Unlock this video + Thousands more !

Start your free 30 day trial today!

Join Free Sign in

Conversion Rate Optimization: How to Unlock Your Website's Potential

Video Summary

With 95% of conversion decisions happening in the subconscious brain, motivating action and improving conversion rates is challenging, to say the least. The key is to learn what you’re audience wants and needs using big data and advanced analytics, that helps to surface the right content and shape user experience in a better way. In this session, you’ll learn a secret 6 step process to identify the major boosters and blocker that are likely helping or hurting your messaging and user experience, surfaced from dozens of experiments in different industries and niches. Most importantly, you’ll learn how to remove the guesswork for conversion rate optimization with proven tactics that deliver results for your hard work and marketing investment.

Speakers

I live and breathe internet marketing. Conversion rate testing and optimization is my passion and area of expertise. My company has created innovative testing programs for businesses, with a tried and proven process for evaluating and executing high-impact tests.

Also an avid CRO speaker, having spoken at many conferences, events, and corporate trainings, including:
Social Media Marketing World 2018
Content Marketing Conference 2018
Social Media Marketing World 2017
State of Search 2016
Searchlove Boston 2016
Mozcon 2015
StartFest 2015
SLCSEM DMC 2015

What you will learn?

  • Why testing your site is important?
  • The six things that will help you optimize your website
  • How mobile users consume content differently from desktop users

[00:00:00]

Okay. Well, we’re going to go ahead and get started. Thank you guys all for coming today. How many of you guys have had fun at content marketing conference so far? Show of hands. How many guys have learned something today that you plan on applying to your business? Some of you? Okay, and if you have too many things to apply you don’t think you’re actually going to remember.

[00:00:29] Well, hopefully today I can give you guys a few actionable items that you can take home and apply and hopefully we’ll keep it simple enough that we can all remember these things as we go home and potentially even apply as you’re sitting here in the session. So that’s my goal. And so what I’d like to do, I’m just going to jump right into a test that I ran and have everybody guess which one was the winner. So basically we’re looking at two different designs. That’s really the major changes. There’s a cup of coffee at its but it’s mostly design changes. So by a show of hands how many of you guys like the version on the left better the Green Design?

[00:01:11] Literally, no one a couple people that are probably just trying to be brave. How about the one on the right the Blue version? Okay. It’s pretty unanimous and I’m with all of you guys. With the one on the right, it looks better, you know from a design standpoint. In fact, when we proposed this test, our designer forbid us from running it because the Green version just looked so awful.

[00:01:33] But when we actually ran the test the Green version won by almost 14%, which was fascinating. It blows my mind is one of the reasons I love doing A/B Testing is because I’m constantly surprised and this brings up a major problem, which is, it is impossible to predict what is going to work best on our website.

[00:01:56] Especially when we’re talking about design elements, you can go through all the committees in the world, you can talk to testing experts, you can talk to UX Designers, but nobody can predict what our audience is actually going to like better. So there’s obvious justification for testing but it goes a step beyond that, which is well, how do we know what to focus on when we’re testing. So that’s what I hope to be able to answer for you guys today and to give you guys a few kind of food for thought elements to look at on your website.

[00:02:29] So in terms of my background, I’m from Utah, anybody here from, Utah? Couple people. Yes, thank you. This is my beautiful family. I’ve got a wife and two kids and my background my very first full-time job was in search engine optimization. I spent two to three years doing search engine optimization and some social media marketing and I had a huge problem, which was we were getting a ton of people to our website.

[00:02:58] And nobody was actually converting. Nobody wants that problem. I’m sure there’s a lot of SEOs here in the room that have had that problem. And what sucks even more is when you are not getting people to convert and nobody knows why. Right, that is very frustrating. And so I’ve spent the better part of the last six years, creating a process of analyzing landing pages or websites that I would like to share with you guys that’s been instrumental in helping me create a testing process that consistently finds winning experiences. And so there’s going to be some overlap for from those of you who went to Justin and Taylor’s session.

[00:03:39] I think I’m going to be reiterating and diving into some of the points that were made earlier. And hopefully I’ll give you guys kind of a holistic picture of your website. But in terms of when you’re looking at what should I test on my website? What should I test on my landing page? These are the six areas that I’ve found typically have the biggest impact on conversion rates.

