The Unofficial Shopify Podcast: Entrepreneur Tales

The Boneheaded Mistakes We've All Made Running Online Businesses

Episode Summary

with Steve Chou from My Wife Quit Her Job as co-host

Episode Notes

Steve Chou and his wife Jennifer built a seven-figure business... and made a lot of mistakes along the way. We all have!

This episode isn't about successes, it's about the stuff everybody (myself included) gets wrong when starting out.

MyWifeQuitHerJob's Steve Chou co-hosts this joint episode to discuss the boneheaded mistakes we've all made running online businesses.

Show Links


Never miss an episode

Help the show

What's Kurt up to?

Episode Transcription

Steve Chou: Kurt Elster, what is going on? It’s been a while since I’ve spoken to you.

Kurt Elster: It’s been a bit and I regret that. That’s my fault. I should really be following up with someone as fabulous as Steve Chou.

Steve Chou: Well, it’s funny. Last time we hung out was at the Klaviyo conference and we recorded one of these podcasts also and it was a lot of fun.

Kurt Elster: I was gonna say, as soon as you said, “The last time we hung out was at Klaviyo,” and the first thing that popped in my head was, “That was so much fun.” Like when I think about that conference, that’s really the memory I have, is recording with you, and Youderian, and Tony, in that cool but weird little recording booth setup they had. It was a ton of fun.

Steve Chou: And what I remember is you guys like busting on me the entire episode and yet I still published that one.

Kurt Elster: Sometimes you just have to speak truth to power.

Steve Chou: All right, so Kurt, today’s a really exciting episode. We’re gonna be talking about common mistakes that store owners make, and let’s start out with your number one pet peeve. What is that?

Kurt Elster: Obsessing over page speed and now, as of today, Core Web Vitals.

Steve Chou: Yes. So, actually, can you talk about Core Web Vitals first and what they are even, just in case the listeners don’t even have any idea what that is?

Kurt Elster: Okay, so originally Google said, “Hey, you gotta look at your page speed,” and they had this arbitrary… what I feel is a largely arbitrary metric that much to its detriment gives you a score, a grade. It grades your website and it’s based on what I think is fairly flawed data. And the implication was always if it doesn’t perform well, at some point you’re not gonna rank on Google if you can’t get a good score. And the reality is that is an overstatement of what is really happening, and so as of now, coming out, a Google update is around the corner, it may be about to publish, in which they’re going to take into consideration three metrics from the larger Google PageSpeed score called Core Web Vitals.

And probably the most important one there is first content paint, so really, it’s like how long does it take. Not necessarily like how long does the entire website take to load. How long until I can see it and use it is really what Core Web Vitals looks at to simplify it. Because you could be on a website, it starts loading, and you can start using it and scrolling before it’s finished. Stuff can load in the background. That’s fine. They don’t actually care about that part. It’s how long until I can use the darn thing.

And so, people naturally freaked out, and the reality is what’s really going on is if you look in like the fine print in the description of what Google’s saying they’re gonna do, it’s they’re saying, “Hey, if we have two sites that rank for the same placement, the one with the better Core Web Vital rating will take priority. We’re gonna use it as a tiebreaker.” Okay, so speed is important, but it is not the be all, end all, lose sleep over it thing that I think it’s being presented as and people are treating it, and really my issue is I’m so sick of the anxiety and the hair tearing out that’s occurring with merchants needlessly because of page speed.

That’s my problem.

Steve Chou: You know what the problem with it is, is that it gives you a letter grade from zero to 100, and agents everywhere are trying to get a frickin’ 100%, and it really can’t be done without jumping through a lot of hoops, and I think it was you… Was it your podcast or one post on Facebook where you actually looked at some of the top Shopify stores and all of them had PageSpeed scores under 20.

Kurt Elster: Yes! Yeah. Yes. Yeah. So, Shopify’s making an effort in trying to get people to have more performant websites, and as part of that, they said, “Hey, you can see in your store. Here’s a Page Speed grade.” And it’s a little better than like the regular Page Speed grade because it’s an average of like a homepage, a product page, a collection page, so a little more accurate, and then in my partner portal in Shopify, where I can see like, “These are the stores I have access to. These are my client stores.” I can rank, just sort them all and see what the scores are.

And what’s interesting is very few people get above 30, and the stores I have that get above 60, they’re stores that we built like five to 10 years ago and then really haven’t changed. They just worked and that was the end of it. And so, I thought that was interesting. So, I know part of the issue here is app JavaScript. We’ve done the research here and discovered largely it’s JavaScript, it’s apps, and it’s not deferring the load on that JavaScript that creates the problem. But in our seven, eight, a couple nine-figure stores, all of them are in the low double digits, like 12 is a pretty typical score.

And Shopify will tell you like, “Oh, you score the same as similar stores.” So, we know for the highest performers, the teens… So, like F-minus-minus if we’re applying a letter grade is the typical average score.

Steve Chou: You know what’s frustrating about this, is that since I teach eCommerce and teach SEO, like I felt like I had to do it for myself, so all my sites are in the 90s now across the board.

Kurt Elster: Hell yeah.

Steve Chou: But it took me three weeks to do that, and I had to jump through hoops and get into the code in order to do that, and I can’t expect a non-technical person to possibly do that, which is ridiculous. Google cannot expect you to do this stuff, right? Would you agree?

Kurt Elster: Yes. No, and I just don’t understand. I don’t understand the initiative. I don’t understand why this has become a priority all of a sudden. I would love to know internally at Google, what is going on here? Is it some altruistic desire to make the internet more accessible to slower devices?

Steve Chou: Never. It’s got to do with ads somehow.

Kurt Elster: Probably not. Yeah, like I wonder what the real objective is here.

Steve Chou: Well, here’s my take on that. You know, one of the Core Web Vitals is cumulative layout shift, which is basically whether your website shifts up and down when it’s loading, right?

Kurt Elster: That’s a weird… Who cares?

