Podcast: Play in new window | Download
When considering a website redesign when you already have an established audience, the last thing you want to do is change everything on them that’s going to give off not only a different look, but a different feel. Chris Dayley urges us to remember that your website is not for you, it’s for your users, your customers, and your clients. And you need to keep that in mind when you’re undergoing a redesign.
There needs to be a strategic approach when it comes to website redesigns. That includes a great deal of testing, research, and analysis to figure out what works and what doesn’t. It also helps to take advantage of tools like heat maps and customer surveys.
Rich: My next guest is a digital marketing entrepreneur, speaker, and neuro marketer, who gets excited about helping businesses learn what their users want on their website using psychology-based testing and analytics. He started his conversion optimization agency, Dayley Conversion, in 2014, and he’s now launching a brand new company to help businesses with conversion optimization, that has yet to be named. If I know the name by the time this episode drops, I’ll mention it in the outro. But in the meantime, please welcome Chris Dayley. Chris, welcome to the show!
Chris: Thanks for having me on the show, Rich. I’m excited to talk today.
Rich: Yes. I always like our conversations. In fact, you and I had talked before about some potential topics for today’s episode and we ended up on website redesigns. Which is perfect actually because flyte is currently involved in our own long overdue website redesigns. But why are website redesigns so critical, in your opinion?
Chris: Well there’s a couple reasons. Website redesigns are arguably one of the biggest parts of marketing team strategies. I’ve worked with hundreds of businesses in the last couple of years, and at any given time at least half of the businesses I work with are either planning a website redesign, are in the process of a website redesign, or just finished a website redesign. So it’s a big deal. And the average website redesign – at least in my experience – takes about 6 months, so this is a significant amount of time and effort and resources that are poured into website redesigns.
The funny thing about it is, when I go and I start engaging with businesses and I see their process for website redesign – or lack thereof – there’s not really much of a process that most businesses go through. Like when you’re going to spend so much time and resources into a project, you would think that you’d have a really good strategy set forth for how you’re going to do it. And it seems like most business are just kind of flying by the seat of their pants.
Rich: Right. Just winging it.
Chris: Yeah. And so to me this is a big deal that businesses need to be thinking about more than they are.
Rich: So let’s talk about just the first steps. What is a good trigger for a redesign, and maybe what are some not so good triggers? I mean, people choose to redesign their website for all reasons. What are some good ones, and what are some not so good ones?
Chris: So a few triggers, these are common ones that I think are no-brainers. Number one, if your business model has changed somehow. So if you’ve added products, you’ve removed products, you’ve added services, something like that. If your business has changed in some fundamental way, that’s kind of an obvious signal that you’re going to need to change things on your website, and so that’s a good time to look at a site redesign. And to kind of go along with that, if you’re rebranding your company – new logo, new creative assets – those kinds of things, that’s also kind of a no-brainer time to look at redesigning.
Another thing that I see a lot is if you have sufficient data that shows that your website is not performing as well as it used to. This is something that I’ll occasionally see businesses have and it always makes me smile, because I love that they’re using some kind of data to drive their decision. But if your conversion rates have steadily declined over time and your other marketing efforts have remained consistent, or if you’re other marketing channels are getting better but your site is performing worse. There’s other signals, things like high bounce rates, metrics getting worse over time. That’s usually an indication that your site is not resonating as well as it used to with your audience. And that happens, I mean, web behavior is changing constantly and so you’ve got to adapt with your audience’s behavior.
And then another one is kind of a gray area for me, and I’ll explain that. If you feel that your site design is outdated, and the reason this is a gray area for me is if you are the only person that doesn’t like your site design, that probably isn’t the best reason to redesign. I’m not saying it’s a bad reason, I’m just saying if it’s only you that doesn’t like your site design, I would start gathering some additional data points.
