GIỚI THIỆU VỀ TẬP NÀY
Show Notes:
Get Your Conversion Rate Optimization Starter Guide Courtesy of Disruptive Advertising.
Tell Us Your Ecommerce Pain Points
Check out our episode with Nick Disabato
Use Unbounce To Build Beautiful Landing Pages. (FOR FREE if you can be considered a startup)
Optimizly or Visual Website Optimizer or Adobe Target to create a better version of your current landing page
Transcripts:
Michael: Hello folks, and welcome to eCommerce Q&A, this is a podcast where store owners, directors of eCommerce, and eCommerce managers can stay up to date on the latest tools and tech in eCommerce. I'm your host Michael Bower, self proclaimed eCommerce junky and my guest today is Chris Dayley, the VP of testing inside optimization at Disruptive Advertising. I'm reading Chris's bio here. It says, "When he's not fixing his hair, you will most likely find him pushing the boundaries of AB testing. Disrupting the website design space, Chris frequently speaks on the topics of CRO and other things like that." Chris, welcome.
Chris: Thank you guys so much for having me on the show.
Michael: Absolutely. Tell me what the most disruptive thing you did this year was.
Chris: Well, it's actually, I'm not even going to give you an example in the website realm. The most disruptive thing I've done this year is, and I haven't even uploaded this to the site yet, but I started growing out my mustache. I decided this might be something that I need to try before I'm an old man, so I've got a long curly mustache going on and I just-
Michael: I've got to see this. You got to see this. Can you send a picture?
Chris: Just this week, I will say ... I will send you a picture. Just this week, I got a full curl in. It goes all the way around. It's a big accomplishment, it's pretty disruptive.
Michael: Nice, nice. You know I'm imagining a world where you can have a curly mustache and people not think you're a firefighter.
Chris: Or a painter.
Michael: Painter, okay. I don't know about the painter thing because what I've seen with painters, at least the painters that I know, they're all a little bit quirky because they've been drinking too much paint. What I mean is like ... A painter, to get those nice edges on his brush he'll stick his brush in his mouth to get it just the right moisture. At least this one painter I did a lot work with 20 years ago, he was a total weirdo, okay ... So mustaches. Well Disruptive Marketing, can you give us the big idea in a few words?
Chris: Sure.
Michael: Start with advertising and advertising marketing.
Chris: Yeah. There's a couple reasons I love the name Disruptive Advertising. Number one there are so many marketing agencies out there and there's a lot of marketing agencies that you can go out there to get an all-in-one solution. Lots of agencies that offer any kind of marketing service you want. We are just focused on one approach, and that is we are a data-driven marketing company that helps companies build advertising campaigns that convert and then convert that traffic on the website. So we take a very focused approach that is completely centered around conversions. So that's obviously a disruptive approach. That's one reason I really like it, obviously, our approach is different from everyone else in the space.
But also, we're also very disruptive in the sense that we don't use traditional tactics. We're not a rote agency that goes in and does the same thing with every client. We are absolutely an agency that looks for anyway of getting in front of your target audience. If that is using Facebook or using Google Ad words that's great. If that is re-targeting, if that's figuring out e-mail campaigns, if that is running tests on your website. We dive in and we look for ways that we can learn what is actually going to convert each of our clients' audiences. That is also a disruptive approach. I am a huge fan of adapting ... We're in such a rapidly changing industry that you've got to be able to adapt and think on the fly and be strategic and that's our core competency.
Michael: That's great, thanks for that. My understanding is that you mostly focus on lead gen, is that right?
Chris: No, actually the majority of our clients and the space that we do the best in is eCommerce. I have a lot of background in legion and that may be what you're referring to. But I would say over half of our client base is eCommerce.
