualified people went to the site. Like, I never make up for Glossier, you're probably not buying makeup from Glossier. I bought a bunch of powder from Sephora, like five years ago, and I'm still using it occasionally on camera, like,

William Harris 23:01

oh, yeah, I need to do that, too. I've got a little thing going on here.

Kurt Elster 23:06

I'm not doing it here. But I've just, you know, I'm an oily oily man. And so now a 40-year-old with acne, I need my powder. So but like, it's kind of interesting, like everybody within the e-commerce Twitter community. And of course, you know, Twitter thrives on outrageous opinions, oh, had opinions on this glossier site, because there's high standards, this this huge brand, that there are lines like around the block when they open a retail store. And for the president of Shopify to say, hey, let's welcome this brand to our platform with the site that we developed with them, then suddenly, there's very high expectations that like, this is the best of the best. When the reality is, it's perfectly alright, to have, you know, to not follow best practices blindly and strictly in favor of, you know, let's have some personality, let's have some branding. Or occasionally, let's just try stuff, let's throw it against the wall and be a typical, because we want to see what happens, you know, we're not paying by the pixel, this, the page doesn't expire or go bad. I can make changes at any time. That is, like one of the wonderful parts about web development is it's very malleable as a meeting. And so it's it's okay to play around and test. But with glossier, the criticism, and it looks like they might have addressed this a little bit like

William Harris 24:34

yeah, I'm wondering, it's like, this looks like this has changed a little bit since the original one went live.

Kurt Elster 24:39

The original one that main menu, was it was plain text. It still is. And it was like this thin font. It was gray on white. It was small. It was it was shockingly difficult to read. And it wasn't for like lack of space. It wasn't like they had too many elements. Right, it feels like a site that's very mobile-first. Right? We're looking at it on desktop.

William Harris 25:07

Yeah. And that's part of the problem, right? Like, I mean, like we just called out, sometimes we're evaluating things on desktop. And I'm willing to bet a lot of other people were evaluating this on desktop and not looking at the mobile version of this.

Kurt Elster 25:18

On mobile, it's a little weird if I'm honest.

William Harris 25:20

Sure. I mean, by now button this far to the left, like just randomly over here, like that is different, right? Like glossier, right? Like that's different. It's just, you can't say that it's best practice, but it might work very well for them.

Kurt Elster 25:33

I think sometimes you get that works to like they're the human brain craves novelty. And so if I can present you with something unexpected, new and different, I have your attention. The problem is with those novelty is, it's a trick, and novelty wears off. And so at some point, it just becomes like, oh, yeah, that's what it looks like. And it's kind of annoying.

William Harris 25:55

Yeah, I think one of the things that you said that I appreciate the most though, out of all of this, the biggest takeaway is be aware of the metrics that you're looking at when you analyze whether this is or isn't working, because you call it out perfectly when you launch a new website, the conversion rate for them was down likely significantly, and it had nothing to do necessarily, with whether or not the site was good or not. And even if you were looking at, you know, you know, spying on them, like you said, with, what was it Microsoft, what clarity, or Microsoft clarity, you still would maybe see like a lot of erratic behavior, because it's people like you and me who are not going to buy it that are just browsing everything. And so it can create a lot of let's just say like false realities that you might try to make decisions on that data. And it's just, it's not the right data that you need. It's not the users who are supposed to be coming to your website buying. And I think that can lead people towards making bad decisions.

Kurt Elster 26:51

What's funny is a like a new site launch with an accompanying PR campaign. A hallmark of its success is if traffic goes up and conversion rate goes down, because it means we drove a whole bunch of new new eyeballs to the site. Sure, I've seen it before. And then maybe our remarketing kicks in whether that, you know, and there's so many ways to remarket. Now with customer identity resolution, you know, I can, they don't have to put in their name, email or any info. I can identify who they are. And I can automatically mail them a postcard and send them an email. These things work and they're, they're legit and legal. They get my ad retargeting going, and then, you know, maybe 10 days later, they make a purchase as a result, it Okay, so there's a longer tail on that. Bia, we often see that with like a new site launch, especially if we we publicize it and promote it as like, oh, traffic went up, conversion went down. Oh, no, the site launch wasn't successful, wasn't it, though? Yeah, it could be conversion rate could be misleading, because it's so dependent on its denominator? Yeah, if I change, if I change the quality of traffic in any way, that number swings wildly.