Steve Chou: Well, have you ever gone on one of those sites where you’re trying to click on the next button, but then the ads… all of a sudden it appears where the next button is, and you click on an ad by accident?

Kurt Elster: Yep.

Steve Chou: I think they’re trying to reduce that and click fraud. That’s their motivation.

Kurt Elster: Oh. Wow. I like that theory. Never would have occurred to me. That’s pretty brilliant. Like if you look at how they generate their revenue, that makes a lot of sense.

Steve Chou: Right. And also, this page speed, like if it loads faster, that’s more ad clicks, right?

Kurt Elster: Oh, man. You cracked the case.

Steve Chou: So, that’s what I really think this is for, and so they’re trying to automatically algorithmically take out those people who are just trying to scam the system for AdSense.

Kurt Elster: This is really good.

Steve Chou: It’s my take. That’s my take. All right.

Kurt Elster: I love it. So, I think the other interesting thing is Brian Dean from Backlinko published data yesterday where they looked at something crazy… It was like 200,000 sites. Something nuts. And he said there was very little correlation, if any, between UX metrics, which I think they measure… It sounded like they measured in page views, and Core Web Vitals scores. So, obviously like stuff that was way screwed up, it would correlate. But for the most part, it really did not have an impact on UX.

Steve Chou: I don’t think the rollout has fully happened yet, right? I think they delayed it until like next month, I believe. So, we’ll see what happens, but I suspect you’re right. It’s probably not gonna have as much of an effect. It’ll have a similar effect as to what PageSpeed did when it first got announced.

Kurt Elster: So, tell me sir, what… PageSpeed is clearly the thing that makes me crazy. That’s my big one, is every time someone brings up the ghost of PageSpeed I’m like, “Oh, here we go.” What drives Steve Chou nuts?

Steve Chou: Okay. You know what drives me crazy, is when someone comes up to me and says, “I tried Facebook Ads and Google Ads and it didn’t work.” And then I go over to their site to take a look and they’re driving paid traffic to a site that just isn’t ready. And we can expand upon this, because I know this is part of your pet peeves, too. But how many times has someone said that to you, you go over their site, and it’s just terrible?

Kurt Elster: Yeah, but then Facebook ads gets blamed. You can’t blame the traffic store because your site doesn’t convert.

Steve Chou: Right. And they spend all their time on the copy, and all this stuff, where they’re just driving it to something that has poor copy. It actually doesn’t even make sense to me.

Kurt Elster: Yeah. Often, the disconnect there is frustrating. Or really, like how often have I been on or have you been on… I like Instagram. I’m a millennial, I suppose, so I like Instagram. I go through my Instagram stories and then I’ll see an ad for something, and it looks interesting. I’ll swipe up on it and I just end up on like either they send me to the homepage in which case I’m just clicking out. Heaven forbid. Do not send anyone to your homepage. Or I just end up on like an entirely standard, unoptimized product detail page with like a one-line description. And I’m like, “What did you think was gonna happen here?” And of course, I just bounce.

Steve Chou: Well, you know what my pet peeve is? Since we’re talking about product descriptions, my biggest pet peeve is when I see a product that has zero likes, zero shares, zero everything, zero reviews, and they didn’t even take the time to hide the zero part. Because when I see zero, that indicates to me that no one’s buying anything and that this is-

Kurt Elster: Yeah. Zero social proof is what you just advertised.

Steve Chou: Exactly.

Kurt Elster: That’s negative social proof.

Steve Chou: It is. It actually causes me to leave because that means no one’s bought that product before, right?

Kurt Elster: It’s funny that product reviews apps don’t know to hide themselves, especially like the review stars widget, if there are zero stars. For most of these apps, that’s a pretty easy customization that we have to go implement. It’s not hard, but that should just be an option built into them. Because yeah, zero reviews is just like a red flag if I’m a brand-new visitor.

Steve Chou: Actually, you know, that’s really interesting. So, you actually have to muck with the code to fix that?

Kurt Elster: Yeah.

Steve Chou: Interesting. And then same goes with social proof, right? Like Facebook likes, and shares, and that sort of thing? You physically go in?

Kurt Elster: You know what? I almost always scrub those social share buttons because what are the chances that you actually get someone to share a product page to their Facebook feed? That’s a tough sell.

Steve Chou: I agree with you, actually. I had considered removing those altogether, because I know from my own site, very few people click on those. However, I will say that people do click on the Pinterest button, though, I guess since I’m in the wedding industry. What have you found?

Kurt Elster: Yeah. For some industries, the Pin It button, that’s a win. But you need to be in the right niche and wedding is perfect for that. But beyond that, like, “Oh, a product page! I’m gonna share this to my timeline!” No one’s doing that. It’s just not gonna happen.

Steve Chou: Interesting. So, do you take away those social buttons actually for your clients?

Kurt Elster: I do. Yeah. I always just… At this point, I just strip them out, and then if there’s pushback on it, I say, “Well, we can heatmap it or split test it.” I say, “But if I heatmap your best-selling product, I guarantee out of 2,000 visitors, you will get… Flip a coin. One or zero clicks.” And sure enough, that’s what happens every time. And it was so consistent that I don’t even think about it anymore. I just strip them out unless it’s a niche, like wedding, in which case just give me a Pin It button and that’s the end of it.

Steve Chou: You know what’s tricky about this, is like the default Facebook buttons always have the number next to it, right?

Kurt Elster: Right.

Steve Chou: And so, unless you know how to manipulate the code a little bit, it’s actually kind of hard. I had to code up my own button to do that. To remove the shares.

Kurt Elster: Oh, really?

Steve Chou: Yeah.

Kurt Elster: Yeah. Most of the time, yeah, it’s just like, “Facebook share,” and it doesn’t have the numbers. I mean, if people are actually sharing it, more social proof, that’s always gonna be my priority. If the question is, “Should we add more social proof?” The answer is yes. I don’t think you can do enough.