Michael: You know what I'm seeing and what we're pushing right now for a lot our clients is use these strategies and tactics of info marketing in lead generation that have been proven over time, over the last multiple, honestly more than a dozen years and apply those to your eCommerce site. The main thing that I see when we're talking about info marketing or lead gen is that you have a single page that lays out a case if it's a persuasive argument that you're doing. A lot of us in eCommerce think that if we just show you our stuff that that will make you buy it, which is not necessarily true if everybody else is doing that same thing. You have to be persuasive, you have to give you a reason to buy this thing. Or if you're earlier in the customer journey, let you know why you need to be considering this type of thing in the first place.
We're going through this with a client right now, it's an ebike company, which is in interesting space, being a biker myself I'm in to that and actually have been considering buying an ebike, which is freakishly expensive right now. Which is the point, right? When you have a ... When you're not competing on the basis of price, you're in, maybe a better industry or a newer industry or something like that. There's a lot of people that are going to be competing on the basis of a lot of other attributes other than just the price. That's where all these strategies and tactics can be really valuable.
Chris: Well and what you said that I loved that's a big big part of what we do, you're talking about value proposition. And for me, someone's motivation to convert on your site is a simple, simple equation that I didn't make up. But someone's motivation to convert is equal to their perceived value minus the perceived costs. You have to build up enough value for your user that they think that that value is greater than the cost of either the dollar amount cost or the time cost it's going to take or the cost of potentially getting blasted with emails and texts from you. There's a lot of perceived costs that's in someone's mind and so almost immediately when someone comes to your site, you're working at a disadvantage. You're working with a debt that you need to repay them in the form of value. You need to show them that value. So I love what you said there.
Michael: How are we working at a disadvantage? Can you elaborate on that?
Chris: Yeah, so people in general, especially people that are web-savvy, which is everybody, is people are so familiar with the standard tactics. People are very familiar with the things that they encounter on websites everyday. So there's a lot of things that people, when they come to your website are immediately going to tune out. They're going to tune out probably 80% of your website experience. So you have to make sure that the things that you are really drawing attention to, the things that stand out to your audience, you have to make sure that those things within two to three seconds can communicate a tremendous amount of value. Because that's about all you have. You have about two to three seconds when someone comes to your website for them to psychologically draw context from your site and decide if they want to stay.
Michael: Let's dig into this some more if you don't mind.
Chris: Yeah.
Michael: The whole idea of getting value within two to three seconds, let's say we're selling a product and that product's a physical product and I can't get that product to you within two to three seconds. So what do you mean exactly in an eCommerce site let's say?
Chris: Ecommerce is probably the very best place and sometimes the hardest place to do this because we want to tell them so much about our product, right? It doesn't matter what the product is, it could be, I have optimized pages for doctors that are buying hereditary cancer testing kits. Extremely detailed, lots of components to this package and very scientific. There's lots of information that we could just barf all over these users. That's what we as marketers and we as business owners are really tempted to do, is just give them a ton of information. Whether it's doctors that you're advertising to or kids or mothers or whatever it is, you need to be able to show the value of the product in two to three seconds.
There's two things that need to happen. Number one, you need to make it seem very easy to take action. It needs to be obvious what the user is supposed to do. The very best way to make it obvious what they're supposed to do is to use color contrast to your advantage. My eye needs to be drawn to the call to action immediately. Because psychologically if I feel like I have to figure out something. If I feel like I have to figure out what I'm supposed to do, if I feel like I have to scroll down your page or read a bunch of stuff, that immediately causes anxiety for me. That's never good for conversion rate. So you want it to be very obvious, what they're supposed to do.
And then you want it to be very easy for them to figure out why they should buy your product. There shouldn't be a ton of information that you just spew all over them right away. You can have a lot of information on site. You can have a lot of information on an eCommerce product but you don't typically want to barrage them with all that information immediately. You probably want to have two to three things that are going to stand out to them, above the fold so that they can go, okay, there's a little bit of information here. If I want more I can scroll down. But it needs to seem psychologically easy for someone. If they can figure out what they're supposed to do and they can figure out why they should buy your product and they can at least just draw some quick context from the page you have a much much higher chance of converting them.