William Harris 28:01

Yeah, yeah. And I think that's true. For a lot of metrics, the metric itself can be very misleading. returning customer rate is one that I absolutely hate as a metric. Because I don't think it tells me anything worthwhile. For example, if all of a sudden you doubled your ad spend, and you're driving a ton of new traffic to the website, your returning customer rate, as far as new customers divided by or you know, new customers compared to returning customers is wildly different. That's not necessarily a bad thing. It doesn't mean that you all of a sudden have less returning customers, it just means that like weighted compared to the amount of new traffic that you have coming in, it's different. And so you might think that you have a problem with returning customers when you don't you just have actually done exactly what you need to do, which increased your top-of-the-funnel that's not a bad thing. But the metric changed.

Kurt Elster 28:48

Instead, they have no I don't know where I was going with it. That's all right. So I'm gonna fire up a product detail page.

William Harris 28:57

What Oh, sure. Because I was saying otherwise, we could jump over to the crux. Want to I want to talk about that one. Let's see.

Kurt Elster 29:05

Yes, I think yeah, look, you get to say now I feel like they proved the criticism on the fonts by bumping up the size or the font weight one of the two because even in a screenshare like suddenly it's easy to read. Yeah. But you've clicked through to a product page it says the Lip Color Trio. It was not obvious to me where that Add to Cart button was going to be the add to cart buttons below the fold along with the price that's all right, that definitely it's a typical it's not best practice. And our products are hidden the left Yeah, that the Back button is gray. Oh light.

William Harris 29:41

Does that mean it's exactly what you said? It's like it's not even black text. I don't think I think it's like dark gray on light gray on white.

Kurt Elster 29:50

Yeah, so like, you know, if I'm, if I'm saying let's conversion rate optimize this, my answer would be take that Add to Cart button. Let's make it way more obvious that it's button buttons should look like buttons. And then above the fold, I want that like with that little product description up there. Hmm,

William Harris 30:07

yeah, yep, right up here. Now this make simple.

Kurt Elster 30:11

Same deal. Maybe this looks different and better on mobile.

William Harris 30:14

Probably not. Too bad. Okay, so yeah, but you know, the button is such a silly thing. I can remember there being outreach, I wrote an article on Optimizely years ago, maybe 2014, about, you know, even buttons and some of the things that we did. And we did see a 40% increase in conversions just by making some edits to the button, where it was at the color it was. And I can remember a big thing that people talked about was, you know, at least at this time, the hot thing was button colors need to be yellow. And I think that was because maybe there was something that Neil Patel put out there about like button colors being yellow that had the highest conversion rate. It's just to make sure that I clarify like what I think and you can absolutely refute this. This is I'm not a CRO expert, by any means. We focus on the ad side of things, but it's not about yellow, it's not about what color this should be. What I like to see is contrast, I like to see that there's at least something that makes this stand out to the point where it's like, I My eyes are at least drawn to realize, oh, there's a button there and that I can click this.

Kurt Elster 31:17

I remember someone telling me, like buttons can't be red, because red means stop and add. And they were convinced of this. They're like so yes, to be green. All right. Sure. Like what if Green is the color of birth? About that green is the color Barker subconsciously? I don't want to click your birth button. Like if that's the direction we're going here. You know, like, how far do you want to take it? The fortunately, that's a fairly easy thing to test. It's funny is I for the number of split tests I've run I have never one split tested button color. That one just seems so inconsequential.

William Harris 31:53

Sure, maybe well, yeah. Well, like I said, I don't think that the color itself matters. But I do think that the color compared to the background matters. That's where I'd say like gray on white light gray on white wrist. Exactly. It's just contrast. It just does. Does it stand out enough to where you can see that there is a button there.

Kurt Elster 32:11

Well, and you also have to consider like, you know, people with different visions, so people are colorblind. Like marcher, famously, is colorblind. That's why Facebook uses a lot of blue is what he could see.

William Harris 32:21

Yeah, yeah, that's a really good call. I'm not colorblind. And so I love color coding everything, you know, Asana tasks, and everything. And there are people on my team that are not there that are colorblind. And so I have to think through that and say, Wait a minute, is this going to work for that person as well? Um, I've got the crocs added pulled up too, because along the lines of let's just say best practices, this is not on the on the website. But I it was so interesting, and it just fit with the theme of let's just say, conversions and best practices and what you expect out of a brand. I loved this. And have you seen this one? Did you see this go around?

