Peep Laja from ConversionXL.com, is the go-to guy when it comes to converting customers. If you’ve been wondering why your conversion rates have been sliding or you want to learn some best practices on getting customers to convert, Peep is here to help.
We also cover when you should run an A/B test and discuss the most crucial things eCommerce store owners should be focusing on to increase their eCommerce conversion rate.
Andrew: Thank you so much for tuning in to today’s show. And today on the program, I’m joined by Peep Laja from ConversionXL.com and if you’re an eCommerce, which of course, you probably are listening to this, you’ve undoubtedly heard of Peep. He is the founder of ConversionXL.com, which is one of if not the most well regarded and respected conversion blogs online. It puts out a ton of great content there, runs a conversion agency so he really is in the trenches and knows this stuff. And also holds an annual conversion meeting and events Conversion XL Live in Austin every year. And we sit down and talk conversion, talk a lot of different things, discuss why conversion rates have been sliding for a lot of eCommerce stores over the last two, three, four years and part of it is due to mobile, part of it is due to something else, which we get into.
Andrew: We talk about the approach, when you should and shouldn’t A/B test. A lot of people think you should be testing all the time, sometimes it doesn’t make any sense to you. We cover what the most crucial things for an eCommerce store to focus on from a conversion aspect are, delve into a lot of different things as well as talk about his upcoming events in Austin. So, I’m going to go ahead and get right into it with Peep Laja from Conversion XL.
Andrew: Peep, people always talk about how you have to be split testing. Split test, split test, split test, it’s kind of this mantra but you came out recently with an article that talked about maybe why you shouldn’t be A/B testing, times when it doesn’t make sense to do that. Can you talk about when you should be pulling back and maybe not be diving headlong into that approach?
Peep: Sure. So with A/B testing, we’re trying to understand whether there’s a significant difference between different variations of a webpage. In order for us to understand whether there is a difference or not, we need data samples and we are only able to tell if there is a difference if the sample size is big enough. So low traffic websites unfortunately just don’t have enough transaction volume. Basically, we won’t have enough evidence to know one way or another. So if you have less than probably speaking 1,000 transactions per month, and the transaction could be either purchases or whatever you want to test, like can be also email subscriptions, if you have less than 1,000, it’s probably too early for you to test. Yes, you could get away with less transactions for a successful A/B test but then the treatment needs to win by a huge margin like let’s say 30% better or more because if the treatment is better by 4% and you only have 200 transactions in each variation, then you don’t have enough data to know whether it actually is better. Hence my advice that if you do not have enough transaction volume, you’re better off investing your energy in other things like improving your overall service or figuring out your customer acquisition channels, building traffic, stuff like that.
Andrew: So Peep, when you don’t have enough conversions like you’re talking about to really dive deep into split testing, how can you cheat a little bit? What kind of things can you do that can kind of give you some of the similar insights but without maybe the cost and the time of running all those split tests?
Peep: A/B testing is for validation whether something worked or didn’t. So we have two unscientific ways to kind of do the same thing but with less validity. Now, before we run a test, we should know what the problem is that we are trying to fix. So, whether we’re doing A/B testing or one of those other things I’m about to tell you, you need to do your research first. Your qualitative, quantitative and so on so you would know where the problems are, where people are dropping off and why they’re doing it. So, essentially beforehand, you need to identify a list of highly specific problems that you’re now trying to solve and you validate the solution through A/B testing. So, when you can’t A/B test because you do not have enough transaction volume, one thing you can do is that you implement a solution to those identified problems all at once.
So, let’s say you’ve figured out 100 problems or the clarity of your valley preparation was poor or your product images were low quality and maybe not clickable, your “Add to Cart” was too small and so on and so forth. So, all these problems that you identified and now you want to fix them, so you fix all of those problems at once and then you go to your digital analytics tool, whether it’s Google Analytics or whatever and you want to see a lift, a change in your results basically overnight. So if you had 100 transactions per week, now you want to see 130 or 140 transactions per week. You need to see a lift at least 20 or maybe 15% or bigger in order to know that the changes that you made actually had a positive impact. If it’s 10% or less, you really don’t know if that is just noise or regular traffic fluctuation and all that stuff. So that’s one thing. Just change everything at once and hope for a 20% or bigger lift in transaction account.