Michael: Let me make sure I've got this straight. There's two things that you mentioned are crucial for making a conversion friendly product detail page, which is first making it super obvious what you want them to do. Putting it right in front of them. And second making it really obvious why they should do it, is that right?
Chris: Yeah. What I see happen a lot with eCommerce sites, here's an example, I'm actually going to tell you about one of our eCommerce clients that sells a bunch of products. This client is Diesel Power Gear, they have a show on the discovery channel. So they get lots of traffic to their site and their monetizing on this show with a bunch of knick knacks. Like hats and beard oils and t-shirts, whatever. He's got a bunch of stuff that you could buy from them. When you go to their homepage, they were like, "Hey we've got all these cool products, there's tons of different types of people that are coming to our homepage so we just want to show them as many different kinds of products as possible". So they had this very long scrolling page, you could just scroll and scroll and scroll. They're like, "It's kind of like Instagram, people love Instagram". So they thought that this was the ultimate homepage.
When we went in, one of the very first tests that we ran that I am a huge proponent of and it's called an existence test. Existence tests are basically, we want to figure out what content should be on the site. So they have all these products that are one their homepage, for an existence test we go through and we just create a bunch of different variations of the homepage where we remove some of those products. So we had eight or ten different variations where we removed one or two or four products at a time. And we launched that test and by the end of the first week we had generated an additional 28,000 dollars in revenue. Just from removing stuff from the site.
Michael: I'm not going to ask you to share real numbers but can you give us an idea of roughly how much revenue this site's doing? Is it ...
Chris: It fluctuates from one to three hundred thousand dollars a week.
Michael: Okay. So that's a roughly, almost a 10% or even more actually, it's 10% plus ...
Chris: Yes.
Michael: Actually that's 10-30% increase.
Chris: Yeah, just from removing stuff. And you have to remember too that we were splitting traffic to some of these variations. Once we implemented one of the winning variations it represented, I believe something like 40 to 50 thousand dollars in increased monthly revenue.
Michael: Just from removing stuff off the site? Not changing any of the messaging, wow ...
Chris: Exactly. That really speaks to this point of when you give people too much stuff, it makes it really hard for people to make it a decision. When you make it simple, when you make it easy, when you help them and guide their attention where it should go, you have a much higher chance of actually converting them and them buying something.
Michael: This is great. Honestly this ties in, Chris, really closely with a series that we just recently did around timeless UX principles. This idea of taking things out, figuring out what really matters, focusing on that, pushing that, that was one of the main things we talked about. I'm glad to hear that this is further confirmation of that. Chris I want to change gears slightly and talk about you guys, was it you personally or your company won the 2014 Gold Witch Test 1? Was that you?
Chris: That was my company back ... 2014 I started my agency, Daily Conversion, this is prior to Disruptive.
Michael: Got it. So can you tell me about that? What exactly did you do, how did you win that, what did you learn from it?
Chris: This was a really fun test. The reason I submitted it was because it was a test that I think a lot of people might have rolled their eyes at when they saw what we wanted to test. And it had a tremendous impact. What we ended up testing was we had a landing page that we had optimized a lot around. We had been testing on this landing page for over a year.
Michael: What kind of company was this, what was being sold?
Chris: It was for, this was a lead generation, it was for universities. So getting people to go back to school. So we had a landing page with a giant form, a ridiculously giant form with 20-something form fields on it. Like I said, we had been testing on this for years so we had figured we had a best practice landing page here. We've been testing on this for so long, we've seen so many improvements, we've learned so much about the audience, we've got it pretty dialed in.
