English Español Deutsch Français Italiano Português (Brasil) Русский 中文 日本語

The Best Dashboards For Uncovering Customer Behavior On your Website




Tristam Jarman: Good morning, good afternoon, or good evening from wherever you're tuning in today. Today we'll be talking about the Best Dashboards for Uncovering Customer Behavior on your Website. I am joined by the fantastic Clayton Wood and Jason Barnard. Guys, do you just want to quickly introduce yourselves. Clayton?

Clayton Wood: Sure, yeah. I'm Clayton Wood. I am going to talk a little bit about some of what we can do to look at customer behavior. I've been doing SEO for about 10 years. Jason?

Jason Barnard: I've been doing SEO for 20 years. Significantly older than you, I think. I come from the world of music. I'm a double bass player musician. I used to make cartoons, I was a blue dog in a kids’ cartoon. SEO is my third job.

Tristam Jarman: Just let you guys know, this is actually the second part of a two-part webinar series. The first part we did was a couple of weeks ago. I don't know if you just want to run quick through that, Clayton, for everyone who missed that?

Clayton Wood: Sure, yeah. This is the second part of a two-part webinar, like Tristam said. I'm going to try and make it so that it works independently by itself. The first one, we talked quite a lot about tracking and analytics, and how to uncover what people are doing and how that affects SEO. 

This one is around customer behavior on your website. At the end of this one, I'm going to show some Google Analytics dashboards. And we're going to have a healthy discussion about if we think customer behavior affects SEO.

Tristam Jarman: We've got people from all around the world, Dubai, Brazil, Mongolia, South Africa. Wow. Yes, we've got quite a few people already here. Clayton, do you want to jump into your presentation?

Clayton Wood: This presentation, I had the idea to walk through it, it's about 20 slides long. I'll try and spend a minute or two on each slide, will go 15 to 20 minutes. 

Overview of Online Customer Behavior

Let's get started. Over the years in my career, there's a couple of key metrics that I've learned really help uncover customer behavior. I've put them together and found a couple of dashboards that I think give a lot of insight and tell the story of what the folks on your website and in your leads funnel are doing. That's why we call this webinar The Best Dashboards for Uncovering Customer Behavior.

In the last webinar, in part one of this, we did how to uncover not provided keywords. If you've been an SEO for a while or webmaster, you've known that over the last four or five years Google has just totally shut down almost all that keyword data in Google Analytics and has replaced it with a not-provided label there. 

There's a couple of tools that we learned how last time to uncover that behavior. And these dashboards that we're going to show at the end of this one work best when you've got that system in place in your Google Analytics. We'll talk a little bit more about that coming up.

All right. Here's the deal. Four parts, and we'll run through them: Customer behavior, how does it affect SEO, how to read it, and the dashboards that I was talking about.

Uncovering this data and looking at the behavior of your users on your website is only the first step. This may seem obvious to a bunch of you, but I want to drill it in because it's that important. You've got to do something with it...What you do with it after...that’s what affects your business’ bottom line.

All right. What does customer behavior affect? The obvious stuff. It affects conversions, right? This is a snapshot of a heat map. You can see where all of the clicks are. You can see where the customers liked to click and don't like to click. That kind of behavior will affect how they perceive your brand, how they interact with your brand, whether they buy your products whether they sign up. 

How Customer Behavior Impacts SEO

Number two is; this one that's debatable these days. Does customer behavior affect SEO? This is a screenshot of just the typical SEMrush dashboard, which you're probably familiar with. And I'd love to stop here and ask everybody what they think, or how much they think customer behavior affects their SEO or their organic rankings?

We'll get into it deep here in the next couple of slides, but before we do, I'd love to ask Jason. Do you think that customer behavior affects the organic rankings of a website?

Jason Barnard: Yeah. A few really interesting points that come up over the last couple of weeks. Today I was talking to Andrea Volpini from WordLift here in Rome and he's 100% on-board (that) click-through rate affects rankings. He has actually no doubt about that. 