Number two would be that you set up an A/B test like you normally would but in addition to measuring the final transactions and revenue, you will also start measuring all the micro conversions. So let’s say you run a test on your product page, so you measure the clicks on the add to cart button and you measure each final step, the cart eCommerceFuel Podcast, step one, two, and so on and so forth. And so, now you run the test maybe for four week, 28 days and at the end of the 28 days you’ll have 60 versus 70 transactions, so not enough evidence to know whether the treatment is actually better. But now you look at the micro conversions and said hey.
So, we have 30% more add to cart button clicks at very high count, maybe thousands or hundreds and then also more people get to the check out and start filling out the form. So then you can use those micro conversions as indicators that so even though I don’t enough evidence to know that it’s better but micro conversions are also all up, so it’s probably better. It might not be but it’s probably better or at least maybe it’s not worse. And if you’re on your A/B testing with VWO, with Belgian [SP] statistics, they’ll also tell you the probability of loss. So they’ll measure two things, not just how confident we are that we have a winner but how confident are we that it’s at least no difference so it’s safe to implement it no matter what.
Andrew: So really what you’re doing is you’re just kind of looking at the matrix side of the most impact on your conversion rate going upstream to get a larger sample size so you can be more confident.
Peep: Exactly right. But again, this is highly unscientific and you’re going to get stuff wrong. Like sometimes all your micro conversions are up, but it actually has negative implications on your bottom line because you imagine you put everywhere signs free beer, free beer, free beer and they get to the final phase, there is no free beer, they have to pay for it. So the micro conversions are all up but the final transactions are down because they are pissed at you.
Andrew: What percentage of either big site re-designs or migration things like that do you see taking kind of the change everything at once and see how it works out? Because that’s historically…that’s what I’ve done just because thinking through, you talked about thinking through, or from our side just more of the testing. Let’s say you’re changing 30 or 40, or if not 100 things on your website trying to isolate the impact each one of those has individually is a really time consuming process. So, what percentage of the redesigns that you do for clients is a change everything and see how it impacts the bottom line versus a let’s test everything we implement and see how that independently affects the outcome.
Peep: A radical redesign where you change all the website is extremely risky and often I would say 50% of time it backfires meaning in most cases, I would say 60, 70% of cases, there is no difference whatsoever in conversion rate. So you spent all that time developing and designing and wasted all the time and money for nothing or in another maybe like 20% of cases, the conversions actually go down and this is where the changes were made because of silly reasons, “Oh, we need some flashy sliders and we need some other random ideas I read about in a blog post.” Radical redesigns need to be managed and done extremely carefully. You again start with proper research. And because you want to figure out which bits of your website are actually working and which are not working.
So the layouts that are working you want to keep the same. You might do a facelift but what goes where is the same where some pages where it’s really performing poorly, you want to completely rethink. So for instance, on your cart page the average is 50% drop off in the funnel. So if your cart is 30% drop off, that means it’s good. You don’t want to change that. Or if it’s a 80% drop off on your cart page, you absolutely want to change everything about the page because that page sucks, and so on. So you use this average benchmark, you’re like your product page add to cart ratio 10% is pretty typical so if it’s much lower than that, it’s crap. If it’s much higher than that, it’s good and so on and so forth. And of course, it’s all very contextual. It depends on what you’re selling and how expensive it is and so on.
Andrew: You just said it’s contextual and so completely understanding that it varies based to industry, it varies based to pretty much every unique business and situation. But inevitably I and I know other people listening, what are the averages? Give me some averages. You said 10%, could you give me a sense of what going through the funnel maybe from homepage to product, well, maybe we can take a product page down to the eCommerceFuel Podcast? Can you give us a sense of product page to add to cart to starting to check out to all the way through what percentage you’re seeing so people can get a sense of that and some of those average matrix?
Peep: Every single page has one goal. Every single page has one goal. So goal of a category page is to get them to click on the product page. On the category page, that’s what you measure and on the product page it’s add to cart click. So on the add to cart click, typical is 10%. So if less than 10% click on the add to cart button, it has room to grow probably. If it’s like 15%, you’re doing fine. On the cart page, 50% advancing to check out from the cart is the average, typical so if you do less than that, you’re in trouble, if you do more, congratulations. And in the checkout, now, there is no correlation between eCommerceFuel Podcast steps and performance. I have seen seven and six steps checkouts work extremely well and one step eCommerceFuel Podcasts that perform terribly and vice versa.