But I decided to shake things up a little bit so I ran a test, I just called it the color test. That's what we called it. Literally all we tested, is we tested 12 different variations of colors on this page. So we tested the background color, the value proposition colors and the button color. So different color combinations. And we used any color we could come up. So we had purple backgrounds with orange buttons. We had orange backgrounds with purple buttons. We had blue backgrounds with ... Red backgrounds ... We had green, any color you could think of we tested in this test. At the end of this test, we ran this test for ... We had, I think, close to 500,000 visitors that we ran through this test. So a significant amount of data. What we did was, after we finished the test we went in and we segmented the results by traffic source. We wanted to see, okay, do different traffic sources respond to colors differently?
What we found is that each traffic source had a different winner. So our email traffic had one color combination winner-
Michael: What was it? I got to hear this. What was the color combination that email liked?
Chris: This was so interesting. Email liked purple background with orange buttons. Like Phoenix Suns. It was the most ugly, when our designer saw that page, our designer actually told me I shouldn't even run that as a variation of the test. Because he said it looked so bad there's no way this is going to win. Not only did it win, but it had over a hundred percent increase in conversion rate.
Michael: Okay, so that's email.
Chris: Email for organic, I believe it was a blue background with an orange button. Then we had our, I guess what we called affiliate traffic, its re-marketing and then affiliate advertisers. I believe on that one it was the flip flop of the email. We had an orange background with purple call to action buttons. What came to me out of this test, was something that was very interesting. Number one, color obviously played a huge role in how people responded to that page. And different types of audiences respond differently to different colors. We had just assumed, well we'll just run the same variation for all traffic. From this point on, we learned our traffic sources respond very differently to our test variations, so we need to test them all separately. We need to have a customized landing experience for each of these traffic sources. It was very very interesting and some of the craziest results I've ever seen.
Michael: That's so weird. We just did a call the other day with Nick Disabato, whom you probably know from the industry. We had a very interesting conversation about the whole dynamic tension between activities that basically make your site look worse but convert better. There is a balance to be struck just to be clear for anyone whose new to this show, you don't want to only go off of what tests are better in the short-term, because that's like all the low-end fast food restaurants that just start randomly adding things to their menu until it's so cluttered you can't even figure out what's going on. Basically you lose your brand eventually. So you've got to watch out for that. But at the end of the day you have to balance those two things because you need to sell stuff. You need to sign up leads. This is a transactional scenario here, it's not an art gallery. Dynamic tension. Wow, that's really interesting.
One of the things that we're running into is when you start getting more targeted you get a much better conversion rate. Astronomically better, right? Every single factor by which you segment in an audience and then market to that makes it way better. What's a modern tool that you guys can maybe recommend for managing the, all the different versions of everything that you have to have? Landing pages, let's start with that.
Chris: A couple things on that note. First, I always recommend testing before you start personalizing all of your landing pages. Because I have also run very similar tests in different industries where all the traffic responded the same. Or all the traffic didn't really seem to care. I wouldn't get too complicated before you know that you nee to. But if you've tested and you find that these different audiences respond differently ... I can't even tell you, the idea of personalization gets a lot of business owners really excited. They're like, "okay yeah we can have a campaign for, we can have a landing page for women, we can have a landing page for men, we can have a landing page for Asians". This idea sounds really exciting and it also is a ton of work. And it only makes sense if the audience cares, right? If the audiences actually perform differently.
But assuming that you have tested and these different audiences actually want a customized experience, there's a variety of things. We used, for landing pages, we use Unbounce. Unbounce is a great way, very easy tool to set up landing pages. You can copy and duplicate landing pages, create variations. It's a really really easy way to manage tons of different landing pages. And it's very very simple. It has a drag and drop editor so you don't need a developer to go in and develop these landing pages. So highly recommend a tool like Unbounce.
