441: Website Optimization, A/B Split Testing and Heatmaps: How to Double Your Customers, Sales and Profits with Conversion Fanatic Justin Christianson

Robert Plank 0:01 Marketer of the day number 441, website optimization, AB split testing and heat maps, how to double your customers sales and profits with conversion fanatic Justin Christensen. Hey everyone, and thanks for listening. We're talking right now with Justin Christianson, and he is the number one best selling author of conversion fanatic, how to double your customers sales and profits with AB testing. And he's the co founder and president of conversion fanatics, which you can find at conversion fanatics.com and they're a full service Conversion Rate Optimization Company. And he's also the host of the weekly Marketing Podcast, the CMO roundtable. So we're gonna have a lot of fun stuff to talk about. How are things today? Justin

Speaker 1 0:46 not too bad. Thanks for having me on the

Speaker 2 0:48 show.

Robert Plank 0:48 Glad to be talking to you. And so, so, yeah, so you do, you optimize conversions, you do a B testing and things like

Speaker 2 0:56 that.

Speaker 1 0:57 Yeah, that's so what we do is we basically take companies who are already doing well, doing some amazing, fantastic things, and we help them do it better, essentially, and really help them identify and understand their audience behaviors and how they interact with their brand and their sites and their landing pages, and just get them to take more of the desired actions that our clients want them to take.

Robert Plank 1:20 All right, so when you say you work with companies that are doing well, what's the criteria for that? How do people know that it's time to look into optimizing the conversion rate and doing tests and things like that.

Speaker 1 1:32 You should always do it, you know, I would say, do it qualitatively from the beginning. You know, really just asking your visitors what's going on, but you hit kind of a threshold where you need enough traffic to actually justify doing an actual AB testing campaign and a process and an actual optimization process, and that's usually about 30,000 monthly unique visitors, is kind of the minimum threshold. And then after that, it just gets easier. If you have less than that, then it just takes forever. Takes forever to conclude test results and it just you're wasting your time. So I generally just tell people, if they have less than that, to do qualitative analysis. First, you know, just surveys and understanding and using heat maps and click maps and just kind of changing things to massage it and tweak it to where you can get to a point of growth.

Robert Plank 2:21 Okay, great. So there's, there's the in the early stages, we can have some of these tools like heat maps and things to eyeball it, but after that magical 30k monthly number, then we can really dial it in and get more scientific on this.

Speaker 1 2:35 Yes. So,

Robert Plank 2:36 so great. So just to get everyone caught up, can you explain to everyone listening what heat maps are and what AB testing is.

Speaker 1 2:44 Yeah, so heat maps are essentially a software tool that, you know, things like hot jar, Crazy Egg, true conversion tools that you plug in just add a JavaScript to your page and tell it where you want to read a heat map, and it'll essentially tell you how far people are scrolling on the page, where they're clicking on the page, what they're paying attention to, what they're ignoring, so it just gives you a glimpse. And you can even do screen recordings where it'll actually record people's mouse movements and where they're clicking, just to get a better understanding how people are interacting with your landing pages. And then AB testing is really our tool to either prove or disprove any of our assumptions. So you never really want to make any changes to your page unless you're actually split testing it. You know, when you've got enough traffic to actually split test. So maybe we're going to test a headline, instead of just changing the headline and then just seeing kind of what the numbers look like. We're actually measuring it side by side, so we're splitting up. 50% of the traffic is going to headline a, 50% of the traffic is going to headline B, randomly, and we're seeing how many conversions it produces on the other side, whether positive or negative.

Robert Plank 3:59 Great. And thanks for explaining that and getting us caught up on that. Because, because, yeah, and that seems like a really important thing, because, because, like you said, somebody might be tempted to say, well, let me just make a tweak to the website and see if that makes a difference. But what if someone made a tweak to that website, just as the holiday season was ramping up, they might change the headline and say, whoa. Suddenly, we got all this set, all these extra sales, all this extra stuff, and it was, you know, it was because of an increase in traffic, or maybe, you know, some, some maybe, like, the website was on, like the radio or the news or something, and some other unforeseen event caused the boost. And then we're kind of shooting in the dark. So by AB testing, we can for sure tell that that little tweak you made to the website, to the headline or something, made all the difference, right?