So rather I would say that don’t focus on the number of steps here but rather on the percentage of people dropping out once they get to the checkout because now they should be pretty motivated because they found the product, added it to cart, confirmed that they wanted to start checking out and now it’s about filling in their data and stuff. So, 90% plus is highly possible. So I would say that if in your checkout under 80% of people are complaining their purchase, there’s something wrong. And that’s good news because that means there’s a huge possibility for your business to do better. It might be that the checkout has terrible usability and what not, possible. It also might have to do with their motivation or fears like physical, emotional side what’s going on inside their mind and also distractions, external distractions. So, on mobile phones, you see bigger drop offs because they’re more easily distracted because they might be driving or whatever. Those are kind of the averages.
Andrew: You mentioned how conversion rates change a lot over mobile than desktop and one thing I’ve seen in my own business, one thing I think is happening to a lot more store owners because I’ve heard this from them is, especially here in the US, is the slow erosion of eCommerce conversion rates over the last four to five years. And part of that, if you look at the great part of that’s due to this explosion of mobile traffic. Just almost every mobile site or mobile conversion rates for store owners I’ve talked to is well below the desktop on average, right? And…
Peep: It depends on what you sell. But yes, in most cases that’s true.
Andrew: Yeah, which makes sense, that makes sense. You would expect to see, let’s say if you have a fixed pool of people, they’re moving from shopping desktop to mobile and maybe some of the purchase on mobile but a lot of them just find your site on mobile and then they come through and convert on desktop because it’s easier. That’s a theory a lot of people have and you would assume that if that’s the case, your mobile conversion rates, okay it’s going to be lower but you’re going to see likely an increase on the desktop side for the conversion rate because you have more people coming further down in the purchasing process. They come to you and purchase more quickly without visiting two or three times to learn. But that’s not what I’ve been seeing and what other people have been seeing. Have you been seeing this in the clients you’re looking with? And if so, do you have any theories on where these shoppers are leaking out of? Are they going to Amazon’s mobile site? Is it just…what’s happening there?
Peep: So I think there are two…yes, I have seen this and have had many, many discussions with my colleagues and clients about it. So there are two underlying trends that I think we can blame or three even. One is that there is just so many more eCommerce stores around. So there used to be way less but now anybody can go to Shopify and open one that looks pretty decent overnight and start selling. The money is distributed among a bigger range of online stores. Or some when we look at the really big online stores, let’s say Amazon, and you look at their annual revenue growth whether globally or in the US and it’s growing by the billions every year. So it used to be in a 20 and now it is 25 billion. So where did that extra five billion come from? Well, from all the other stores.
So, Amazon is eating everybody’s lunch and it’s true. Because with Amazon, people are not doing price comparison and stuff like that because one click and you have it. You don’t even need to fill out a form. I personally don’t shop in any small businesses because I just hate putting in my credit card data, my address, I have it in my Amazon, why would I bother? So that’s two and three, the number of mobile traffic data has exploded. It has brought in more people using the internet especially in the poor income classes where they might not have a desktop or a laptop computer at home but everybody has a smartphone and they’re surfing. So that kind of more people, less purchases brings your average conversion rate down even though your absolute numbers might not be down.
Andrew: I think it would be interesting. I’d love to see, I’m sure we will see some kind of technology like Apple Pay but in a browser because right now you can use it but you have to have your own app. But if we see that come over to the browser and also some kind of technology making it easier for eCommerce store owners to capture lead, capture an email address or something like that on the mobile version and then convert them more on the desktop.
Peep: Yeah, yeah, yeah. And I would encourage all eCommerce site owners to experiment more with alternative payment methods other than credit card like Amazon payments and stuff like that where somewhere where my address and my credit card is already stored. When we’ve experimented with a number of payment options on a site, so two always kicks one’s ass. So basically adding PayPal to credit card already makes it better and adding a third one typically makes it even better. And if you add a fourth and a fifth one, well, now it’s like, sometimes it gets better sometimes it gets worse. In Europe, that’s complicated because in Europe you have a plethora of payment systems because people want to pay with their bank essentially and so on. So we’ve done some experimentation there.
Andrew: We actually just got on adding Amazon payments to our Shopify flow and probably 10% plus of people pick Amazon payments for…which makes sense because everyone’s got an Amazon account set up.