Steve Chou: Here’s like an obvious pet peeve, like it’s intuitive to me, but sometimes I’ll land on the site, and I’ll have no idea what the heck they’re selling or why I should buy from them at all.

Kurt Elster: Oh! The story, and the description, and the positioning. Those are the fundamentals. I swear to God, all of the fundamentals are in copywriting, and that’s including the social proof. Like you gotta nail those before you worry about design and page speed. I think that’s my other issue, is like sites with one-line descriptions and no reviews, and like we’re worried about our page speed score. You got bigger problems.

Steve Chou: Actually, you know, I’m actually curious. Since you work with a lot of clients, what is your hierarchy of priority when you land on a site, or what you work on?

Kurt Elster: Okay, so 100%, number one, it needs to be clear. The positioning or the tagline needs to be clear. Because you said like, “I land on a site, I got no idea what they’re doing, why I’m here.” That’s a big problem. And if you saw it, everybody else did too. So, when I land on a site, I need… Within seconds, I need to know what they sell, why, or at least you’ve been able to spark my curiosity. You know, like I land on and they’re like… That’s not a real site. But certainly, like a name like that, I gotta know more. So, sometimes you get lucky, and you can get away with not having more of a story initially.

But for the most part, you gotta have a tagline, a headline. You need something to orient to me as to why the heck I just got dropped in here and what I’m supposed to do next. And so, I really… The absolute cornerstone of everything is knowing and nailing and communicating that positioning statement. And that could be like-

Steve Chou: That needs to be above the fold.

Kurt Elster: Yes. And usually it’s like, “All right, I want your logo, and can we put a three-to-five-word tagline underneath it?” Oftentimes, that’s a great way to do it. Maybe you get lucky, and the brand name has what you sell in it is obvious, like Harney and Sons Fine Tea Company. I’ll give you a guest what they sell, right? So, that one, like you get lucky. But yeah, I want a tagline. I want a headline. I want something that’s gonna make it really clear to me, above the fold, what you sell.

All right, and then that makes it easy for me as a visitor, as a customer, to go, “All right, I want to know more,” or, “This isn’t for me.” One of the two, but either way, we’re separating the wheat from the chaff. And then from there, okay, tell me a story. I like stories. Tell me a story! I want that brand story. And it’s even like in a product description, you still want to be able to tell a story. On an about page, you’re telling a story. On the homepage, you’re telling a story. If you’re not storytelling throughout this journey, you’re leaving money on the table. That is the next fundamental and that really comes down to, “All right, all of this falls under copywriting.”

We’re not talking about design or development. This is just you gotta type on the keyboard and make the click-clack noise, right? Nobody wants to do it. It feels like homework, so it gets pushed to the bottom.

Steve Chou: Here’s the perfect example. So, our mutual friend, I think, Michael Jamin, he runs TwirlyGirl, and they sell kids’ dresses.

Kurt Elster: Oh, I love Michael Jamin.

Steve Chou: Right? Girls’ dresses, very commonplace, very saturated, but their story is so amazing. It was about how his wife was abused as a child and she just wants to express… Of course, I’m not doing a good job of explaining this, but she wanted to create these dresses to make girls really happy, and he has just really clever videos on his site that clearly express this amazing story. I’m not doing it justice. And that video makes almost all of their sales.

Kurt Elster: Oh, yeah.

Steve Chou: It’s on the about page, it’s on the front page. It’s an amazing story. Every time I watch it, I want to go buy a dress.

Kurt Elster: Well, there’s quite an unfair advantage there in that Michael Jamin is an actual, honest to goodness, Hollywood screenwriter who wrote for King of the Hill, and Beavis and Butthead… I mean, he’s just brilliant and hilarious, and he applied… He knew what Hollywood storytelling worked, like what that looked like, what that process was, and when it worked, and he turned around and just applied it to copywriting on this website. And the results are million-dollar videos.

Steve Chou: For most humans, though, you should work on your about us page a little bit. Tell the story. Just something about yourself that’s a little bit more personal. Because you don’t want to appear as like a big, faceless, big, large company, right? You want to play off the mom-and-pop aspect because people want to shop from mom-and-pop stores.

Kurt Elster: Yes. And especially now. I think in the last two years, we’ve really seen the rise of conscious consumerism, where people want to know who’s getting their dollars and why. Why should I care? I have… At this point, I now have what feels like infinite options for just about anything I want to buy, so being able to tell that story, that’s the competitive advantage against Walmart, against Amazon. They can’t compete with that. They’re not a person. Jeff Bezos, richest man in the world, going to… The guy feels like a supervillain at this point, right? You’re not. You’re a person. You could be their neighbor. If they can relate to you and relate to their story, your story, then you are going to get those dollars. People are gonna vote with their dollars and say I believe in what you’re doing. And if you keep your mouth shut and never share your story, you guarantee that that’s never gonna happen.

Steve Chou: So, a common misconception is that people don’t even look at your about us page, but I guarantee you, and this is for our store, the about us page is probably the second or third most trafficked page on the site. Because when you’re shopping at an unknown boutique, people actually want to know the people behind the store.

Kurt Elster: Yes. Yeah. Well, think about it. It’s like if somebody jumped out in a parking lot, “Hey, give me your credit card number! I’ve got these t-shirts! You want to buy them? Give you your credit card!” That’s insane, but that’s functionally what a lot of online direct-to-consumer stores do. And so, I need to know who it is who’s asking for my credit card number, and that’s why that about us page is so important. And if you run screen recordings on a site, it’ll probably be true for just about everybody’s, is you’ll see a fair number of people will add to cart and then go check the about page, and then you can see whether or not they’ve made a purchase decision on if they go back to the cart and proceed to checkout.