[00:04:00] And we’re going to dig into each one of these specifically. I’m going to show you guys a lot of case studies today to hopefully give you some ideas of things that you can try on your site, but the six areas that we’re going to that we’re going to look at are your value propositions, your call-to-action your content. We’re going to talk about any diversions that may exist on your website. We’re going to talk about things that might cause anxiety. And then we’re going to talk we’re going to dig into not just responsiveness, but how to create a great mobile experience. So I’ve got a lot of slides and I probably have more than I’ll be able to get through.

[00:04:33] Value Proposition

So I’m going to dive in here to the first one which is value proposition. So, I’m sure that if I asked any of you guys what your value propositions were, you could list off 10 or 15 or 20 of them for me. It’s not hard to identify what our value proposition is. It’s what am I offering? What do we have? Specifically we want to look at what’s unique that we’re offering?

[00:04:58] And, beyond that there’s a few questions I want you guys to ask which is how many value propositions should we display on a page?  And which value propositions will resonate best? When I asked a lot of business owners or marketing people those second two questions, that’s where we kind of fall apart, right we can tell what is unique about our business what’s unique about our products, but when it comes down to well, how many do people actually want to see if I have free shipping and I have non-GMO products and their cruelty free and they are cheaper than my competitor.

[00:05:37] What are they ultimately going to care about? Do they just care about the fact that they’re cheap? Do they care about the fact that they have free shipping or they have a hundred percent guarantee? You know free returns all that kind of stuff. What do people really care about? So these are major, major things that you need to understand when you’re creating a website because most of us when we are creating a website, we just throw all of them up there and hope that something sticks.

[00:06:01] So, I’m going to show you guys a fairly simple test that everyone should run on their website and the way I like to think about just general conversion, is somebody’s motivation to convert is equal to their perceived benefits minus the perceived costs. Now, there’s a lot of costs that you guys don’t even know about when somebody comes to your website. There is obviously whatever the prices of something that they’re buying from you or the cost of giving you their information on a legion form. There’s also the time that it’s going to take there’s also the potential hassle of getting called by your people and emailed. There’s all kinds of perceived costs, and so the point of your value proposition should be to increase the perceived benefits. Right, help them understand what the benefit is to them.

[00:06:52] So, here’s a simple test that I think everyone should run on their website and as a quick side note, every test you run on your website or landing page or email or adwords ad should have a business question associated with it. And I like to use open ended questions like this, “What value propositions will resonate best with our audience?” What this does is it helps me approach my test from different angles and say well I have one hypothesis that these value propositions will resonate best but let me think about some other ways and you know what even if our original experience wins, we still have answered this question. So it’s important that you set up your test right from the beginning but this is a test that is fairly simple to set up and I’ll show you how we ran this for one client.

[00:07:44] So, this is Penny Mac and I have pointed an arrow here to the three value propositions that we’re going to be testing. Okay. This is literally the homepage for some of their traffic and the value propositions that they have on there were just picked by whoever made this page. So literally as I started asking questions about how did you decide what value propositions to put on there? They said, I don’t know what else would we put on there? But they have a ton of things that they do. This is basically just some really generic things here. So we wanted to test a couple different approaches at these value propositions. So we literally just changed out the icon and the verbage. We had a version that focused on, we’re only changing the bottom value propositions here, by the way.

[00:08:30] So the original experience had more than 1 million happy customers conventional FHA VA USDA home loans. So we changed that in the second version to 65 billion dollars in loans funded in 2016 and the fourth largest lender in the country. These are like, you know, some credibility statistics, right. Then we had another version where we are focusing on maybe some potential pain points unparalleled customer service and a fast easy quote process. Again fairly simple tests right? All you do is swap out the text, swap out the icons. You can build this test in under an hour. This ended up having about a 20% increase in conversions. This is bottom line of the funnel conversions, people that actually clicked on the button, filled out the form and have submitted their information. Mind blowing right? Just from changing a couple of things on the site.

[00:09:25] Now, I’m not saying that every test you run is going to have an impact like this. But what we found from this test was these value propositions really matter to the audience and by focusing on different types of value propositions were able to answer a lot of questions that people have and potentially resolve some concerns that they have without making them dig deeper into our site.

[00:09:52] I’m going to I have a slide here to try it out and grade your value proposition. I’m not going to pause too long here. But typically the questions that I want to have you guys ask are, “Are our value propositions easy to identify?” “Could they be communicated in a better way and do you know with data, which value propositions you should be focusing on?” Because as soon as you get that data as soon as we find an answer to that question.