Speaker 1 4:45 Yeah, we want to see how much it improves, or maybe it reduces conversions. And we want to see by how much, and like you said, historical data you can't measure month over month and see improvements, because it changes day to day in our digital landscape. So we have to constantly be evolving and changing and measuring those things, instead of just saying, oh, last month or last week, we had a 2% conversion rate, and I made this change, and this week we have a two and a half percent conversion rate. And that really isn't an apples to apples comparison, because, like you said, the traffic changes and all sorts of other different elements play a role into that?

Robert Plank 5:21 Well, great. So as far as your ideal client and things like that, is there, like a favorite kind of website that you enjoy working on? Do you enjoy a certain niche? Do you prefer e commerce versus direct response, sales, letters, anything like that?

Speaker 1 5:35 We kind of gravitate more towards e commerce, but we also do lead gen and software as a service as well. But I say majority of our clients are in the E commerce space of some fashion. Some are a little more direct response than others, but that's kind of where we typically fall.

Robert Plank 5:55 Okay, so in E commerce, is there a starting point that you like to look at first. Like, do you start with the heat map, or do you look at the navigation? Or like, how do you begin to attack that problem? If you have a client and they say, I want to convert more with the same traffic, or I want to make more money with this traffic, where does your thought process start? So

Speaker 1 6:16 we're going to look at the data first. We're always the data is not going to lie to us. So we're going to look at analytics. We're going to look at heat map data, if we have it available, and we're going to really just try to uncover where the key problem areas are. Maybe mobile has got majority of the traffic, but 60% of the revenue is actually coming in on desktop, so we might have a mobile engagement issue. So then we're going to look at the mobile experience and really try to identify some key problem areas where we want to start testing. And, you know, it could be forward to backward, where we actually start on the checkout page and work our way back. Or it could be that we even start on the homepage and push people down the pipeline. You know, the incremental increases all the way down throughout the pipeline, or it could be maybe we have an abandoned cart issue, or we've got other different issues. So each business is really different, but we're just looking for those kind of fall off points and those friction points that are causing the visitors to get hung up on the entire process.

Robert Plank 7:18 Okay, great. So you start with you look at that data, and do you see that typically, when a client comes to you, do they already have their Google Analytics and they're tracking and stuff? Is that already in place, usually, or are people missing that?

Speaker 1 7:31 Um, I would say the majority of the companies that we work with already have analytics in place. In some fashion, some are missing some key elements, but for the most part, they have it set up pretty well where it's tracking, you know, revenue and everything in in econ, in the actual e commerce platform, within analytics. But a lot of companies don't use heat maps. So we'll actually add that into the fold and just make sure tracking is already set up for but for the most part, most, most companies, have their their stuff together.

Robert Plank 8:00 So they're, for the most part, they have their stuff together, and a lot of them don't have the heat maps in place. And so I'm curious about, like, what do you think that there's some sort of a missing piece that these companies don't even know, that they don't know about when they come to you? Like, do they have the analytics in place, but they don't really know how to make sense of it? I'm just kind of kind of curious about, like, where is the hurt when these people first come to you?

Speaker 1 8:25 Well, the biggest part is people are so close to their their company and their marketing. Oftentimes, you know, they're in the trenches. They see it every day, and they often will overlook some of the small things that could be causing them to not get what it is out of their marketing that they need, or maybe click costs have increased on Facebook or something, and they're just left dumbfounded, and they need a fresh set of eyes on things is usually where it's at, and they know that they need to be constantly changing and constantly evolving With the ever changing digital landscape. So the big hurt there is the big fluctuations, and just not really knowing what they don't know about their visitors. So it's a little bit of an education process before we actually get started on an optimization plan, because there's a lot of moving pieces, and a lot of companies are just so close to it that they're overlooking some of those key areas of improvement that they might have overlooked because they're so close to it,

Robert Plank 9:29 that makes a lot of sense. I also imagine that they might be resistant to making some of the changes that they need to make, like maybe they spent a lot of time on a really perfect, beautiful landing page, but then some parts of it need to go and maybe they're too attached to those things they've spent so much time on.