Steve Chou: Here’s what I actually like to do. So, usually when you ask your friends for an opinion on your site, they’re your friends, right? So, they’re gonna say, “Oh, it’s great. It’s great. It’s gonna do great.” But they’re probably lying to you because they don’t want to hurt your feelings, right? So, what I like to do is I’ll use a service like PickFu, it’s basically like a polling service. You can get 50 responses within like 15 minutes. And I did it with my site maybe a year ago and you’ll get real, unbiased comments, like, “Hey, I hate the popup. I hate this. I hate this. No, I would not shop there.” You just ask a simple question, like, “If you landed on this site for the first time, assuming you liked the products, would you buy from here?” Simple question like that and you’ll get a lot of eye-opening answers.

Kurt Elster: Oh, I love it. How do you spell PickFu?

Steve Chou: P-I-C-K-F-U.

Kurt Elster: F-U. All right.

Steve Chou: Maybe in the show notes, I actually have a 50% off coupon off of that, so if you want to try it, it’ll be the best $25 you’ll ever spend, I guarantee you.

Kurt Elster: Sweet. No, I know this works, because we’ve used… I’m not familiar with PickFu, but I’ve used Hotjar, which will let you do… They call them incoming surveys. Same concept.

Steve Chou: Oh, it’s the same thing. There’s a lot of companies that do similar things. Yeah.

Kurt Elster: Yeah. I don’t use it, but I know a lot of Shopify merchants like Lucky Orange. I think it can do it.

Steve Chou: Oh, really? Okay.

Kurt Elster: I’m not sure. You know what? I shouldn’t say that. I really have no idea.

Steve Chou: Well, so the students in my class, I have them do it before they submit like a… before they ask me to critique their site, basically, I ask them to do it. And it’s eye opening. It’s eye opening.

Kurt Elster: Well, so this brings me to one of my other pet peeves, not talking to your customers. Nobody talks to their customers. It’s like, “Well…” And oftentimes you’ll ask like, “Oh, why do your customers buy?” And they’ll have an answer. And then I’ll say, “How do you know?” Oh, we just know. What? You just know? That means it’s a best guess. And you might be right, but until you ask, it’s still just a best guess. And that’s what I love about talking to your customers, and one of the best ways to talk to your customers, man, pick up the phone. Call your customer. Use It’s the best app. Call your customers on the phone and talk to them. And be like, “Hey, why’d you buy? What’d you hope to get out of this? Tell me about the experience.”

Just talking to a few people is really enlightening. But at the same… Will have a similar effect to asking your friends, where they’re not gonna tell you to your face like, “It kind of sucked.” They might, but they’re less likely, whereas that instantaneous popup, you’re getting that immediate reaction, unfiltered, and man, people tell you the craziest stuff in those popups.

But it’s so… It’s incredibly helpful, especially when it’s like, “All right, I got a few weirdos, but eight people all kind of said the same thing was their initial impression and it wasn’t great.” Oh, all right, so now you know that’s a thing you need to address.

Steve Chou: You know what I do every time I release a new product? And this actually drives my wife crazy. I have an abandoned cart script on my site where if they abandon, like I get emailed saying, “Hey, this person with this phone number abandoned.” And I’ll call them up and I’ll say, “Hey, we just noticed that you tried to check out, but you didn’t complete the process. Was there anything wrong and is there anything that I can help you with? And more often than not, they’re actually willing to talk to you and tell you what’s wrong, and then usually what I’ll do is at the end I’ll give them a big coupon, or sometimes I’ll just give them the product for free after that. Just compensate them for their time.

It’s really hard to do this. It’s really hard to cold call. It’s definitely not cold calling, right? Because they were on your site, and they gave you-

Kurt Elster: It’s not cold calling. Okay-

Steve Chou: But it’s awkward.

Kurt Elster: Who wants to pick up the phone and call a stranger and be like, “You tried to buy from my website, and you bounced, and I want to talk to you about that.” The first several times you do that I’m sure was nerve wracking.

Steve Chou: It is. Actually, you know what? If you want to make it a little bit more comfortable, what you can do is you can lead to the offer. You can lead with the offer and say, “Hey, you know, we noticed you didn’t check out, and don’t worry, I’m gonna compensate your time with a big coupon, or give you the product for free, but I’d just like some honest feedback on why you didn’t make the purchase the first time.”

Kurt Elster: And do you represent yourself as like, “Hey, I run this site with my wife.” Do you make it personal?

Steve Chou: Yeah. I’ll say, “Hey, I’m the owner of this business. I run it with my wife. And we just launched this product and we just noticed that you tried to check out, but you didn’t finish it, and we’re just wondering if you’d give some feedback on the process.” Something very straightforward like that. And I would say seven times out of 10, someone will probably be willing to talk to you.

Kurt Elster: Interesting. Oh, I love this idea. And so, when you call them, what kind of objections do you typically hear? They’re like, “Well…” Because they’re gonna tell… You’re essentially saying like why didn’t you buy, and so that you can then bust that objection for other people in the future, and then you seal the deal with, “Hey, I’ll give you the thing for free.” And now you’ve got a customer for life. Or like, “Here’s 50% off. I’ll just give it to you at cost.”

Steve Chou: Right. So, I could tell you this, because maybe a couple years ago we launched like a new apron line, and sometimes you’ll get a question. You’ll get an answer like, “Hey, I was on my mobile phone, and it just wasn’t convenient to check out.” Which could be a problem in itself, right? Like maybe you need a better mobile payment system, so you don’t have to enter in all that stuff. If you’re not using PayPal one touch or-

Kurt Elster: I was gonna say, yeah, like you need… If you hear that objection and you don’t have an express payment option, you know what the solution is.

Steve Chou: Right. But so, in this particular case with the apron, what ended up happening was the woman was like, “Hey, you know…” So, we sell a mother-daughter apron set, just to kind of give you some background, and there’s sizing for the adult, and then for the kids, we just have an age range, like… I can’t remember what the age range is now. Two to four and six to 10 or something, two to six or six to 10. Anyway, the problem was that the person wasn’t clear what the exact sizing was for their child because they have a child who’s larger, percentile wise, and they weren’t sure that the apron was gonna fit. And we didn’t do a good job of conveying that.