Now if you get feedback from your clients, your users, your customers, now you have additional data that says our site really isn’t the best. And then of course if you have additional data – you have analytics data, conversion rate data, other things to further support that – then you might go, ok, our site looks outdated from my perspective, other people have shared the same sentiment, it doesn’t seem to be performing as well as it could, then that seems like a good time.
But the reason that I shy away is, a lot of times you might hate your website design, but your users might like it. And if it’s converting well and it’s working, then instead of thinking about doing a full blown redesign think about making a few tweaks to the site instead of just throwing out the baby with the bath water.
Rich: I agree, because very often we tire of our marketing long before anybody else, because we’re seeing it on a daily basis. We’re like, “Oh my god, I can’t believe that photo is still up”, but I do also agree with you. We went to our website last year and we were like, “Oh my god, we’re not mentioning the fact that we do Google ads or Facebook ads anywhere on this website”, and it’s probably two of our biggest areas in the past two years. So going through that was obviously a wakeup call that everything needed to be reviewed at that point, which is why we’re going through this process.
What mistakes do you see people making when it comes to a redesign?
Chris: So one of the biggest mistakes that is actually rule #1 for me – I’ve outlined 5 rules that I’ve kind of walked through with my clients – the biggest mistake I see is forgetting who your site is for. My wife was watching this show on Netflix about people that renovate their house to turn into an Airbnb. So they would bring in these designers and they would help them redesign their apartment or house so they could rent it out as an Airbnb. And they start taking this one guy to all these furniture stores and are showing him a couch they think would look great in his living room, and he kept saying he didn’t like it. And they kept saying, “But this isn’t for you to sit on, this is for people who are going to come and stay at your house.” And then they go and find some patio furniture and he says he doesn’t like that. And they tell him, again, this is not for you. They just had to keep beating him over the head with this.
And that’s the exact same thing that I hear all the time with people when they’re designing their site or redesigning their site. It’s, “Well I don’t like this. I don’t like these colors. I like pictures that look like this, I like these kinds of sites“. And so the most important thing is that you remember this site is not for you, this site is for your users, for your customers, for your clients.
Rich: Chris, you’re preaching to the choir over here with me right now. And I have to ask, if somebody is woke enough then they realize that this website is not for them, how do they find out what their ideal customers want in a website?
Chris: Yeah, that’s a great question. That’s actually part of my process for creating an effective redesign. You’ve got to start gathering data. So I’ll tell you what most people do – and this is not the most effective way to do this – but what most people do when I talk to businesses and they say, “Oh, I know what my clients want”, I’ll either get “My client is in the fitness industry so I went and looked at other fitness sites and I’m just going to copy them.”
That’s a bad model for a few reasons. You don’t know if these other fitness sites that you’re copying even work well. I mean you could say well it’s Nike, of course it works well. But the thing is, you’re not Nike. Nike has a lot of things going for them that you don’t, like they have a very strong brand. They probably convert well no matter what they put on their website. And so just assuming that because another business is big or is doing well, does not mean that their site is the perfect site. So that’s the first thing that I see.
The second thing that I will see is, “Our users come from Instagram, Facebook, they have these sorts of interests, they’re coming from these destinations, so therefore we’re going to design a site that looks like that.” Again, that’s not horrible, you’re at least getting information about your users. But just because you know some information about them and just because you have a customer persona, doesn’t automatically tell you what people want to see on your website. It gives you a really good idea of who you should try to speak to in your content. It gives you a really good idea of who you’re trying to resonate with through your imagery and that kind of stuff. And so it at least gives you a good starting point. So building up your customer persona of, “This is Jackie and she’s a 30 year old mother that works from home”. If you can put together things like that it gives you a good angle for your content.
Some additional things that I don’t see businesses do very often that are really helpful. Number one is, understand what do people like about your site currently and what do people not like about your site currently.
Rich: And how do we know this? Do we interview our best customers to see what resonates with them, do we go to Google Analytics and see what our bounce rates are on certain pages, is it a combination, or is there something else entirely?