[00:10:15] Then we can run all sorts of follow-up tests on that. Should we have an image that helps illustrate some of these value propositions over here? Where should we highlight those value propositions? How much attention should we draw to them? There’s all kinds of follow-up questions that will want to ask after that original test.

[00:10:32] Call to Action

But that gives you just a couple of things to consider when you’re looking at your value propositions. So value propositions are good to establish the benefit of our product. But our call to action is arguably the most important thing on our page because that’s the thing we want people to do right and so a few things to think about with your call to action is, “Am I telling people what they should do?” This is like one of the most missed things on websites today. A lot of call to actions are really passive in nature. Like learn more. I don’t really exactly know what I’m supposed to do. Right? I have a form that I want someone to fill out and at the button says submit and the thing at the top says, may not even say anything.

[00:11:18] I might just have a form of the button that says submit. We want to tell people fill out the form and click this button to get your free quote or whatever. You know, it’s a super long call to action, but you guys get the gist. You want to tell people what to do and so a couple of keys and Taylor just touched on this earlier a couple tricks that you can use to your advantage.

[00:11:39] So the two second rule I think about it in terms of human psychology. So people have done all kinds of experiments on how much attention people give to a new website. So if somebody comes to your website and they have never seen it before they will typically give you about two seconds to figure out what the heck is going on.

[00:11:58] They’re going to look at the page and they’re just going to do a quick assessment and there’s hundreds of things that are going through people’s subconscious brains right now, and during that two seconds they need to be able to identify your call to action. Because what that does is it reassures us.

[00:12:14] It reassures us emotionally to go, “Oh, I know what I’m supposed to do.” Okay, that’s one less thing I have to worry about. I don’t need to feel stupid because I actually know what to do. I mean have any of you guys ever been to a website before and you get there and you’re not exactly sure what’s going on.

[00:12:28] Like it feels like you’re in the wrong place. Right? Like I thought I was going to be able to sign up for this thing, but I don’t even see a form, that’s frustrating it causes anxiety. So you want people to be able to identify that in two seconds and one great way of doing that is using color contrast.

[00:12:47] Okay, making sure that and again Taylor and Justin earlier both touched on this, making sure that your call-to-action really stands out. I think I’ve been labored that enough. Let me show you a couple of tests that we ran to illustrate this. So here is one question that is great to ask, and you can ask this for any page what call to action is the best?

[00:13:09] Now I see this this is a major challenge that people have is what should our homepage be especially if we have multiple different products or services, what call to action should we be giving people? Sometimes we show them every single product or every single call to action that we possibly have. Sometimes we just picked something or sometimes were like pointing them to a blog post, right? So you want to figure out what should we be showing people? So here’s a test that we ran for a company called Digit Cert they do digital security.

[00:13:40] This was the original homepage that had a single call to action on it, and it was to take people to a comparison page, to compare all the different certificates. Now if you scroll down on the page, there were other call to actions down below the fold, but again we’re talking about, in the first two seconds, can I identify what I’m going to do.

[00:13:58] So we tested a couple of different versions against this. We wanted to see should there be a single call to action, or should we show our three most popular products, our three most popular packages? So we had one version where we had split people’s attention between three different call to actions.

[00:14:15] Then we ran another version where we kept the original call to action and then also added in the three additional top packages. By running all three of these, this can help us understand how many different options should we give people on the homepage? Now, you would think that just having one would be the best because it’s the simplest right?

[00:14:38] But here we actually found that having the most call to actions out of all of our variations here literally doubled revenue. It was insane, right and then you know the obvious questions that come after that. At least for me are well, how many should we put on there if this if this doubled Revenue we should try 15 call actions on there right?

[00:15:01] But so those again are the kinds of things you want to think through and when you get a test result like this. You want to dig deeper into what did we find here? Why did this impact revenue so substantially? But these are questions that again, you’re going to want to ask on the homepage, you’re going to want to ask on your landing pages. Am I giving people the best call to action or the best options of call to action?

[00:15:28] Another test that I think is just really fun to run is the best way to feature a call to action. This is a client that’s in the vacation packages space. This is the search results page after you’ve told them where you want to go, or the lack thereof. Here, we haven’t told them anything. But this is all of the potential vacation destinations I could go to. Now if I was to ask you guys what the call to action is here, can you identify that quickly and in two seconds?