Speaker 1 9:47 Yeah, we have a couple clients that have creative departments that they're really close to their design, and they think it's the best ever, and it's, it's, it's a little tough getting them to realize what we're doing. How we're going about this strategically, and we're measuring those changes, because at the end of the day, it doesn't matter what you or I think of the website. It matters what the visitors think, because they're the ones that are your customers or your potential customers, and we're just trying to better relate our marketing message to them in a way that gets them to take the desired actions that we want them to take.

Robert Plank 10:24 I agree completely. If I had to choose between having the best looking site ever that was making $0 or an ugly site that made tons of money, of course, I choose the ugly site. And so you mentioned that. So in one example, you looked at all the data, and you realize that the mobile traffic wasn't converting at what it was. So how do you come across something like that? It's just, is it just a matter of you've done this long enough where you have an eye for it, or do you hit it from different angles? Or are there specific things you're looking for? How do you reach those sort of conclusions?

Speaker 1 10:57 Well, we're just, we know that typically, that it'll be an engagement issue most of the time, or something is confusing to the visitors, but your analytics is going to tell you that. So just go look at the mobile view on your analytics and see, okay, you got revenue coming in from mobile and maybe desktops converting at two and a half percent. Well, mobile is only converting at point 6% but you have 70% of your traffic is on mobile, so you're spending a lot of money to acquire all these mobile visitors, but they're not converting nearly as much at nearly a higher rate as desktop. So you're making less money. So we're just really looking at where those key kind of friction points are, and that just points us in the direction that, hey, maybe we might have a mobile issue here, because the revenue swing is so different, and it's just kind of analyzing the numbers and figuring out, you know, where our kind of starting point is going to be

Robert Plank 11:53 okay. And you mentioned a few minutes ago that, so, yeah, you figure out some kind of an optimization plan, and what's the timeline like on all this like, is it? Is it ongoing? Is it take weeks or months to start making changes? What's the typical sort of turnaround?

Speaker 1 12:08 Well, optimization never stops, because the markets are constantly changing and evolving, and it's the only way to really get and keep a competitive competitive advantage is to constantly be testing to try to beat your high score. But a typical optimization plan from start to the first test live is anywhere from two to four weeks, once we're analyzing, come up with a game plan and really just dive in and see where we're going to start first and then building out that first test experiment. And depending on how complex we get with that first test. It could be anywhere from two to 10 days to develop a test. You know, if we're going to redesign a page or something, it might take a little longer,

Robert Plank 12:49 okay? And you mentioned that optimization is never done. There's always something new to work on. And so do your clients usually stay with you month after month? Or do they like to have you for a little while and then they're done. Or did they leave and come back? Or how does that usually work?

Speaker 1 13:05 The majority stick with us for a minimum of a year. And I mean, we've had clients for three years now. I think we have one that's been with us for almost that long. And, yeah, I mean, we it varies. Some businesses, they'll hire us and bring us in, maybe a larger organization they want us to do ingrain and optimization process into their company DNA. So we'll help them flush that out and help train, you know, the natural progression is to train their staff on kind of the process of it. So then they might choose to take it in house, or some just choose to maybe get the benefits out of it, and then they come back maybe, you know, two to six months later, once they've had time to digest. But the majority we just keep going. I mean, we're still getting wins for the company that we've been working with for three years. So it just never, never ends.

Robert Plank 14:03 Yeah, and I imagine if you double the company's money, or you increase their profits coming in by 10% Why wouldn't they want to keep hiring you again and again, right?