Chris: Yeah, so those are super important pieces. So first of all it is really helpful to either survey your customers – you can either call them by phone, you can have a popup survey – we’ve gathered some incredible data from ‘thank you” page surveys. And it can just be one question at a time, you just ask one question at a time. And the more specific the question is, the better.
So for example it’s not a great question to say, “Why did you buy from us today?” That’s really open ended and so it’s better to say something like, “What were you looking for when you came to our site today?” And then they might give you some good information. Or you might say, “How long did it take you to find the product you were looking for today?” Things like that that give you an idea of was this a very straight forward experience for people, did it take them a long time to actually purchase.
So that you page surveys are great because you’re capturing people that have just purchased, it’s also just as important. I use exit traffic as great serving traffic, because if somebody is bouncing from your site if they’re leaving and you can give them a little exit popup that says, “Hey, what were you looking for today that you didn’t find?” If you’re able to gather some data from somebody that’s leaving and say what was it about our site that you didn’t find, that can give you also information about what does our site does or doesn’t have going for it.
Rich: If we can just talk about the mechanics of that for a second, and I’m getting into the weeds here, but literally how would you set that up for people who either bounced or maybe just tried to exit the website? What are you doing to capture their attention and throw that popup – I assume – in front of them?
Chris: So there’s a lot of tools out there that will allow you to throw up popups like that. So I know Mike Stelzner at Social Media Examiner uses OptIn Monster, which is a very easy tool to add to a WordPress site. You can actually use SurveyMonkey to create popup surveys on your website. So first you’ve got to find a tool that can do it, and then you add the tool to your website. And then you’re going to create inside of the tool it will give you different rules that you can set. So you’re going to say ‘where do you want this popup to fire’. It can be site wide, it can be a thank you page survey, you’re going to set the location up for the survey, and then you’re going to actually write the questions.
So like I said, I would suggest no more than one question at a time. So you can run one question for a week and gather some data, and then change out your question to something else. But if you give people too many questions they’re just not going to answer them.
Rich: Right.
Chris: Because it takes too much time. So that’s the first piece of gathering some data. You do want to gather some of that qualitative data. Qualitative, meaning that you’re getting some qualifications rom people that help you understand why they’re saying what they’re saying.
You also want to gather some quantitative data, some stuff that you can analyze. So you’re going to gather stuff from Google Analytics so get an idea of which pages have the highest bounce rates, because higher bounce rates usually means that people did not find what they were looking for on that page. So which pages on your site have the highest bounce rates, which pages on your site have the lowest time on page?
And again, these are general things to look at. Of course you might have some pages on your site where you want a high bounce rate. For example, your exit page is going to have a high bounce rate, or your thank you page. You might have some pages on your site that are just intermediate pages. So maybe your category pages on your website have a shorter time on page, and that’s just because people go there and then they’re clicking out to somewhere.
Rich: Right, they’re transitory.
Chris: But looking at these things can be some good triggers. So things like that can be very helpful. And then one other tool that I find is really underutilized in site redesigns are heat mapping tools. There’s a free heat mapping tool people can use called Hotjar, and basically you can put it on each page of your site and it will show you where are people clicking, how far down the page are people scrolling, it will just give you another layer of data to say, what do people like on this page and what do they not like.
If they’re not scrolling super far down the page, then that either means they really like what’s up above the fold so you did a great job up there. Or it means that your content down below the fold is not very engaging, and so they’re not going very far down.
But again, this will give you a trigger, it will give you something to look at and say, ‘why is this happening?’, ‘maybe we can adjust that in the site redesign’. So those are like three areas that I typically use to start to inform the site redesign.
Rich: Ok, and that makes a lot of sense. As you get into the redesign itself, how much of this is just about putting a pretty face on content you already have, and how much of it is about just getting rid of everything and starting from scratch? Or does it depend?