Kurt Elster 32:56

I've seen it because of you. Okay, and I love it. I love I call this style of ad shitposting lack of a better term, but it shitposting I love when a brand does it because you know, they thought about it, they thought hard about it, and they approved it. And they it went through multiple layers to get to the point and ad spin before you ever saw it. And they went like this is it. They have selectively designed a like meme type shitposting style ad. And that's why it gets your attention because you don't expect it from a brand. But it is what you expect to see in your feed from friends potentially. Sure. That's your like, you know, this meme style. But it's really, too as a brand to do a meme style add stakes go up like you don't want to see you do it wrong. And the my Gen Z kids will say it's cringe because you're trying too hard. Like, God forbid, it looks like you put effort into anything if you're

William Harris 33:55

absolutely, yeah. Including, including how you dress or anything, right. It's like it's funny, like some of the things that like my dad will say that are in style. And it's like the literally the goal I feel like is to look as if you tried not at all, but you really did try but you but it look like you did not try at all.

Kurt Elster 34:12

Like remember grunge in the early 90s. That's what I had. That's what I grew up with. And we look back on that same deal. I'm sure our parents were like, What in the heck is wrong with those kids?

William Harris 34:22

It's coming back. It's coming back actually a little bit. I think I just saw some people starting to wear gene codes or something. And I was like, Oh man, like we're going way back into like, the hell it was cool when I was in like Junior High in high school.

Kurt Elster 34:33

Mad gencos Do you want some jeans that are just really really?

William Harris 34:39

I did show my daughter those and she was like, no, please don't let those come back and stop.

Kurt Elster 34:45

The butt. So if you're, if you're listening, that's okay. Yeah.

William Harris 34:48

Yeah. Well, and if you're listening, I just realized it was like if you're listening, you have no clue what this ad looks like. So I want to explain it for you real quick here. Crocs did an ad and it has just like a side profile of the shoe. And then there are every possible bad font that exists on Windows 95. There with like gradients on the fonts and things like that. And it says, Did you know Crocs has sneakers? Very sneaky, very soft. Mom will approve. I mean, this is the epitome of like you, you took this into Canva you didn't even let Canvas AI do it, which they have some really nice AI lately, but you took it into Canva You didn't let the AI do it. You just took the basic cheapest fonts that you could find and threw some colors on it and just said, Yep, that looks good. At

Kurt Elster 35:37

church flub, that's the style. That is should be on a bulletin board at your local community church.

William Harris 35:45

That is absolutely accurate. I loved it. It got so much attention in. I think that it it caught my attention. I even liked the words that they use, though. Whereas like very sneaky, very soft. Mom will approve. It just it feels fell to me, doesn't it? It does. And it felt like it brought back nostalgia to me. And so like if their target audience is me. It absolutely did a great job because I remember playing around with every one of these fonts. I don't know if I can tell you what these fonts names ARB, I can remember playing with these and all of those colors. When I was on adding like that drop shadow behind the Crocs. They're

Kurt Elster 36:19

playing with like brighter bund print shop Deluxe.

William Harris 36:26

That is wow. That's exactly right. That's what it feels like. And so I'm like, Oh, this is cool. This is like my childhood.

Kurt Elster 36:31

I call basically a graphic designer. Yeah.

William Harris 36:35

So I want to transition a little bit because there's something else that I wanted to talk to you about that we talked about before, which was empowering mental health, there was something you brought up about cognitive biases. Perception is reality. Reality isn't reality.