I was talking to Gary Illyes who said, "But we can't allow click-through rate because if we use click-through rate we gotta say goodbye to personalization." My guess is it does influence, but it isn't a very strong signal in most cases simply because it's such a messy signal for Google to deal with. I do get Gary's idea, that if they rely on it heavily, personalization is going to be a big, big issue.

The second point is something like, I saw Jono Alderson from Yoast talking about ranking signals and he's basically saying what is a ranking signal? He uses a really nice example of saying, for example, for a restaurant, is the food they buy from the suppliers a ranking signal, is the chef they have a ranking signal? 

And the answer is no, not directly. But the fact that they're buying quality food and the fact that chef is a very good chef means that people are very satisfied. They're satisfying the customers and those signals come into Google and affect the rankings.

Then the third point, which is something I've been thinking about looking for great gigs about local search. Local search has always been based on entities and local search has always been based on customer behavior and customer satisfaction.

Those three points for me indicate the customer behavior has to be a very important aspect. Whether it's direct or indirect, whether it's a heavy influence or a light influence, and how much of an influence it is in different circumstances, different context is all up for debate. 

But it is clear that if you don't satisfy your customers, if your customers are not acting in a positive manner towards your business, offline or online, then you're going to be sending out negative signals to Google and it isn't going to help your rankings.

Collecting Data to Assess the Impact of Customer Behavior

Clayton Wood: The biggest thing that we can do is sort of piece together the breadcrumbs that tell that story that you're talking about. Google doesn't have something that publishes all of the ranking signals, but there's so much data out there by SEMrush, by some of the top news sources that you can see whether it affects it or not.

I loved the point about indirectly affecting it. Let's follow some of those breadcrumbs.

Jason Barnard: I do love the idea of breadcrumbs, that's a brilliant idea.

Clayton Wood: I pulled a few out when I was doing the research for this and I think they make pretty compelling cases for, yes, customer behavior affects SEO. The biggest thing, I think we can all agree, that customer behavior affects, is the business's bottom line. This is the point that Jason was making with the restaurant and the quality ingredients or the skill of the cook. 

One of the breadcrumbs is how RankBrain works. The slide that you're seeing now is an idea of the flow of how people search. What happens when they search, what does Google's show on the search results page. And then Google answering the question: did this page satisfy the user, yes or no?

If you think about how Google could tell if a page satisfies the user or not, one of the things would we click-through. Right? I agree with Jason that, if you're on this search result or the query and you're trying to rank for it and it's a query that would get a lot of click-throughs and you're the result that's not getting CTR, then it would stand to reason that the ranking might go down.

Another breadcrumb is by SEMrush itself. SEMrush in 2017, two years ago, released top ranking signals. There's probably 30 or 40 on the list, but at the very top is what I've screenshot here and squared off in red is visits to the website, time on the site, pages per session, bounce rate. This is all customer behavior. It's interesting that two years ago SEMrush was evangelizing this customer behavior either directly or indirectly affects SEO.

Jason Barnard: If I come back to the idea of is user behavior a ranking signal? Maybe not, but it doesn't actually matter because at the end of the day if your bottom line is being improved by improving customer satisfaction on your site you're making a win. And the fact that indirectly those customers keep coming back is going to affect Google, that is going to make Google think well of you. 

Bill Slawski is talking about the quality signals who repeat visits to a local shop. We're coming back to local SEO and rankings again. They track the Android phones, they track Google Maps. And if I go back to the same coffee shop that's in my area time and time and time and time again, that's a quality signal. That's the type that Bill Slawski Google is looking at.

They are looking at customer behavior in local search. And if we go back to what I was saying early on, if you're doing local search today you're probably, or you've been in local search for the last few years, you're in a very good position today because that Google is currently pushing out this argument toward more and more in the way the Google local search has been working for years.

Google local search is based on customer behavior, it's based on customer satisfaction with the reviews, and so on and so forth. And it's now starting to be based on repeat visits. 