Chris: Oh it totally depends. So if you know what people like on your website currently – and like I said, you’ll get some of that from the surveys, you’ll get some of that from the heat maps, you’ll get some of that if you have done any kind of A/B testing on your site – you don’t want to get rid of all that stuff. You want to take that stuff and modify it and alter it in the context of a new design.
If you don’t know any of that, or if people don’t like anything on your site – which is really unlikely because you probably wouldn’t be in business if you had a horrible website – but if people don’t like most of what’s on your site, then you might just want to scrap it all and start from scratch. But again, you want to have enough data so that you’re not just guessing.
But one of the things that is really important here on a site redesign, is I usually will recommend that you have some kind of a measurable goal, like what is my goal of this new site.
Rich: Right. Why am I dropping $5,000-$25,000 or more on a new website? What am I looking to accomplish?
Chris: Exactly, yes. And so it could be I want a new design that converts the same. And that’s a fine goal, but it needs to be something that’s measurable. Because if it’s just, while I want the site to look better, it’s really hard to gauge whether or not you’re successful on that. That comes down to your penny. And so it needs to be something that you can measure so that once the new site goes live you can go, “Awesome, this is a homerun!”, or, “Eh, this is ok”, or, “Crap, we totally failed”.
So I’ll give you an example of a failed experiment. A large billion dollar company here in Utah recently spent about 9 months doing a site redesign. They didn’t have a goal, they just wanted a new site. So they went through this 9 month process and they put their current site on lockdown. So they didn’t change their current site at all, so the current site stayed pretty stagnant. They went through this process, they launched the new site at the beginning of this year, and conversion rates tanked.
So they went, “Crap, that didn’t work! Revert back to the old site!” So they reverted back to the old site and then they spent January through March trying to tweak it, so they said that didn’t work very well so let’s try to keep the design but tweak some other things. So they spent another couple months tweaking, launched it again, conversion rates tanked again. So again, they didn’t really have a goal. And then after they launched it they were like, “Oh, this was a blundering failure”.
And so you need to understand that goal before you start the redesign, or at least when you’re pretty early on in the process, so that you can use that to inform all of your designs.
Rich: Are we launching this in one fell swoop after we’ve done all this work? So, here we’ve tried to design something based on data, based on maybe some customer interviews, we try and build something we think is going to work. I know that you’ve talked about launching in phases, what does that look like exactly?
Chris: So I understand that this is not going to be practical for every business, but I would say when possible – if possible – before you even launch your design. So you’re going to start creating your designs, and I suggest that as you’re creating your designs you’re going to go page by page through the site and you’re going to create new designs. Here’s our new homepage design, here’s our new product or information page design, you’re going to start creating these designs. And at that point, if you can, what I like to do is extract the ideas – so what did we change on this page – and throw it into an A/B test with your current site.
So this way you can start to refine your new design before it’s even launched. And so we did this with another client of ours here in Utah that had actually gone through several failed redesigns. I say “failed” because they didn’t’ even launch them. But they started to create some of these designs and they said, “Hey guys, here’s what we’re thinking for this page”. So we said, you changed the way the content is formatted, you changed the way the page is arranged, and you changed the design for the principles on the page, let’s run an A/B test and let’s test each of those concepts, all of them together and then separately.
And what we found fascinating was some of the changes they were going to make were awesome and had a significantly positive impact on conversion rates, on website behavior. And some of them completely destroyed their conversion rates, I mean some of them had a 50% decrease in conversion rate. And so what this did is it gave us that data early on before they had actually built the thing so that they could pivot and make some changes.
So if it’s possible, I recommend doing that so you can start early in the process and start testing these ideas as they come up. If that’s not possible or if you already have the whole design done and built, if that’s the case then I still suggest doing one of two things.
Either a page by page launch, like A/B tests. So for example you’re going to say I’m going to test old homepage versus new homepage, old product page versus new product page, old info page versus new info page, and you’re going to do a page by page split test. Or, at least test the entire old site versus the entire new site.