Kurt Elster 36:52

Yes. So I think for, for mental, both for improved mental health, as a human, not just a business owner, but I think with entrepreneurship, the stakes are higher. And so it's easy to get more stressed. And it's like, you know, the buck stops with you as an entrepreneur. And so you tend to be harder on yourself, you tend to bring in, internalize a lot more work stress, and that can be quite detrimental. And so I think one of the most powerful things I've done is often interrogating my own decision-making. And like, if you're talking to ChatGPT, it's kind of boring if you tell it, let's think step by step. But you should really do the same thing yourself. It's like, okay, if there's a thing that's bothering me, let's back up here. What led me to think about it this way. And there's a lot of cognitive biases that humans engage in, like, the thought that we are, by nature rational, just as a true like, we're really just, you know, clever mammals try not to get killed. The reptilian brain set up. Yeah, yeah. Like your reptile brain is back there trying to keep you forget by a snake or something. And so if you start thinking through a lot of like, cognitive biases, and you learn them, you start to recognize them and realize, oh, no, a lot of what I thought might have been good decisions, or things that were bothering me simply aren't the case. And my, one of my favorites, that I see a lot of people fall victim to is recency bias, where you got like, I had an electrician over, did some work, and then half the outlets in my basement, quit working. And I was like, clearly this man broke my outlets that obviously, based on the timeframe, that's what happened here. And the reality was, you know, my house was built in 1992. And the circuit breaker when he did the proper thing of turn them off individually before the main breaker and then turn the main breaker back on turn them on individually, it's just this one breaker had never been flipped. It was so old, that just flipping it on and off, it was enough to break the thing. That was all that happened. You know. And so that recency bias, though, I was like, oh, you know, this guy probably does. It wasn't the case at all. And we see that all the time with websites, like a bug crops up, and it's like, whatever the last change I made, oh, well, clearly, that must be it, or you go up, you know, changes button color hurt, alright. Sure. And the next day, versions are down, it must be because we changed the button color. Not necessarily at all, those things just happen to have occurred within the tape, same timeframe. Not that they're necessarily related. Like that's one of my, my favorite examples. And of course, like confirmation bias will just tend to seek out information that agrees with us. You know, I don't want to go Google things that don't agree with me who wants to be like getting over that, you know, being able to say, hey, maybe I'm wrong here. Maybe I like completely blew this. Or maybe I need to say sorry to apologize. Like, being able to identify those cognitive biases just makes it so much easier to go Oh, I got that wrong. But here we go. We can fix that for the better I think that's where it starts is like really just there's a Wikipedia articles you go through, here's all cognitive biases, and you start recognizing very quickly, like, the ones that you'll you'll make, it's also fun to recognize it and other people, especially in like, how they're dealing with you. Or like in, you'll see especially in like online arguments, you know, like the straw man arguments, you know whether they're gonna tax up they are set up something else that either they dismantle and go see, you're like, but Okay, great now that we've torn that straw man down, let's go back to our original argument here. Yeah. But life just becomes much easier when you engage in these things. And like it, maybe you learn this stuff in college, I didn't, I did go back and learn it through Wikipedia as an adult, that's the recipe,

William Harris 40:45

I wish that it was almost a required course in high school or something to go through, let's just say logical arguments, right? And all the different types of like use of straw man being one of them. But like there's, you know, a lot of different ways in which we are lied to on a daily basis, sometimes intentionally, sometimes unintentionally, by let's just even say, by ourselves, that we lie to ourselves, we are lied to, by others, by the media, by the government, by whoever it is that we're not against. Yeah,

Kurt Elster 41:21

like, that's just the nature of human rhetoric, human words.

William Harris 41:25

Exactly. Yeah. And I feel like this shows up in data all the time. And this is exactly I think, to the point that you you bring up where, you know, we'll see this in, let's say, the ad side where somebody can say, well, our ads aren't performing as well, because the CPM is up. Well, that might sound like a good reason for why the ads aren't performing well. But what's interesting is if you go chart the CPM against your, you know, revenue, CPM is positively correlated with revenue. And every time your CPM goes up, your revenue typically goes up. So well, that's interesting, that didn't chart out the way that you had hoped it did. And so there's a there's a bias that we sometimes have in all of these things. And we, like you said, we seek out the data that tends to tell us what we want to believe. And I think that causes a lot of a lot of problems in our analysis. And I actually wrote a couple articles that I think would be good to talk about with this, there's two big ones that I remember putting together. One is about Bayes Theorem, and how that shows up in advertising. And so I'll link to that in here as well. Um, the other is about Simpsons Paradox in advertising. And both of these kind of come to the idea where you can have these presuppositions about what you expected to see in the data. And the data will actually confirm what you wanted to see, unless you look at the underlying variables. And so an example of this is, we'll see this in not just advertising, but let's say let's say like medical results, and so they do a study or something on whether or not a medication was effective. And they could say that it is effective, maybe in general, if they look at just all of the trials, but if you look at like all of the different sub-segments of every single person, it actually wasn't effective for every single one of those groups compared to the placebo. But in the aggregate, for some reason, the way that the aggregate amount of stuff worked out because it maybe was more heavily weighted towards one group than another group. The aggregate numbers, the averages actually say that it was overall but each symbol single group it wasn't and it's a wild thing. So if you haven't seen Simpsons Paradox, please go read this article and see how this is playing into your, your your e-commerce store your ads, and to bring it back the ads. The example there would be maybe, maybe there's an ad that you're saying ad A and B, and maybe ad B says that it's beating at A when you look at it just across every campaign, but when you break it down, you look at well which one's doing better for prospecting versus retargeting. Maybe it's A that's outperforming in both of those groups. But because it was weighted more towards the retargeting, that's why ad B in aggregate ends up looking better across the whole; it's a very interesting phenomenon.