[00:16:06] No, I didn’t see anybody. Everyone’s taking their heads right? Now, what this company actually wants people to do, is to call them.  Right, and so that’s not very obvious. If I was to look at this page, I’m thinking okay I need to click on one of these search results here. I need to get more information.

[00:16:25] But if the best course of action is actually to call in. We need really need to draw attention there. So we tested a variation where we drew a ton of attention over to over to that call section. We added we did change some text as well. And we had some other variations in there where we did not have the text changes and they also performed well, but we ended up with an 83% increase in phone calls, and I might add and I don’t have this in here zero impact to online sales.

[00:16:58] Online sales, I think actually increased surprisingly. So what happened here is what we drew a ton of attention. It was obvious that I should call in and we probably saved a lot of people that were just bailing from the site that were really frustrated because there’s a shish ton of information on this page and I don’t want to look at all that right.

[00:17:20] So, again, think about. How should I be featuring my call to action? Am I drawing enough attention to it? Am I drawing enough attention to the right call to action? There’s all kinds of things that I could go through there, but I’m not going to spend much more time on that. So A few questions to consider is your call to action easy to identify? Could it stand out more? Do I know what the best call to action is for the page? Those are some questions to consider with your own call to action.

[00:17:50] Content

So content. Hello Content Marketing Conference. This is a big deal right? There’s a lot of things on your website that are content. Your videos are content. Your content is content. Sometimes your pictures or other things that you have on your site is considered content. I would say the vast majority of your website is content. So this is a really critical element to think about and so a couple things to consider here.  Is my content relevant? Some of the previous presenters, you know covered this topic.

[00:18:23] Am I able to even understand what you guys do that also ties in with your value propositions, but how much content should I have? Where should my content be? Is it readable on all devices? That should go without saying but I see this all the time where I’ll pull up a website on my phone and their banners shrinks down and they didn’t size it properly for mobile, so I can’t even read it.

[00:18:45] So stupid so that should not be an issue if it is then fix it, but how much content should I have and where should that content be? These are really interesting questions because we often assume that people want all of our content from the very first time they hit our website.  So we’ll look at a test question here how much content and in what format does my audience want?

[00:19:10] Okay, so we’re taught we’re touching on two different things here that we can address with one test. This is for social media examiner. I don’t know if any of you guys read their blog. I met Mike Stelzner three years ago or so and we’ve been working together since this is one of their popups on their website to get email addresses believe it or not, this pop-up generates like 20,000 email addresses per month for them. This is a huge, huge deal. And so he’s got this social media marketing industry report and he has a paragraph of content here that explains what you get in it. This paragraph is basically just chock-full of value propositions.

[00:19:50] Okay, but that’s the content in question here. So we tested a couple of different formats because I looked at this and went okay, that’s a lot of content. Right? Let’s test a couple of different ways of formatting this so let’s test one version where we just pull out. the most relevant value propositions, so it’s short. It’s sweet, it’s to the point. Then we took another version where we literally had the exact same content to the T, but just bullet pointed it out. Okay. So we look at these three versions and I’ll have you guys guess on this one. How many of you guys think the bullet point version 1?

[00:20:32] How many guys think the shortened version 1?  How many of you guys think the original version performed best? Two people. So interestingly both of our test versions decrease conversion rates.  This was surprising to me. I thought that the bullet point version was going to work best because it literally said the exact same thing as the control. Just to me, it looked like better formatting.

[00:21:00] S,o again do not make assumptions when it comes to content. Don’t take like the best practice, don’t take, you know, your Executive’s opinion that says we need to say everything in one sentence because sometimes that doesn’t actually work the best.  So with your with your content, you know, I could I could go through dozens of case studies with this.

[00:21:24] Again, look at your own website and ask some of these questions. Is it relevant? Do I know how much content I should have? Do I know how to format it? Do I know where it should be? These are very fun tests to run.

 

 

[00:21:41] Diversions

Let’s talk about diversions for a second. So this is without a doubt, some of the easiest and most surprising tests to run.

[00:21:51] What if I told you guys that 95% of you in this room have at least one thing on your website that is hurting your conversion rates? And you’re going to sit there, you’re going to think what would that one thing be? And there’s probably more than one. So out of all the clients that you know, I’ve worked with hundreds of clients at this point and I have not seen a client yet that has not had at least an element on the side that’s hurting conversion rates.