If you're testing an actual page on your site, so not a landing page, you've got to have a testing tool in place. So something like Optimizly or Visual Website Optimizer or Adobe Target. You've got to use something and these testing tools ... What we do for a lot of our clients when we've broken out landing pages like this and we have four or five different audiences with different site experiences, we won't even make the changes on the actual site. Because it can get complicated if you've got five different versions of a page that your developer has to code up based on traffic source. A lot of time we'll just have one version of that page and then we'll use that testing tool to pull out those specific audiences and customize the experience for them. So that way you don't have this nightmare of a back end on your website that you're developer cries every time you need to make a change to. He can just manage one version and then you use the testing tool to manage all the rest.
Michael: Absolutely. We're actually seeing the emergence now, this is a bleeding edge area, but of tooling that will not only do what you said but it will do it dynamically from a data source. It will be like personalizing your page based on data that we are pulling from our [inaudible 00:21:06] or whatever it's pulling. So that's going to get really really fancy in the next, I would say, year where we'll be able to do dynamic personalization without having to do a ton of manual work.
Chris: Yep, personalization on the fly.
Michael: Cool, that's interesting. I don't want to spend too much time talking about the tech here. I want to here more about the disruptive side of your business. What are some other really disruptive things you guys have done?
Chris: Well, in the testing space pretty much anything you do is going to be disruptive. Especially because most of the time you're dealing with a business owner that has very firm opinions on what is going to work and what isn't.
Michael: Mm-hmm (affirmative). Because they started this thing and it's their baby and they know this industry. Maybe they're part of the market, that kind of thing.
Chris: Oh yeah. In fact, I'll give you one example. Just the other day, we have a fashion retailer, she's got a lot store boutiques and then now she's launched an online website to sell her clothes. So she's very very very involved in the fashion industry. She flies out to Europe every year, multiple times per year to figure out all the latest fashions to make sure that she's on top of the latest trends. She's sources out all of the materials herself, so she's very involved in this space. Because she's very involved in this space and she's very on top of the trends, she was positive she knew what her audience wanted.
One of the things that we proposed to her ... One of the things that we've seen work very well on eCommerce sites, if you're looking for a test that you should run, it's not the easiest test to build but it can have tremendous impact, is having some kind of a pop-up experience when you add something to your cart. Going back to this idea of making it very easy for your customers, you want it to be very obvious what the next step of the process is. On this particular website, if I clicked add to cart, I added a shirt to my cart, for example, I would just get a little notification that says, "successfully added to your cart". And now it's great. We recommended to her, we've seen this work on other sites, "instead of that we should have a pop-up that comes up and says, you added this to your cart, now let's go checkout and buy this", and she said "no that will never work". So we said, "okay why will it not work"? And she said, "I tried this three or four years ago, my audience just doesn't like pop-ups, plus nobody else in my space is doing pop-ups. If it worked, someone else would be doing it."
We said, "well let's go ahead and let's run it anyways". So we ran three or four different variations of pop-ups with different designs, different content, different information, different emphasis on checking out versus continuing to shop. Anyways, by the end of the test all of the variations that we ran were beating her existing website. The one that ended up winning, even though she wanted to push people to continue shopping, the one that ended up generation by far the most revenue was the one that just had, in your face a big red button that says "proceed to checkout". And then had a little tiny link that said "continue shopping". So making it very obvious, you just added something to your cart, let's get you to checkout now and actually buy it.
Again, this was very surprising to her. It went one hundred percent against what she believed her audience wanted. And went against what her industry, quote/un-quote, best practices were. But it totally worked for her audience. Again this whole idea of being disruptive is all about challenging assumptions. You can't assume, just because no one else in your industry is doing it, that it doesn't work. Chances are, no one else in your industry is even testing. They're probably just launching a new website every six months and everyone thinks that what they're doing is working. So everyone else is copying them. You've got to test what works for your audience. You've got to be willing to challenge those assumptions and be disruptive in your industry.
Michael: That's really interesting. Again I want to refer back to the fact that it tested well in the short-term, do you think that's going to have a negative impact on the brand in the long-term? Is there a way of testing that?