Kurt Elster 44:02

There's that phenomenon, issues like that occur in your daily life, whether you realize it or not constantly, and oftentimes, you just you're doing it to yourself, and you have no idea. And until so yeah, until you're aware of it. You can't change it. And I think that's just, it's just a lifelong way of thinking is learning to question your assumptions about decision making.

William Harris 44:27

Totally. Kurt, I got one more thing that I want to do with you here. So I like to have a little fun. This is an ad lib script. That so ad libs, not Mad Libs, because it's about advertising, but I'm nice to do that on my I need to do that on the road caster Pro. But so so I'm not going to tell you what it's about yet. But I just need you to give me a few words. We're gonna deal with this and then we're gonna act it out. So this is very, like Jimmy Fallon style where we're you're going to be personalized I believe person to or vice versa, whatever we want to be here. So I need a name. Any name, fun name, make up a name. It could be your name. It could be Harley Finkelstein. I need a name.

Kurt Elster 45:10

I'm gonna go for a name. I'm gonna go with the Keebler elf. Does that count as a proper job? salutely

William Harris 45:16

That's great. People are off. Okay, and I need a sound that an animal makes.

Kurt Elster 45:24

Well, Moo is immediate.

William Harris 45:27

No, that's good. Verb you would use in sports. So it's like a verb that you'd use in sport. Tamron. Nice. You're good at a verb that you would use in a workshop or a machine shop. Drill. Okay. a plural noun as a thing. So like a plural thing. Bottles.

Kurt Elster 45:56

Is there a theme? Is the models work?

William Harris 45:59

No, no, no. Like, I mean, there's a theme to this, but I don't want you to know that they might specifically want you to just pick random things. Right? A feeling or an emotion that ends in ING sweating. That's good. That's interesting. Okay, idiom like it's raining cats and dogs kill two birds with one stone.

Kurt Elster 46:25

The customer's always right.

William Harris 46:27

Nice. Almost an animal. Rabbit. Nice. Shout out in Orndorff, the rabbit Matthew C. To write an animal body part not for a rabbit though. different animal.

Kurt Elster 46:49

And animal body part.

William Harris 46:51

An animal body part. So I mean, it could be from a rabbit's foot tail tail. Sure. I just meant it didn't have to be from a rabbit. Let's see. A verb you would use at work like email. It's like I need to email this work. Verb Diem. Nice. And then same animal from before here. And a human body part. Know

Kurt Elster 47:22

What are like specifically human body parts?

William Harris 47:24

ni ng ankle. I mean, yeah, it doesn't have to be specifically human body parts, I guess. But it's like I didn't want you to say like trunk. And it's like, well, that's an elephant.

Kurt Elster 47:32

Alright, I'm going with kind of goes here.

William Harris 47:34

Yeah, absolutely. It's actually really funny when I see that. And then a business phrase or call to action.

Kurt Elster 47:48

Add to Cart.

William Harris 47:51

It's good. Okay, and then last too. I just need any verb, just just a verb. I don't have a theme for it. Chat. Okay, and then. And then the last one is just one more verb. Ad verb, share, I ran out of categories. What was it? Share? Share? Actually, if that fits really well, okay, I'm gonna share my screen. One more time for you here because we're going to act this out. Pull that up. You should be able to see it. We're gonna go back to the top. Do you want to be person one and person two?