[00:22:17] So with diversions, there’s some things that you want to think about. You want to ask yourself, obviously, what could be distracting people from my call to action? Here’s some things that are typically distractions. The most likely distractions, contrasting buttons that are not your main call to action. Images can be very distracting, especially when they’re unnecessary images. Other offers that you might have on your page.

[00:22:45] This is a big deal for e-commerce companies because I’ve got so many products. I want to show all of them. Your menu can be a distraction links pop-ups anything literally anything on your site that is not your call to action could be distracting people. Your content might be distracting. So how do we figure out what is actually distracting people, what’s hurting our conversion rates and what’s helping? This is one of the easiest test that you can possibly run. I love this test. So we want to find out what’s hurting our conversion rates. What’s helping our conversion rates? I call this an existence test. You literally have your original version on the left and you take out something on the site.

[00:23:30] So in this case, we took out that accessories row and in our variation over here, we have no accessories. This takes 30 seconds to build this test, you go in and you hide that element. And so we actually tested on this particular e-commerce site. We tested nine different versions of the homepage where we just went through and removed each row on the homepage each row of products. Out of those nine different versions seven of them increased revenue. And we ended up with a version that increased Revenue per visitor by almost 60%. Now think about that, revenue per visitor. So if I know I make $1 per visitor on my site, if it increases by 60%, and I get a million visitors per month to my site, that’s a lot of money. Just by removing stuff, like these just blow people’s minds and it’s such an easy test. And what this does again is not only helps you increase revenue which it definitely does, but it also helps you to identify what stuff is helping conversion rates?  Because if you know that removing the accessories row hurts your conversion rates, you know that that’s an important section of products for people you might want to focus more attention on your accessories. You might want to figure out what accessories we should show there. It’s just you can go really deep down this rabbit hole, but when you have results like this that are increasing revenue, You’re going to be testing your homepage all day long because it has a huge, huge, impact on bottom of the funnel revenue.

[00:25:14] So again, if you go on your website, I would ask what are the things that could potentially be pulling attention away from my call to action. And if you have many call to actions like it’s an e-commerce site then I would ask do I know what products and or call to actions are helping my conversion?

[00:25:35] Anxiety

So now we talk about anxiety. This is kind of interrelated with distractions, but believe it or not, you don’t want people to feel anxiety on your website. I see a lot of websites I go there and they have a big flashing red banner that says warning your computer might have a virus click here to download our thing, like that is the last thing in the world I want to see and that’s the last thing in the world I want to be feeling on a website, right? So with anxiety again, we want to ask could anything potentially be confusing, unclear to me, alarming or frustrating? I like to also think about this in terms of is there anything on my website that people aren’t expecting right?

[00:26:20] So if they click on a link and I take them to a page, is that what they’re expecting to see on the next page or are they clicking on learn more and I’m taking him straight to a checkout page.  So with anxiety, there’s lots and lots of tests that you can run. One area in this is just again an example is comparison boxes comparison charts, right?

[00:26:46] We  love using these in the sales process, especially if we have like a service and we have lots of different options that people could choose we like to use a comparison chart to highlight our best package right? See all these features. So this is for social media marketing world the conference of social media examiner, and they have this ticket comparison page where they’ve got three different tickets and three, you know three different levels of what you get with that.

[00:27:17] And there is there is some anxiety inherent in these kind of charts, right? So we actually tested removing this chart and conversion rates went down. So we knew that this ticket comparison chart was important and there is some you can sense some anxiety with looking at this right. There’s just so much information here.

[00:27:38] And, if I find a ticketing option that I like, what do I do?  So we tested one option which was adding a call to action in there having a button that was at the bottom so that once you see the ticket you like you can click on. We tested another version where we had a button and actually have the price of that ticket and this one was a little tricky because we go.

[00:28:02] Well, the ticket prices are pretty expensive. This might cause more anxiety, but let’s find out what happens. So we tested those two different versions against our original and the version that had the prices performed twice as good as the version that did not have prices, which was surprising.

[00:28:23] Right. And so we dug down this a little bit deeper, right? We know that this section is helping conversion rates. We know that having the price on their helps conversion rates. Let’s iterate off of this. So we tested another version where we call attention to the most popular package, which they didn’t have on there and we increase sales by an additional 9%.