And so, after talking with her, I was like, “Hey, so what information would you need? Because we actually have the measurements there.” We had the measurements in the product description, but what she would have liked to see was like if my daughter is five foot six, or something like that, like a height scale. And so, that was something that-

Kurt Elster: Yeah, they want like a table.

Steve Chou: Exactly. Yeah. Based on height, not necessarily age, because age is ambiguous.

Kurt Elster: Yeah. I don’t like age. All of my children are tall, but my daughter is in the 98th percentile for height at age four. So, people are always like, “She’s four?” Well, when sizing stuff is based on age, it’s really hard. I’ve found for all of my kids, just add plus one or plus two, depending, when they use those age sizes.

Steve Chou: And this might be common sense for apparel, but this was the first apparel item that we’d actually ever carried in our store, so we didn’t know, right?

Kurt Elster: Okay. Because I was gonna say with apparel, the number one objection by far is sizing, like will this thing fit me, and then the follow-up objection is, “Well, if I buy it and it doesn’t fit because this sizing is ambiguous,” so really, a lot of people feel like I’m gambling when I buy apparel online. Still, to this day. And then, so you need to be really up front with, “Here’s the returns and exchange process if this doesn’t fit.” And I flat out put it in there as, “What happens if it doesn’t fit?” Hey, we’ll pay for the return shipping, and we’ll send you a different size.

Steve Chou: Here’s the rub on this, though. So, we do personalized aprons, so-

Kurt Elster: Oh, no!

Steve Chou: You can’t return them, which was a further objection. I mean, these are all things that we learned, and this might be common sense to people who sell apparel, but we weren’t experienced in that, so these are all things that we learned just from cold calling people.

Kurt Elster: Yeah. No, like you know, no matter what you do, you don’t know what you don’t know.

Steve Chou: Yeah. Yeah.

Kurt Elster: And that’s why we’re hammering on you have to talk to people, because otherwise, you don’t know.

Steve Chou: All right, is it my turn or your turn? Was that yours or mine? I feel like we have the same peeves, pet peeves.

Kurt Elster: I have no… I don’t know anymore.

Steve Chou: All right.

Kurt Elster: Well, so we made some app recommendations, and that leads me to the other issue that I see, and this, especially with Shopify stores, is people… I call this app roulette. And you know, regardless of what platform you’re on, there are apps, plugins, and scripts, and like tools, and there is no end to the amount of really cool stuff you can use on your site, and these are shiny toys, and I call it app roulette because it’s like, “Well, if I just get the right combo of tools, if I have the right tool stack, my conversions will explode! My average order value will explode! I’ll make more money.” I totally get what’s going on here.

But, I mean, it just becomes detrimental, where you end up… You can see the sites that are doing this when you land on it and A, the site takes a while to load, so like you know that PageSpeed score is tanked, and then it’s like, “All right, spin to win! Punch the monkey! Enter your email!” I see someone who is familiar with punch the monkey from the ‘90s. Those ‘90s flash banner ads. That really ages me.

And you know that’s what’s going on is they’re like, “Well, if I can just throw more widgets at it, more shiny toys will fix it.” And it isn’t the case. It looks cluttered. It’s confusing. And then on top of it, when they go to exit, my Hotjar exit intent survey pops up and they’re like, “Not another effing popup!” That’s a real thing that I see when I run those on sites that have a lot of stuff. They’re like, “Why are there so many popups?”

Steve Chou: So, you probably see this a lot more than I do, but sometimes I’ll open up a student’s site and I’ll see like 40 apps installed.

Kurt Elster: Yep.

Steve Chou: And some of them don’t do anything except like install a small piece of code in there because they don’t want to touch their theme files and whatnot.

Kurt Elster: Yes. That’s scary.

Steve Chou: Like it literally just inserts a piece of JavaScript. Yeah.

Kurt Elster: Yeah. I don’t… Oh, edit my theme? Oh my gosh. You may as well… It’s like pop the hood and adjust the timing on our distributor. Like it’s a scary thing if you don’t know what it means, so I get why those apps exist. There’s a lot that’s just like you’re right, it just… It injects one line of JavaScript code and now you’re paying $8 a month for the rest of forever.

Steve Chou: I do want to say something that I’ve seen, and maybe you have more experience in this since you work with a lot of clients, but I’ve seen people uninstall apps thinking that it was clean, but in fact they leave a piece of JavaScript in there, and whenever you have a piece of JavaScript in there, they can literally track everything about your site if they want to do that, right? Have you seen that?

Kurt Elster: Yeah. If they’re insidious. All right, so Shopify specifically, when you… You have an app installed. When you click delete on that app… I think it used to say uninstall. I think it says delete now. But either way, it’s a misnomer. Really, what it does is it just immediately severs the app developer’s connection to your store. And it’s a security thing, like that’s, all right, you want these guys out? They’re out. The end of it. And the problem with that is depending on how the app works, some add theme code to the theme. I know I have four apps and all of them, we just put theme code in the theme. And we do it because that’s the most performant option.

Steve Chou: Right.

Kurt Elster: But as soon as you delete the app, guess what? All that stuff is still there. So, you know, we’ll get… Our solution is we immediately, we fire off an email, a transactional email to the merchant that says, “Hey, we saw you uninstalled our app. A, tell us why,” so we’re looking for that objection, and then, “B, okay, if you didn’t remove the theme code, here’s exactly how to do it and you have to do it. And if you don’t want to do it, just hit reply, we’ll do it for you.” I just… I don’t want to be part of the problem when it comes to performance, and so we do that.

Steve Chou: So, you have seen this before, right? With certain apps leaving?

Kurt Elster: Oh, yeah.

Steve Chou: Okay. Yeah.