Chris: That's a really interesting concept that you bring up. This whole idea of branding is one that's near and dear to my heart. I'm very passionate about, obviously I was a business owner myself. I do care about brand. But to me, I think a lot of brands get way too caught up in the idea of their brand. To me, whether or not you're pushing someone to check out or giving them the option to continue shopping, your customers most likely are not going to remember that long-term. And if it's working on your site, if it's generating consistently more sales ... If you're really worried about it hurting your brand long-term, one thing I've done in the past is run what I call a perpetual control where you always have a very small percentage of your traffic that's going to a version of your website before you started testing. And then you've got your latest and greatest version of your website where you have 95% of your traffic going.
So you can track and make sure that these improvements you're seeing actually play out long-term. But for most of our clients, we see these continue to play out long-term. And that idea of brand being so concerned about what your audience thinks about your brand, if they like your site experience you've got a good brand. It doesn't matter whether you're using pink or purple or red or orange on your site to your audience. They don't really care what your site looks like and feels like. They really just care, is it easy to use? That's one thing that Amazon has found. And Amazon, I think, is the king of, is they're constantly tweaking their site. It has nothing to do with, "we want people to think that we are the most sophisticated website in the world. It's really just about, how do we sell the most damn product in the world?" And everyone loves Amazon because it's so easy to buy crap. It's almost too easy.
Michael: You know what's interesting, what this is reminding me of, to be honest with you, I don't know if I fully agree. But that is because I haven't done the testing that you have done. And I feel anxious about this idea of making changes that may feel so dangerous or whatever. I'm feeling really challenged in my assumptions about that and I appreciate that challenge. What I was thinking about just a moment ago was that the concept of, in genetics, in biology, I'm not a biologist ... But the idea that a genetic population, when it's exposed to a certain environment will change and morph. The idea of hopeful mutations basically, that genetic traits expressing themselves in a particular way within an environment will allow the population to survive better in that environment. I think maybe we can analogize from the idea that an audience is like an environment and that your website is like an organism or maybe a specific subpopulation of that genetic community that then can find it's safest and best expression. And it's going to be a little bit different depending on the environment, depending on strengths of the company and so on.
What I'm getting at is you have lizards that develop larger pads on their feet to be able to grip things better. But then in another scenario it's actually gets smaller pads so they don't have as much surface area with the sand so they don't burn their feet or whatever. I think that we can basically, maybe take that idea and just thinking out loud here, but apply that to changing ... When we think of testing and changing our websites and really any of our digital properties, or any of our marketing, any of our messaging, anything about the company that's customer-facing, maybe we can be a little bit bolder and see what's going to stick and be more straightforward. So that's a challenging thought for me.
Chris: And just going along with that, it's important that we not get so tied to what has worked in the past that we are unwilling to try new things that may work and let the audience surprise us. Because I will tell you this, I run hundreds of tests every month right now for a variety of different clients in every industry you can imagine. I am surprised every day by a test result. Something that I didn't think would work that did or something I thought would work didn't. You've got to give your audience room to surprise you.
Michael: That's great. Well that's a great place to stop and Chris we so appreciate your input and just your enthusiasm on the show today. Before we go can you let us know how folks can follow up and learn more from you and your company and maybe work with you guys?
Chris: Yeah, you bet. I'd love people to reach out if they have any questions. I'm on Twitter, @chrisdayley. My last name is D-A-Y-L-E-Y. We have, we've put together actually a free starter guide if anyone wants to start testing and they don't know exactly what to do or what tools to use. So if people want to check out our starter guide, they can go to disruptiveadvertising.com/guide. If they want to work with us there's a little check box to click. If you don't want to be contact by us just don't check that box and we won't pester you. That's a great resource if you want to get started.
Michael: Excellent. And we'll leave it there. Thank you again Chris. And everyone, start testing. Start breaking things. If you're afraid of what's going to happen, maybe work with an expert like Chris and his team. Let's improve our websites in some really amazing ways.