Speaker 1 14:13 Yeah, exactly.

Robert Plank 14:15 And so using you mentioned that sometimes it's fun to beat the high score to do some really cool stuff. Do you have any interesting stories like that where some company came to you and they were doing okay, but you guys really knocked it out of the park with some huge, big changes?

Speaker 1 14:32 Oh, let's see. We had one that we came in and they had a really high cost per acquisition. They were up around the $80 mark, so they weren't profitable on their recurring customers until about month five, we came in and did some tweaks and did some stuff, and in about 10 months, we grew their recurring monthly subscription, like customers from just about 6000 to two. 22,000 and we cut their cost per acquisition from $80 down to close to 10.

Speaker 2 15:06 Wow.

Speaker 1 15:07 And then another one, we tested for about two and a half months and didn't really see any significant improvements, and then came out and hit a 20 700% win, yeah. One test page redesign actually did that. So we've, you know, we'll hit the home runs, but they don't always happen. You know, most of the time it's some incremental improvements and really learning from that data and really building upon those use cases and really better understanding the visitors. And then we'll hit those 20, 3040, 50% wins.

Robert Plank 15:38 So how much of a time commitment to you and your team spend on a given client. Like you mentioned that you know you might be testing for months and months and nothing changes. I mean, is this the kind of thing where you guys have to tweak it every day, or you set it up and only check it once a week? Like, how involved does a person have to be when they're testing and tweaking?

Speaker 1 15:58 Pretty involved? So we'll, I mean, yeah, tests need to run. So sometimes we'll, depending on the size of the business, we'll actually set up a test and we want to run it if it shows promise for at least a calendar week. So, but in that time, we're looking at the data, you know, at least three times a week, and then we're maybe building another test in parallel with it, so we're maybe testing on a different point of the of the funnel, or a different point in the process, to where we get to the point where we're running two to five tests at any given time for a client, is what we kind of like to work towards.

Speaker 2 16:36 And

Robert Plank 16:37 that's a great point that you just made, that you don't have to just choose a test and then let it run, and then pick something else and let it run. If you're split testing different page or different parts of the funnel, those can work independently. And that seems like a real time saver.

Speaker 1 16:52 Yeah, you can. It's some places you can't. I mean, depending on what you're testing and where in the point in the process, but you know, you can do multi page experiments and have it track effectively and things like that, but sometimes just one test is all you can be running at any given time. But we try to maximize as much as we possibly can. But yeah, we're constantly in the trenches looking at data and looking at things, and we do bi weekly phone calls with our clients, and we're doing, you know, we have a reporting team that's always in there, you know, reporting, Monday, Wednesdays and Fridays and all the tests. So we're always up to date on what's going on,

Robert Plank 17:28 awesome, keeping your eye on the ball. Well, I don't really do I don't really do as much testing as I used to, and my business partner kind of does a lot of that. But when I was doing a lot of split testing with my own stuff, I noticed that sometimes I would just have my mind already made up. I was like, I'm gonna make this change in my web page. I don't even think I need to test it. I'm just gonna do it, and then I do the split test, and either it would go the opposite way, or once it ran for long enough, it would show that it didn't make a change. So do you have any funny or interesting stories like that where you just you decided ahead of time that something was going to work, but then you tested it and it went the opposite way or made no difference.

Speaker 1 18:11 So I actually use this in some of my speeches too, that we had a campaign running for a very large client, and we were running a video sales presentation for a promotion. We had a female voice over actor, a male voice over actor, and then I recorded one with all the ums and ahs and very imperfect and unpolished, and everybody at the company thought the female voice over actor was absolutely going to knock it out of the park, myself included. And we ran it, and that female voice over actor converted at half as well as the male voice over actor. And my ums and ahs imperfect video converted twice as well as a male voice over actor. So if we would have just assumed that the female voice over actor was going to win, we would have lost out on close to $357,000 over the course of promotion. Geez.