Rich: So let’s just talk logistics here for a second. As a website designer this feels like a nightmare scenario to me. So if I’m just testing one page, the homepage, I know how to set it up so Google will come to one page or the other page and randomly serve up one of the two and measure conversions. But once they leave that homepage – we’ll call it the “new homepage” – once they leave the new homepages and I haven’t built out the rest of the website, then they’re going to the old pages. To me that feels like a complete disconnect for the end user and it’s going to hurt my conversion rates in this testing period.
And the other option is, I build out two completely different websites, I randomly show one or the other, and so it keeps within the old site or the new site. But then I’ve already built out the site anyways. Tell me what I’m missing here. Or do I have it and this is just what we have to do?
Chris: So those are great concerns. So let’s address the first one. So your first concern is if you’re testing page by page and someone sees one new design and the rest of the site is the old design. So that all depends on what you’re changing. If you are redesigning everything like the menu, the header, the logo, the color palettes, if everything is different there can be some disconnect there. So usually what I will suggest in those cases is to keep something common at least during the test, like the homepage. And so that way if there is at least some commonality and the design stays the same form page to page, it doesn’t feel quite so jarring.
And then of course if your color palette is completely different – so they’re going from a site that’s all pink to a site that’s all green – that can be pretty jarring, too. So in that case I probably wouldn’t suggest a page by page redesign.
But usually most businesses that I see that do site redesigns, they’re not dramatically changing everything. It’s usually he same brand, the same logo, the same general color palette, usually it’s just variations on the same brand. And if that’s the case, it usually doesn’t feel that jarring to go from an image-heavy homepage with lots of background, to going to the product page that doesn’t have all that stuff. That kind of stuff is actually somewhat normal on websites.
So again, you want to look at this on a case by case basis, and you’re probably also feeling along the idea of, well what if I run this page and it sucks.
Rich: Right, after I’ve invested all this money and time, do I really want to pull the plug on a brand new website. I’m guessing the answer is, “yes”, because better to have lost all that money to stop the bleeding, than to lose all that money and then just continue to bleed conversions.
Chris: Absolutely. And for agencies that are doing this for their clients – like design agencies and marketing agencies – or we build out a lot of landing pages for clients. It’s way better to say, “Hey guys, we ran a test and this landing page doesn’t work. So we’re going to pivot now”, rather than say, “Here’s the new site, see ya later”.
Now of course nobody’s happy when you bill out a new design and it doesn’t work. But if you can say we know what’s wrong or we have an idea of what’s wrong and now we can pivot, that’s a pretty awesome strategy. I mean, that’s what most marketing strategies are. You try stuff out and sometimes it works, and sometimes it doesn’t. And if it doesn’t work, then you pivot and you try a different approach and you go again. So that’s part of the process.
Now for the whole site against the whole site, I forgot what your question was on that.
Rich: Just what that process looks like. Because if you are changing over the whole website, you basically have both sites as silos, and depending on which one randomly they end up on you just keep them in that silo and compare conversion rates.
Chris: Yeah. So what’s really important when you’re testing an entire site and an entire site, it’s really important that you track a lot of metrics. Because you want to really get an understanding of what happens.
So here’s an example of some metrics that we will track when we do old site versus new site. We will track how many people got from the homepage to the – I’ll just use our agency site as an example – how many people got from the homepage to the conversion rate optimization page, how many people got from the homepage to the PPC page, how many people got from the homepage to the blog, how many people went form the blog to a product page, a blog to a category page.
So I’m going to have a lot of metrics that I’m tracking and those are pretty easy to set up in A/B testing tools. So if you’re using something like Google Optimize, they have pretty much anything you can track in Google Analytics you can track in Google Optimize. So you want to track as many metrics as you can. And then of course you’re going to track overall conversion rates; how many leads did I generate as a whole, how many leads did I generate on the old site, how many leads did I generate on the new site.