Kurt Elster 48:35

Yeah, I didn't want to get into so I guess I'll be person one.

William Harris 48:40

Okay, you pick the right one. Okay, so we are writing an email to a client about why their ROI is down. Okay, so we're gonna set the scene here, your client just sent you this email. Hey, Keebler elf. I just got an email from a Facebook ads rep. And they said that everything you're doing is the wrong thing. They were 100% convinced that the reason I rose is down 3% week over week is because you're targeting new customers instead of just letting Facebook do whatever the heck it wants to do. And repeatedly target the same repeat customers that it knows will buy already. Your person one.

Kurt Elster 49:14

Excuse me. Are you on? That's really dialed up.

William Harris 49:18

You're good here. Let me blow this up a little bit here to make it a little easier.

Kurt Elster 49:22

Boy, that's a tough email. How are you going to respond? Man who gives

William Harris 49:26

a flying Moo? What the Facebook rep said maybe I'll just ignore the email and go tackle myself.

Kurt Elster 49:34

I don't know. I don't think our boss would like that. Why don't you read to me what you have and I can help you drill it a little bit. I tend to be really great with bottles

William Harris 49:44

naturally. Great. Here's what I have so far. Hey, Steve, I can understand how sweating must feel to get an email from your boat. Hang on one second, Siri decided that she was going to jump into that conversation. Thanks Siri. I'm sure you didn't get that. Okay. I can understand how sweating it must feel to get an email from your Facebook rep like this. I just want to assure you that everything is the customer's always right. The Facebook rep is wrong here, it's much more likely that the ROI is down because of a rabbit in your account. The rabbit will often cause ROIs to drop because they use their tail to DM which the Facebook policy is clearly against. I'm going to dig in tonight and remove the rabbit with my own two hairs if I have to add to cart, William Harris.

Kurt Elster 50:33

Well, I can tell you've really chat this through. Sounds good to me. Go ahead and share it.

William Harris 50:41

There we go. We nailed that. I appreciate you entertaining me with your acting skills. Um,

Kurt Elster 50:48

yeah, no, I've heard in the whole class that for you.

William Harris 50:53

Okay, I really appreciate the half-assed thing and like, do you have you seen what is his his MO? I think have you seen his MO? He's? He talks about half-ass Okay, so he talks about he's he's like from Croatia or something. And he's really funny explaining how he doesn't understand like a lot of like things that Americans say and like half-assed doesn't make sense to him, too. It's like, you gotta go watch. But he's like, half as this is a bad thing. You know, full as this is a good thing. Like I don't understand what's going on. But I always have as I understood it to be a let's just say a misunderstanding of the word haphazardly. You did it haphazardly and haphazardly just became half has, and then somebody was like, oh, did you say half-assed and it turned into half-assed, which makes absolutely no sense at all. You can't walk around with one budget, at least not that I'm aware of very easily

Kurt Elster 51:45

English has a lot of weird idioms. But you know, in linguistics, what you've described is called a corruption where it's like someone Miss here's the thing, but then it starts getting repeated. And then now you've that becomes accepted as to like, that's the phrase. So yeah, it could be could be language corruption. Like patina is actually should be pronounced Patna. It's not French. It's Latin, but we say petite. So it's a corruption.

William Harris 52:10

Interesting. There are a lot of corruptions in life. For those who want to be uncorrupted, and follow you and stay in touch and learn from your magical mind. What is the best way for people to reach out and stay in touch with you?

Kurt Elster 52:23

Oh my gosh, Google, me Google Kurt Elster. kurtelster.com is my site. It's got links to everything. But mainly it has my newsletter. And every Tuesday, I send out a nice recap with some what I learned that week, called the awesome recon newsletter. Nice, nice is

William Harris 52:38

a fantastic newsletter. You should subscribe if you're in e-comm at all and you are not subscribed to this. You are missing out. Definitely subscribe to that. Curt, really excited to have you on here today, sharing with us your wisdom.

Kurt Elster 52:51

Thank you so much for having me. Hold on the crowd. Calm down, guys.

William Harris 53:00

Thank you, everybody else for joining us. Have a great rest of your day.

Outro 53:05

Thanks for listening to the Up Arrow Podcast with William Harris. We'll see you again next time and be sure to click Subscribe to get future episodes.

‍