Robert Plank 18:59 So, so yeah, it's like, completely the opposite. You thought the hierarchy would be the female, the male, and then you but then it's completely turned upside down. It's, yeah, it's, you were the winner. Awesome.

Speaker 1 19:10 Yeah. I just, I've, I've learned this. I mean, we've conducted 1000s of marketing experiments, and we can give our best guests and our best estimate, and, you know, we're proven wrong all the time still, and it's just shows you this how much different it can be from business to business. You know what worked for company A, we could test the same exact type of strategy on Company B, and it won't have an improvement. So it's just really understanding what the visitors are doing and building your test plan around that, instead of just going out and assuming something is going to work because it worked for so and

Robert Plank 19:48 so. Well, great. So so people, you know, e commerce, businesses and online companies and people like that, and they need to have heat maps in place. They need to be split testing. Is there anything that you see? Out there that businesses are missing out on, or big mistakes that they're just making all over the place.

Speaker 1 20:06 Well, one is absolutely not testing. They're just assuming and going and changing things. I talked to five companies last week alone that I gave him some suggestions, and then came back two days later and be like, Yeah, I already made those changes. Like, hold on, you need to test those. The other thing is, people are shouting how awesome their business is, or all of look at how great these features are, and look at all of this great things we can do, instead of tying it back to the benefit. And what's in it, for me as a consumer of and what are the biggest benefits of actually consuming that product? And most people aren't leveraging benefits enough, particularly in E Comm, they're just seeing how great their product is. And look at how great this, you know, women's clothing line product is, and look at all those great material, and it's made in the USA, and all of this stuff, and they don't really tie it back to any type of benefit whatsoever, is probably the biggest mistake I see across the board.

Robert Plank 21:05 That makes perfect sense. And it also relates to, as you were saying near the beginning of our conversation, that a lot of these companies are just too close to seeing stuff. And because if they say, like, oh, we work so hard on this, we've spent so much money on this, no one's going to be impressed by that. So that makes perfect sense. And so we also mentioned near the beginning that you have this book called conversion fanatics. So can you tell us a little bit

Speaker 2 21:29 about that?

Speaker 1 21:30 So the book is a little over a year old now, but I wrote it to be very evergreen, in a sense that because the principles even over my whole entire 15 plus year marketing career. It the things haven't changed, just the mediums have changed. At the end of the day, we're all dealing with people. So what I wanted to do with the book is really educate people on the importance of optimization, whether you're a seasoned veteran or a complete newcomer, into the marketing world or into the testing world. And I just wanted to write something that was very tangible and actionable and not really filled with a bunch of fluff. Is really where I came about that. So it really walks the reader through the journey of testing and really the importance of optimization and why you don't want to assume, and a lot of things we're covering here, but in greater detail.

Robert Plank 22:23 Awesome. So, so, as you said, Evergreen. So no matter if you're listening to this, you know, 10 years from now, or if you're a beginner or an expert, something in it for everyone. So where's an easy place to grab that book?

Speaker 1 22:35 You can go to conversion fanatics.com/book or you can just hop on Amazon, search for my name, it'll pop up. Or just search for conversion fanatic on Amazon, it'll pop up there. Too

Robert Plank 22:46 awesome conversion fanatics.com/book or search conversion fanatic on Amazon. And then as far as you guys in your agency, where can people find out more about that?

Speaker 1 22:57 You can go to conversion fanatics.com that's the best place. And you can actually sign up right now, we're doing a special deal where you can get a free analysis. So we'll actually dive into your analytics and really better understand your visitors and give you kind of a game plan, normally, a $4,000 price point, but we're giving that away for free right now.

Robert Plank 23:19 Fantastic. And so just to make sure that you're not wasting their time, they're not wasting your time all that stuff. What's What does someone need to have in place before they go and go to your site and reach out and contact you? Or will you talk to anyone?