[00:28:44] So when you stack those two tests on top of each other, we’re almost 40% increase in revenue, by digging down one of these one of these rabbit holes that we that we discovered. So again, what I’m trying to do here is not tell you a best practice that you should follow but I’m trying to suggest you guys some possibilities of tests that you can try on your website.

[00:29:10] So, again, you want to ask what things might be alarming, confusing, frustrating, too much effort, and how can I potentially resolve that? Can I remove the confusion or can I simplify? can I remove the frustration or can I add something that makes it less frustrating?

[00:29:33] Responsiveness

So the last thing I want to talk about here is responsiveness and specifically I want you guys to ask how customized is our mobile experience? Do you guys know how your mobile users are different from your desktop users? Because most companies when they’re looking at their mobile website are just taking the desktop version and shrinking it down. Right? Making a responsive website and this partially is, Google’s a little bit at fault here because they have that mobile-friendly test that you can do.

[00:30:09] So I want I want us to stop thinking about is my site responsive but ask is my site customized for mobile? Do I know what should be different about my mobile site because news flash, if you haven’t heard mobile users are very, very, different. They behave completely differently than desktop users do.  So let me show you an example of a site that was mobile responsive.

[00:30:36] Okay, so we’re going to ask this question. What’s the ideal design for a mobile user?  So the version on the far left was our control that is a responsive website, right it works on mobile.  But it doesn’t look good. Right and so we tested two variations of literally the exact same call to action, with wrapped inside of a somewhat mobile optimized design.

[00:31:05] And by the way, these designs are completely different than what their desktop site looks like. So again, we’ll do a little vote here and I’ll just give you guys a clue that control did not win on this one. So how many of you guys think the version in the Middle with the image background performed the best?

[00:31:24] How many of you guys think the blue version performed the best? We’re like 50/50 split on this one, so this one’s very interesting. So the one in the middle increased appointments by 10%, calls by 17%, but look at the difference here, the one with the blue background 41% increase in appointments and 84 percent increase in calls.

[00:31:46] Like night and day. I mean it performed four times better than that version in the middle. And again, if you were to take these designs to the executive team, most likely they’re going to go for the one in the middle. This particular client did want to. In fact they almost just redesigned their site and put the middle one up.

[00:32:06] And that would have been a huge mistake.  So so crazy the night and day difference between I mean exact same content call to actions slightly, I mean there’s one difference in there in terms of the content itself but fascinating how big of an impact a design can have and especially when you’re on mobile and you have such a small screen size to work with how you use that screen space can have a huge impact on your conversion rates.

[00:32:38] So again, we’ve talked about a lot of different areas on your website. We’ve talked about your value propositions. We’ve talked about your call-to-action talked about content talked about finding distractions and removing anxiety or resolving anxiety, and we talked about creating a mobile specific website.

[00:32:58] You need to be testing all of these things on your site. And I can’t tell you which one is going to have the biggest impact. Now I’ve obviously pulled out the case studies, I’ve shown here are some standouts, right? They are areas where value proposition had a huge impact. I’ve also run tests where value proposition has no impact.

[00:33:19] And so those are areas where you want to make sure that you are testing enough on your website to know what are the levers I can pull, what are the things that people really care about on my website. And if you will test through all six of these, I guarantee that you will find at least one area where you can really affect your conversion rates if not multiple.

[00:33:39] And, so finally, I’ve got a I’m going to skip past the slide. So I’ve got an A/B testing starter guide if any of you are interested. I have not talked about any like tools or other types of things that you might want to use. So if you go to this website disruptive advertising.com CMC, I’ve got an a/b testing starter guide that goes through some tools that you can use for A/B testing other tools you might want to consider like heat maps and other things that will help you collect data to inform your testing process. So feel free to go and download that and with that I  think I have just a couple of minutes for Q&A, if any of you guys have questions.  Yes.

[00:34:27] Yes. Yes. So now Taylor in his presentation, just before mine, he talked about some of that depends on how much traffic you have. If you have a very low traffic website, you’re going to want to test a lot of things. Now the challenge with testing a lot of things is you don’t know exactly what had an impact right?