What that allows you to do is, you run both of them for a couple of weeks and you see the old site is performing better than the new site. Well let’s go in and see which metrics the old site is performing better on. If it’s performing better across the board on every single metric you tracked, then that’s pretty concerning and this new site design doesn’t work very well. But that’s not usually what I see.
Usually what I see is the old site is performing better, “Oh man, look at that! We don’t get very many people form the homepage to the product pages anymore.” And that’s a good eye opening moment for you to go back and see what you did on the homepage, and then you’ll usually have a big epiphany at that point because it makes sense. The old site’s main call to action was taking people to a product page, but on the new site we just have a nice image and the call to action is lower on the page. And that gives you something to tweak in real time. And usually you can salvage the new site if you can just do that and just make a few tweaks.
Rich: That makes a lot of sense. I don’t know if this is outside of your area of expertise, Chris, but do you have a launch plan for new sites? One of the things I’m always concerned about is search engine optimization, and often you take a hit when you do a major overhaul of the site. I’d love to know any tactics you have for addressing those kinds of concerns.
Chris: Yeah. And that is a valid concern. The SEO concern of, we launched a new site and because it’s completely different, Google will freak out for a little bit. So that piece is sort of unavoidable when you launch a new site. And I say “sort of” because there’s some things that you can do. What I will usually do when I launch a new site – of course as often as possible I’m going to A/B test this – I will slowly ramp up the new site.
So as an example, we’ll start out the new site at only 10% of overall traffic, so the majority of people are still going to the old site.
Rich: When you say this you’re using, what, Google Optimize to be able to determine what percentage people are seeing the new site?
Chris: Yes. Yeah, so you can use whatever testing tool you’re using – Google Optimize, VWO, Optimizely – these will allow you to throttle the amount of traffic you’re pulling in to a test. So I will start the new site at about 10%. And if that goes well for 3-4 days and conversion rates seem to be well and everything else that I’m tracking seems to be looking good, then I’ll throttle it up to 30%, and then 50%, and then eventually you want it at like 90/10 with 90% of the traffic going to the new site. And at that point you can flip the switch.
One of the challenges businesses have is if SEO rankings tank, they think that the whole site just failed. But it might just be the SEO rankings went down, they’ll probably come back up. And if you know your new site converts well and you’ve already gathered all the data we just talked about an you know the new site works and you know the audience likes it, then you can be pretty sure that your SEO rankings will bounce back at some point because you still have a great site. In fact you have an even better site now that’s addressing people’s concerns better, you have better bounce rates, and you have better conversion rates. This is the site that your customers want. And so Google will eventually figure that out, but sometimes they just freak out.
Rich: So what you’re saying is, don’t let short-term SEO problems interfere with your long-term business success.
Chris: Absolutely.
Rich: Alright. Chris, this has been great as always. I love your approach to everything. Where can people find you online?
Chris: So of course I’m on LinkedIn, I’m on Twitter, my name is Chris Dayley. Like I mentioned, I am starting up a new consultancy here in the next short bit, to teach businesses how to run these types of tests. To walk people through and train them on the process, train them on the use of tools. And so as soon as I have that page setup I will let you know and maybe we can add that into the show notes.
Rich: We definitely will. In fact, there’s a very good chance that by the time the show drops Chris will finally have figured out what he’s going to call his new endeavor and we will have a link to that from the show notes. So definitely go and check out the show notes. And in the meantime, Chris, this has been awesome. Thanks so much for stopping by today.
Chris: Thanks for having me on.
Show Notes:
Chris Dayley is passionate about helping businesses learn what their users want on their website through his extensive testing practices. Learn more about his process and strategies at his website, and be sure to connect with him on LinkedIn.
Rich Brooks is the President of flyte new media, a web design & digital marketing agency in Portland, Maine, and founder of the Agents of Change. He’s passionate about helping small businesses grow online and has put his 20+ years of experience into the book, The Lead Machine: The Small Business Guide to Digital Marketing.