Peep: I think the more stores add that as an option, the more people start using it. I think it’s still in its infancy. I think more people will start using it once they experience how convenient it really is.
Andrew: Peep, you had a great article on the Ultimate Guide to Increasing eCommerce Conversion Rates and we’ll link up to that in the show notes. It’s worth a read in its entirety. Would you mind mentioning just the lead on that? What are the two to three biggest things an eCommerce store owner can focus on to increase conversion? Obviously, with the caveat, everyone’s situation varies. You got to test on your own but forcing you to go back into the broad sweeping generalizations, what are those two to three biggest things people should be really hammering on?
Peep:: Yeah. Well, if your product is at all about the design like clothing, shoes, flowers, what not, image is just the key thing. People don’t read the copy, the product description, they all look at the product photo. Whenever I’m working on an eCommerce site, putting an emphasis on photos, the quality and the size, the bigger, the better. And the quantity, how many visibly different photos per product do you have? That’s like a sure. Even one of our clients, Diamond Candles, selling scented candles. You buy it because it smells good. Bigger images increase sales. It’s like pitchless people. People are visual. When you look at how a brain is constructed, we have the old, the reptilian brain and that processes visuals. It can’t read, it can’t understand text and that’s the decision making part of the brain. So, I would often put visuals above all and same goes for category pages.
Having bigger thumbnails and lesser products in a row, typically makes a difference, a positive difference. So that’s one big thing and of course, not all products are visual. Other things that are really easy low-hanging fruits and most eCommerce sites that I see are not doing it, is communicating maybe three key reasons to buy from you on key pages where people land. Because people usually put their maybe, maybe they put their value proposition on their home page but if you look at your top landing pages and when you group category and product pages into groups, you’ll get way more traffic landing on your product and category pages than on your homepage. So if people and people who land on your category and product pages are probably searching for something specific and your SEO kicked in or your PPC ad.
So they’re looking for something specific, they find it on your site, they land on your page and the only thing you say is, “I have this Nike shoes, it’s like $50. Buy it.” Well, that is every other store saying the same stuff. So if what you sell is not unique and even if it is, you need to have on those pages near maybe in a prominent position, maybe on a category page before the product, product page before the add to cart. Basically, three bullet points or it can be a horizontal line or three key reasons why I should buy from you and not the other guys. We have free shipping, we’ve been in business since 1977, you can trust us, and we have five million customers. Social proof, you’re not the only idiot buying from us. Or whatever the case, you have to know who your competitors are and what are the actual three reasons to buy from you. And if you don’t know of course you’ve got a whole new set of problems.
Andrew: Something I’ve been focused a ton on this year is not blindly optimizing for conversion rate but rather optimizing for profit. And is that something that you guys focus on as very based on each customer or each client that you have and if so, let’s say you are taking the profit based approach, what kind of metric do you use? I’ve kind of settled on a profit per visitor one that can kind of…because when you start testing for profitability, you got all these different moving parts in terms of your conversion rates go down but then you got to measure the increase rate on profit but then you also have to control for maybe the different visitors and different seasonality. What do you optimize around? Is it just conversion rates? And if it’s not, what metric do you use?
Peep: So for eCommerce, conversion rate is definitely not a final godly metric to worship because when you slash your prices in half, your conversion rates will go up but you’ll make less money. So, conversion rate is just an indicator. Revenue per visitor is already a better indicator because then you can optimize for a higher cart value or stuff like that or they’ll buy more expensive products and you’ll make more money. So revenue per visitor which is easy to measure and track with a testing tool or Google Analytics is already better than conversion rates. Now, of course profit is even better. So the tricky part of measuring profit is that not every store has their, I don’t even know what the polite way of saying it, has their [beep] together.
Their analytics implementation is lacking. That’s basically…people don’t want to invest in analytics implementation even though it is so important. I encourage everybody to spend money on analytics and I’m not selling analytics. So I don’t have a dog in this race. So, what you can do…first of all you need to use enhanced eCommerce. You get these all new metrics that you can see and when you’re running A/B tests or whatever you’re changing, you always analyze the results in Google analytics not the testing tool. So all the testing data should integrated with Google Analytics and you look at the different metrics across different variations in your analytics tool.