Speaker 1 23:35 No, they need to have traffic for one preferably, if you're paying for traffic, then usually, you know it's fine time to really understand more about your visitors. And if you've got consistent leads and sales coming in, you know, could be leads and or sales, but getting that people into the pipeline, and you know you're getting some action, you're not just a complete newbie. You've got something working. That's usually when we can step in and be the most benefit.

Robert Plank 24:07 Great. So you need to have traffic. Need to have paid traffic. Is preferred leads and sales, if they don't have Analytics set up yet. Is that a problem?

Speaker 1 24:17 Yeah. I mean, it's free. There should be any, no reason why you shouldn't have Analytics installed on your site, at least the bare bones, minimum, because it's free. I mean, just go over to Google and get that takes 10 minutes. So there shouldn't be any excuses why people don't have it installed. So they should definitely have

Speaker 2 24:36 that.

Robert Plank 24:37 Well, great advice. So if you have traffic and all that stuff, but you don't have the analytics and spend those 10 minutes and pop it in there and and correct me if I'm wrong, but the looking into all the raw data and all that that can all be done retroactively, right? Like, as long as they have that the Google code in there now, then six months from now, or 12 months from now, once enough is built. It's up, and they talk to you, even if they haven't even done any deep dive in the data, as long as that code is on the page, you can go back and retroactively look at stuff, right?

Speaker 1 25:10 Yeah, we can look back as far as since they installed that first, first snippet. I mean, it's it can go back as far as we need it

Robert Plank 25:19 to be awesome. Well, great. So, so no excuse left to get your traffic carried out, get it all consistent, get the Google snippet in there, and then once you're rocking and rolling, or if you're already rocking and rolling, and all those departments, then you need to reach out to conversion fanatics.com and get your testing dialed in. One thing is that even if you are super great at whatever you do in business. You might not be an expert in all this testing and tracking this conversion stuff, so why not go to a team that is. Another thing about all this is if it's just you trying to figure it out, and you're spread too thin, you might miss things, especially if you're super close to your business. So you need that outside authority to look, to look in and and talk to you about all this stuff. And so when people sign up for conversion fanatics and go with you guys, is there a price range usually, or is it on like a case by case basis?

Speaker 1 26:11 We're pretty flat rate. We used to be kind of a case by case basis, and really custom, and our solution is custom, but we went to a flat base pricing late last year. So yeah, it's in basically, you can hire our entire team for about the price of a or less than the price of a decent developer in house.

Robert Plank 26:33 Awesome, great. And also, I mean, once you make the money, then it's a no brainer to keep going, because if, if they find extra money that they would have missed otherwise by not testing. Then, of course, they'd want to keep going with that. So conversion fanatics.com and as you're winding down, Justin, is there, like any big, huge piece of advice? Mean, besides, need to be testing. But if you could shoot a message back in time to your younger self, you said you've been doing this for 15 years, if you could shoot a message back in time to 15 years ago, too young yourself. Does anything come to mind?

Speaker 1 27:04 Watch your numbers and don't give up on it, because there's other people that have been there before you, and that's the biggest thing is just because you don't get the home run slam dunk results right out of the gate, especially when you're starting or maybe you're just ramping things up. Don't give up on it, because it's all just a numbers game, and once you massage those numbers in, then it's time to really just pour the fuel on the fire. So that would be the biggest thing. I've given up on too many different ideas and concepts and products and marketing campaigns far too early in my career. So you just need to be patient and push forward, and good things will come out on the end.

Robert Plank 27:44 I love it. Don't give up before the miracle happens. So make sure, yep. So make sure to go to conversion fanatics.com everyone, and thanks so much Justin for stopping by.

Speaker 1 27:53 Thanks for having

Robert Plank 27:55 me. This is Robert plank, and would you like to be a guest on the program? Do you have an idea for a future episode. Or do you just want to tell us how we're doing? Get in touch with us right now at marketer of the day.com/ask, to let us know you.