Kurt Elster: Oh, nonstop. And so, when people are like, “Oh, my site’s slow.” Former CTO of Shopify JML said websites get slow slowly. And it’s because of exactly what you described. It’s cruft. It’s barnacles to a ship. It’s this code just accumulates over time. And even though it’s not doing anything, you know, the web server’s not smart enough to figure that out, and the browser’s not smart enough, and so, it just like… Every new site load has to load all this old garbage that you completely forgot about from two years ago.

Steve Chou: So, let me ask you this. I have students that like I’ll see this in students’ sites. Is there an easy way for them to do it themselves without actually having me step in?

Kurt Elster: Yes and no. So, almost all of the stuff is gonna be sitting either in the beginning or the end of theme.liquid. This is specific to Shopify. So, it’s like I’m just looking for the header and footer that loads in every page, and that’s gonna be in theme.liquid. And most of these apps, there’s either gonna be a comment on it, or it will be evident from the name what it is. You know, it’s like, “Well, I installed Acme Widget Popup Builder,” and you’ll see a line that says, “Include ‘Acme Widget Popup Builder’” so you know, okay, that’s that app I don’t have anymore. And so, you just safely comment that out or delete the line.

Steve Chou: A lot of people have theme editing phobia, though.

Kurt Elster: They do.

Steve Chou: Yeah. But-

Kurt Elster: Knowing just a little… You don’t have to be a theme developer, but if you know enough to be dangerous, and that’s the camp I put myself into, so just know basic HTML, which is not complicated, and maybe a little bit of Liquid and theme structure, you immediately are way better off. There is just so much more you can do that you didn’t realize, especially for like day-to-day stuff, like add and remove tracking code, verification codes. Like a lot of people recently had to verify their domain with Facebook. So, suddenly that’s the kind of thing that you’d be very comfortable and confident with.

And there’s plenty of free classes out there, like Skillshare’s got a ton of stuff. You can learn just like here’s the basics of how a theme is set up, and I could go learn basic HTML and CSS, and you will be in such a better position. Think about it. You’re a web professional. Think, if you are a merchant, whether you want to think of yourself that way or not, and your theme is very much like the face of your business. It’s your full-time salesperson. And you want to be able to speak its language. And so, if you learn those things, not only you’ll be able to do this stuff yourself, and you’ll be able to have better relationships and communication with any theme developer you may hire.

Steve Chou: I mean, you can’t be an online professional without being willing to learn something about the internet or websites. That’s my philosophy.

Kurt Elster: 100%. And like, whether you like it or not, you’ll end up figuring it… Some of this will sink in over time.

Steve Chou: Yeah. All right, here’s my biggest pet peeve. Going for revenue over profit. I one had this student show me their Facebook ads account and they were generating sales but the return on ad spend was terrible, but they were getting the sales, which was making them happy. But it wasn’t returning… Has that happened to any of your clients? I don’t know.

Kurt Elster: Yes, but I think they were more… When it’s happened, they’re more aware of it. They’re like, “Okay, we got it this far,” where it’s like this is a loss leader and we’re gonna make it up on subsequent sales, right? Oh, we’ve acquired them as a customer at a loss, which is what the big DTC brands do because they’re spending other people’s money. You know, they’re not bootstrapping it. And then, all right, on subsequent sales we’ll make it up. Or like drop-shipping businesses where margins are razor thin and they’re like, “Well, just revise, revise, we’ll keep iterating, we’ll dial it in.” But it’s like, “All right, how much time do you give it?” And I don’t know what the answer is.

Steve Chou: Well, I think… I’ll just tell you from with my store. So, we’re in the wedding industry, so you would think that we wouldn’t have that much repeat business, right? But 12% of our repeat business, 12% of our sales is repeat business, but it actually makes up 30% of our sales. But here’s the kicker. So, our average order value is about 60 bucks, and 50% of our customers spend less than half of our AOV, but they represent the bulk of our customers.

Kurt Elster: Interesting.

Steve Chou: Whereas I think only 10% of our customers spend 2X of our AOV but they actually make up 50% of our revenues, right? So, we’re a small business. We’re very small. We have a couple of employees, my wife and I, so where should we be focusing our time? We focus now on those big… We call them the whales. And it turns out that after doing some analysis, that a lot of our cheapie customers were coming from Facebook, and we only have a finite amount of time, so we now focus actually on the big repeat customers that we have in our store. And it just made life a lot easier for us.

Kurt Elster: So, you’re applying the 80-20 rule, Pareto’s principle there, right?

Steve Chou: Yes. Yes. It’s not quite 80-20, though. But yeah.

Kurt Elster: Right, but similar idea. I love that you’re doing this. I think you could apply that same thinking to all manners and areas of your life. But tell me, when you’re doing that customer analysis where you’re segmenting them, and it sounds like maybe you’re getting to like RFM model, how do you do that analysis? How are you identifying these customer segments?

Steve Chou: I can’t remember exactly how we came up with the idea with half the AOV, but we started with the average order value, right? What our average order value was. And then for our business, actually, a good portion of that 12% that I mentioned are event and wedding planners, right?

Kurt Elster: Ah. As soon as you said, “We had 12% are repeat with a higher AOV,” immediately in my head I went, “I bet they’re wedding professionals.”

Steve Chou: Yes. And so, what ended up happening is we start now looking for anomalies in our sales, and then we pick up the phone, like you said. And we’ll call them up and we’ll say, “Hey, we noticed that you ordered a lot. Are you a planner? Are you gonna order in bulk? I’ll tell you what we’ll do is we’ll give you a special coupon code and a representative from our company that will handle all of your transactions and make sure they arrive at the destination on time. We’ll give you handholding, essentially.” And after that, that call, once we have established some rapport, they continue to order from us over, and over, and over again, and they’re basically a customer for life that orders in bulk, they don’t complain about anything. It’s like the best customer ever.