So, custom dimensions and all that great stuff. So, maybe you guys have heard of the terminology single source of truth? Because you know a typical, especially small business owner is that, “Oh, I have my visitors data here and then the purchase data there and then the CRM is our third system.” I have different data in different places and putting them together is really…don’t know how to do it or whatever or it’s expensive. So you can really do that with Google analytics. Of course, there are much more sophisticated enterprise level tools with that like RJ Metrics and what not but with Google Analytics, if you get a Google Analytics implementation guy, he can add this custom information for every product that is in there which is your profit margin per product and all that stuff and all that rich information can be available in Google analytics for you. And so, if you have that now, you can really optimize as you say it for profit.
Andrew: Sorry, just to clarify, are you saying profit per item, that’s not currently supported by Google Analytics, is it?
Peep: You can add any custom data into your analytics that you want, really.
Andrew: Okay. So but it’s probably…
Peep: So you need to be using Google tag manager and through data layer you inject all kinds of information onto your page as you want whether it’s your basically product taxonomy and your product can have individual profit margin information in there already.
Andrew: Yeah. Do you think we’re going to see that at some point be kind of a native feature of Google analytics because they’ve already got conversion rate by product and a lot of these more product-oriented things but out of the box, it doesn’t seem super easy to assign a profit per item.
Peep: Yeah, you should budget depending on who you’re going to get but basically 10 to 25K to get this stuff all implemented with an expert. So yeah, I think it will come available. Maybe it will go to Google Analytics premium first because they’re sure are pushing that one more now and also the new A/B testing tool coming out, Google will convert, that’s also going to be a premium product first and foremost. Yeah, so far I would recommend find an affordable analytics and implementation guy to get it set up.
Andrew: Peep, you had a conference last year down in Austin, Conversion XL Live. Sounded like it went really well and you’ve got one coming up March 31st to April 1st this year in 2016 in Austin or just outside of Austin rather. Certainly a fantastic event. I, unfortunately, wasn’t able to make it but I had hoped to. Can you talk a little bit about how the last one went last year went and what people can expect if they’re thinking of coming this year?
Peep: Well, Conversion XL Live is a unique event in the sense that it’s happening in the middle of nowhere. So it’s one hour outside of Austin, Texas and it’s an all-inclusive meaning your ticket to the conference also includes two hotel nights all meals. So everybody stays in the same place for three days, two nights. Nobody leaves. So that creates the best networking you can imagine because everybody is just there doing everything together. It’s so much fun, great parties and of course the content is just hands-on practitioners only, really kickass people, mostly focused on conversion optimization, squeezing more money out of your visitors and how to do it. Day one is our psychology and persuasion, understanding people’s minds. Day two, we’ll have on process of A/B testing and optimization and day three we’ll have more like traffic, analytics, personalization, retention, stuff like that. So, all things very, very relevant to eCommerce people.
Andrew: And Peep, for someone who’s maybe…you do agency work as well, take on eCommerce clients as well, of course. For someone who’s maybe thinking, “Oh man, this Peep guy he knows what he’s talking about, I don’t have time to do this. I want to hire him to come on and help me out.” Who’s a good fit for your agency? Who do you work with? What kind of eCommerce store potentially is someone that would make sense for them to maybe give you a call to potentially talk about working together?
Peep: So, we run a lot of A/B test, we do done for you conversion optimization. So it’s hands off for the client and we run a lot of A/B test. In order to run tests, you need to have the transaction volume, the traffic. A good fit is somebody who’s annual revenue is more than $10 million and who has thousands of transactions per month. So, that’s kind of the starting point because really that’s where we can do a lot of the fun stuff.
Andrew: Well Peep, it’s been a blast having you talk about this kind of stuff. If you’re interested in either Peep’s agency work or Conversion XL Live, you can learn more about both of those in the blog and the agency at conversionxl.com. And for the live event, that’s live.conversionxl.com. Peep, thanks so much for coming on. I really appreciate it.
Peep: Thank you so much, Andrew.
Andrew: That’s going to do it for this week. If you enjoyed the episode, make sure to check out the eCommerceFuel private forum. A vetted community exclusively for six or seven figure store owners. With over 600 experienced members and thousands of monthly comments, it’s the best place online to connect with and learn from other successful store owners to help you grow your business. To learn more and apply, visit ecommercefuel.com/forum. Thank you so much for listening and I’m looking forward to seeing you again next Friday.
Photo Courtesy of Peep Laja