[00:34:49] We had one of our clients that just launched a new version of their website and conversion rates Tanked.  But they didn’t know exactly why they tanked. They didn’t know if something was wrong with the new site. They didn’t know if it was because people didn’t like the new site design. They didn’t know if it was because they had the wrong call to action on a page.

[00:35:05] They didn’t know exactly what happened. And so the benefit of testing one thing at a time is, you know, exactly what impacted conversion rates and you can understand a little bit more of the why. But if you have a very low traffic website, like if you’re getting less than 10,000 visitors a month to a page, then you’re probably going to need to test multiple things at once to get enough of an impact to actually be able to see the results.

[00:35:31] Yeah another question back there.

[00:35:41] Yeah, so images are a big deal and especially on mobile. I’ll say this people on mobile devices love visual experiences, especially if they’ve come from somewhere like Instagram. A lot of your mobile traffic is going to come from social media, you know, Facebook, Instagram and they a lot of times will respond to mobile to visual like imagery type of experiences.

[00:36:04] So, the types of things that I usually like to think about are should I have a person in my image or not? Because anytime you put a person in an image, they become the focus of the image. If I don’t have a person in the image, what else do I have do I have a picture of my product? Do I have a like a visual representation of my sass product?

[00:36:23] Do I show a screen shot of you know something that you can do with my product or maybe your image is just some text right with some icons or maybe some other things and then also again, I also love testing anytime you’re testing different types of images, I would always throw in a variation where you have no image.

[00:36:42] And so if you have no image, then you have to also ask the question. If I don’t have an image then what do I put there because a lot of times we reserve some of our most sacred and prominent real estate for our weird images. And so if you are going to be removing an image. Then we need to be replacing it with something that is very valuable.

[00:37:02] So then we also have to ask. Well what is valuable? What should be there? So yeah. Those are those are huge. I’ve seen huge, huge results from testing images. Also. Another thing to consider on mobile is how big should my images be? Especially if you have an e-commerce site and you have a lot of products like you have your category page where people can see all 40 of your shot glasses.

[00:37:26] How big should those images be? How many columns should we have of products? Like should we just show them one giant shot glass at a time and they scroll down or should we have three columns and you’re seeing like a little miniscule image of it and I can’t tell you what is best there, but those are great things to test.

[00:37:42] Yeah Taylor.

[00:37:54] Yeah, those are fantastic test to run. You bring up a great point because image, if you remove an image, you don’t know was it because the image was too big and it was just slowing down. Because if you’re on like 3G for example, sometimes loading a high def image can take a long time and so it can be really valuable to run tests where you reduce the size of those images. You have one variation where you just have reduced the size. Another variation where you remove it entirely and that way you can kind of tease out a little bit of what’s causing that decrease but that’s a great question.

[00:38:37] Oh yeah, absolutely. If that is what we’re trying to test right? Now, I won’t say that we won’t always track that for every test but that is something that you should definitely be aware of. Yeah one more question back there.

[00:38:57] Yeah, there’s a lot of information about this online. You can read blog posts that are super super long about how long you should run a test Taylor Rose earlier talked about minimal detectable effect. Which goes to statistical significance? I mean ultimately and this is not like statistical significance is not the end-all be-all.

[00:39:17] But you want to run a test for long enough that you get statistical significance and most testing tools out. There will tell you when you have statistical significance. But I always layer on some additional rules on top of that. I would never run a test for shorter than a full week because you want to get at least one of every day of the week in there and my rule of thumb is I typically run tests for at least two full weeks because sometimes you might have an irregular Saturday in there and we sent out a sale on Friday night, and your conversion rate spike on Saturday but they’re typically very low. So I like to get a couple of weeks of data in there.

[00:39:55] And then it’s also again going to come back to how much traffic do you get so, you know social media examiner. He is getting millions of visitors to his site. And so when he runs a test for one day, he has a hundred thousand visitors to each experience. He’s going to get statistical significance very quickly on his.

[00:40:15] And so that’s where we’ll also kind of we you can talk about throttling some of the traffic you pull into a test and so that you can get a full week in there. But those are a couple things to think about. It’s kind of a non-answer, but there’s not really a great answer for that question.  Okay.

Well, thank you guys for coming. I appreciate the attentiveness.

[00:40:30]

This website uses cookies to optimise your online user experience. Some of the cookies we use are essential for the site to work.
By continuing to use our site you agree to us using cookies in accordance with our Cookie Policy.
Accept