Kurt Elster: So, essentially… This is brilliant. So, what some people do, they say, “All right, I want those wholesale accounts because they have high AOVs.” And so, they’ll set up a wholesale program, but it’s just a link on the site and they expect people to go sign up for it. You are proactive about it. You’re identifying these people that are wedding professionals, and these are your like… Really, these are your ideal customers who have the biggest lifetime value. And so, and then, so once you identify them, you call them up and say, “Hey, we see you and we want to help you, so here’s your lifetime discount code and we’re going to…”

Really, like their big fear is a wedding is a mission critical event, so you’re saying, “Hey, we’re gonna make sure you get your stuff.”

Steve Chou: Yep.

Kurt Elster: We’re gonna look out for you. And then, as soon as you do that and you are so proactive, I imagine that you just have customers for life with them.

Steve Chou: The phone call is actually pretty important in this case. Again, real contact.

Kurt Elster: It’s that phone call.

Steve Chou: Yeah.

Kurt Elster: You’re no longer… It’s no longer a brand. It’s, “Oh, it’s Steve. Steve’s helping me out.”

Steve Chou: Right. Another pet peeve is making a guess at who your real customer is and then writing all your copy to account for that imaginary person when you don’t have any data. I don’t know if that happens-

Kurt Elster: It’s usually like they’re thinking about themselves. Most people, their first customer, they are their own best first customer, and so they’re writing to themselves.

Steve Chou: The only reason I can talk about all this stuff is because I made all these same mistakes. Just to be clear for anyone who’s listening.

Kurt Elster: Yeah. I’m sure we’ve both done some boneheaded things at the time.

Steve Chou: I mean, I’ve got an example on this line, too. I assumed that all our customers were wedding customers, but the reason I found this out is I was looking at my Facebook demographic data and I noticed that a lot of our customers were over the age of 55. And I’m like, “There can’t be people over the age of 55 getting married, like a whole bunch, right?”

Kurt Elster: Yeah. This was an outlier.

Steve Chou: This was an outlier. But it turns out a lot of those repeat customers are people who just like to collect handkerchiefs, believe it or not.

Kurt Elster: Oh, what?

Steve Chou: Yes, right? Just like you like to collect old cars-

Kurt Elster: I do.

Steve Chou: Kurt just got a Volkswagen Beetle, ’79 Beetle, which is pretty sweet. There’s people who collect handkerchiefs, or they’re crafters, and these tend to be older women who do these things.

Kurt Elster: And so, you had this customer segment that you were not aware of. Once you knew that, was there any actionable info there? Anything that you changed?

Steve Chou: Yes, actually. So, in the old days, every single page was about weddings. Every landing page was about weddings.

Kurt Elster: Okay.

Steve Chou: And so, now we have a special section for weddings now, but the rest of it is just talking about either embroidered blanks for crafting, or some of these older ladies who like to collect them. Like the copy has changed, basically.

Kurt Elster: Okay. And so, initially, we talked about like you said, “Hey, the big pet peeve is people just assume they know what the customer is like and then they write to that customer.”

Steve Chou: Right.

Kurt Elster: What’s the correct method? What should they be doing instead?

Steve Chou: The correct method is you’re not gonna know from the beginning. So, you can start with some assumptions, but again, you have to as you said before, talk to the customer. We have a survey in our post-purchase sequence that kind of asks them what they’re using it for and that sort of thing, and then based on this survey data, we make adjustments. We actually don’t call those customers because I guess it would be just too many. We only call the whales. But yeah, we’re constantly trying to get data about our customers so we can do things appropriately.

So, for SMS, for example, sometimes we’ll ask. You know, what would you like us to carry in our store and what would you use it for? Something like that.

Kurt Elster: What would you use it for, I think that’s the critical question. Hey, you’re trolling for product suggestions. Hey, what do you want us to sell? But then like why is really… I think the follow-up question that most people miss. You need to scratch a little bit deeper and figure out what the intent is, because that really adds so much, so much info and tone to why they’re making that request.

Steve Chou: A perfect example of this is my buddy Neville, he used to run an eCommerce store a long time ago, and he used to sell rave supplies. So, there’s this one product that he sold, which were fingertip lights, and I’ve never… I don’t really go raving, but supposedly you twirl these around and whatever.

Kurt Elster: Yeah. I know what you’re talking about. I can picture this.

Steve Chou: He found that plumbers were buying this, these fingertip lights, so they could…

Kurt Elster: Of course.

Steve Chou: When they’re under the sink, they have lights on their fingertips, so it’s convenient and they can see what they’re doing.

Kurt Elster: That’s pretty smart.

Steve Chou: So, he started writing copy to plumbers and it ended up converting really well. Stuff like that.

Kurt Elster: That’s great. Yeah. Oftentimes, things don’t necessarily have to stick to their original purpose. Like you mentioned, oh, Kurt’s got a ’79 Beetle. I have been trying to find the perfect mirror to fit behind, in this tight, hard-to-reach place where I’m adjusting a carburetor. And you know, I bought an automotive mirror like from an auto parts store, didn’t work. The mirror I found that worked, I’ve been using my wife’s compact. I have stolen her compact and that’s actually a much better mirror. It’s much more useful. So, like that’s clearly not the intended use there, but that’s what ended up working.

Not that I’m gonna go remarket compacts as automotive looky-loo mirrors.

Steve Chou: Yeah, right. I was about to say.

Kurt Elster: So, you said post-purchase survey. What are some of your favorite post-purchase survey questions?

Steve Chou: I’m trying to think. Well, mainly what they’re using it for, and then here’s the kicker for ours, and this is specific to our business. We ask them if they are a professional in this business, like are they buying it for business, or pleasure, or as a gift? And if they say business, then we call those people.

Kurt Elster: Oh, yeah.

Steve Chou: That’s actually our primary purpose.

Kurt Elster: They have raised their hand as a whale.

Steve Chou: Yeah. That’s actually our primary purpose for the survey, actually, for our store.

Kurt Elster: That’s pretty bright.

Steve Chou: But everyone else has different uses, obviously. But yeah. So, I guess sometimes we’ll ask them if they like the product, and if so, we’ll ask them for a review, or testimonial, or have them take photos for social and that sort of thing. That’s what we mainly use it for.

Kurt Elster: Do you do any split testing?

Steve Chou: No. Actually, I was gonna talk about that. We actually don’t do a whole lot of split testing mainly because split testing takes forever.

Kurt Elster: It does. It’s frustrating.

Steve Chou: It mostly fails. It mostly fails, like it’s mostly inconclusive, so I’m curious to see what you have to say about that, actually.

Kurt Elster: Okay, so you’re right about both things. Any test, minimum, needs to run two weeks, and that’s a frustratingly long time to wait. And during that time, you’re just like sitting there refreshing it, checking it, and your probability to be best just keeps flipping back and forth between the two. So, for a lot of stuff, that’s true. It really… It turns out, I think the conclusion with split testing is for many split tests, design is way less important than we think. The copy, and the content, and the offer, and the quality of traffic, those really are the much more important factors to conversion.

And with split testing, I think a lot of people just use it to split test design changes, and that’s why they end up with inconclusive results. Or you know, they’re calling it too early, or they’re starting them too early where you just don’t have enough traffic. So, one of the things we do is I’ll usually use revenue as my primary goal as opposed to conversion, because at least now I’m factoring in average order value and conversion by doing that, and I think I’m getting a better signal to noise ratio by using that. I think that helps.

Let it run long enough, like two weeks, and really try to limit the number of tests you run at once, or the number of variants. Otherwise, the thing really takes forever to get anything conclusive. But for the most part, I think a lot of stuff really is just very subjective in that it doesn’t have… You need to use common sense on it. But for other things, it’s like, “Man, should we do this? Is this a good idea?” One phenomenal split testing thing we’ve done is figure out what your free shipping threshold should be, and this is a tough one to do. I found an app that’ll do it called ShipScout, and you have to be on Shopify Plus to do it, so kind of limits who can do this, but it’s really cool. The banner across the top offering free shipping, that threshold will change, and then once you’re in the checkout, the shipping rates will change.

And so, you can figure out through data, okay, here’s… like whether or not you’re leaving money on the table with your free shipping threshold. So, stuff like that, absolutely split test those offers, because that’s gonna have a real, real difference in revenue on the business.

Steve Chou: Let me ask you this. How much traffic would you recommend to even think about split testing? Personally, so our site I think gets about 50,000 visits a month. I don’t even think that’s enough to complete a split test unless I do a whole sitewide split test in a couple weeks.

Kurt Elster: Yeah. It’s gotta be… Well, I think the traffic is an issue, and then on top of it, it’s you also need the conversions. So, if it’s like a high… a moderately-trafficked site with low conversions, you’re probably not a good candidate.

Steve Chou: Yeah.

Kurt Elster: But yeah, once you get to 100,000 visits a month with at least a 1% conversion rate, all right, now we’re cooking with gas, and you should be able to split test just fine at that point.

Steve Chou: Right. I can see the free shipping offer working because that’s like a generic conversion event. But if I was trying to test some UI change on one page, that one page would have to get a lot of traffic, right?

Kurt Elster: Right. And that’s the issue we get into, like I’m running some split tests now and even on a site that gets… that’s in the 1% of websites as far as traffic goes, running a split test on an individual product page, it’s never gonna get… It’s never gonna happen. It will be years before I have a statistically significant result. So, even on these huge sites, if you make the test too specific you run into issues.

Steve Chou: So, I would say, and you can agree with me or disagree with me here, if you’re running a smaller site, I think something like PickFu is probably better than running a real split test. You’ll get results in 15 minutes from random people.

Kurt Elster: 100%. I also find heatmapping, where it’s like heatmap, scroll map, click map… Not click map. Movement map, where you can see where their mouse goes on desktop. Those are very valuable even on sites that have less traffic. That, there you go, that’s a heuristics analysis tool you can use way before split testing starts to make sense.

Steve Chou: And here’s an example of just something… why heatmaps and scroll maps are useful. There’s this one student I had which was dead set on keeping their sidebar. Dead set. Even though it didn’t look good, and it was like compressing the size of the images on the category pages. And so, I was like, “Okay, let’s just run a test.” And we finally discovered that no one is clicking on anything in the sidebar. No one’s even looking over on the sidebar. And so, finally, with that data, they removed it.

Kurt Elster: Yeah. And like we were talking about social buttons earlier. That’s a thing that you can very easily use a heatmap to figure out, like are people messing with this thing or not? Well, the heatmap says no.

Steve Chou: Dude, Kurt, we’ve been talking for 50 minutes about pet peeves. You got anything else to add before we wrap this up?

Kurt Elster: You know, I think ultimately my pet peeve is not starting. I think you need to get out of your own way sometimes, ask yourself what would someone smarter than me do, and then go… Turns out, you already know the answer, and then go do that. I think for a lot of entrepreneurs, the biggest stumbling block is not any of this stuff. It’s just taking action and moving forward. So, even if you’re out there making mistakes, that’s great. At least you’re doing something.

Steve Chou: My pet peeve is not being willing to do the legwork or the dirty work.

Kurt Elster: Yeah. You need to do the work.

Steve Chou: Most people create an online business and think they could just do everything online. But doing stupid things like picking up the phone or actually talking to someone face to face, or being a little bit more personal, I mean, these are things you gotta do. It’s just like a regular brick-and-mortar business in that respect.

Kurt Elster: 100%. Mr. Steve Chou, where can people go to learn more about you?

Steve Chou: Yeah. If you guys are looking to get married, I can hook you up over at That’s my eCommerce store. But if you want to learn more about eCommerce in general, head on over to I offer a free mini course for beginners. It also covers advanced people. If you want to create content, sign up and that’s a free mini course right there.

Kurt Elster: Fantastic. Thank you, sir. This has been a ton of fun.