Clayton Wood: This Rand Fishkin Ranking Factor, it's a really interesting read. If you know Rand, you understand that he's really theoretical, which in some cases is great. Trying to understand what Google's vision is as a whole is really great.

How to Track Customer Behavior on a Website

The next part of this, and let's move on before we run out of time, is just talking a little bit about how to read customer behavior. For all of the technological advancements that Google's making with their algorithm trying to sort out how to rank the most valuable sites that will give the user what they want when they want it, there is still a long way to go on reading customer behavior in their tracking tools. 

I think we can all agree Google Analytics is a beast and it's vast. If you're new to it or if you're stuck on one thing and you can't quite figure it out, it can be frustrating.

How to read customer behavior? Totally different animal. I've really found a lot of relief and success in trying to read customer behavior with event tracking in Google Tag Manager. This is phenomenal. 

I don't know if you're familiar with the specific functionality in Google Analytics, but there is sort of this reverse goal path in the event section of Google Analytics where you can see all the events in chronological order that a customer triggered or didn't trigger and rather dropped.

What are some of the tricks or go-to’s that you use, Jason?

Jason Barnard: I don't think I'm as advanced as you are with this kind of analysis.

Clayton Wood: But Google Analytics is your go-to, right?

Jason Barnard: Yeah, Google Analytics is what I use. I do a lot of segmentation. That's my favorite tool. To be honest with you, this particular part of it, it's really not something I look at it right there. 

Clayton Wood: Segmentation is a phenomenal tool. Don't undervalue it. This is a really great way to try and see some of the demographics, which can lead to tips and some foresight into what your customer behavior is.

One of the cool things about events in Google Tag Manager is you can place them on anything and for any type of behavior, even a scroll could be an event. This is the part of the presentation where I say, "Make a note of event tracking and get into it." It is really helpful, really valuable. 

Another really great way to understand and act on customer behavior, and this one is probably obvious, but I would love if we could get by a show of hands or a show of an emoji, somebody that would admit that hasn't done this. It's simple. 

Go to Google Analytics. Go to the landing page breakdown of the most popular landing pages. Sort out the crap that you don't need and just look at the core landing pages that you know are important for your website. Then go and look for a super high bounce rate and then optimize based on that bounce rate. How simple is that?

You'll know that there's some typical characteristics of bounce rates based on the type of page that it is. You know, blogs are typically longer or lower bounce rate; people stay on the website longer and typically click-through. A sales page might be shorter.

But if you've got an idea for what those standards are for the different types of pages on your website, go and look for the anomalies. Like the one you see on the screen here, 92% bounce rate. This is terrible. The exit percentage is high. 

I would say once a month, go through and look at what's super high, pull it out, and then either put a heat map on it, do some tracking and find out why people are bouncing and optimize it that way. 

Tristam Jarman: Just before we carry on, I'd say that's a really good point, and mixing what you just showed us there with a tool like Hotjar or the events using a Google Tag Manager really starts to seamlessly piece all these kind of little pieces of the puzzle together.

Clayton Wood: Doing this type of research is extremely valuable to make decisions like, "Hey, we've got only a little bit of this particular product over here. How can we sell out of it and get rid of that inventory?" Or, "Hey, we're running this partnership with a major TV brand or other type of food brand and we want to make sure that it gets off to the right start. Where should we put, that not just as a dedicated page on the website, but in relationships to other products that might couple well with them as a package or something like that?"

And the event tracking is the only way in such a large scale type of situation that you're able to place these recorded data points in such a way that it tells the story that lets you make interesting and valuable decisions. Absolutely.

First Dashboard: Tracking Goal Completion

Okay. The last part of this, and we've got 15 more minutes or so I think, are the three dashboards. This first one... and I've put the Bit.ly at the bottom here for you to look at and install on your own. I'm just going to exit out of my presentation and I'm going to show you the dashboard live because these screenshots don't really do it justice. But it's all about goal completion and demographics.

Now keep in mind, before we go there and get into it, this really works best when you're using a platform like Keyword Hero that allows you to uncover that not-provided data in your Google search console. 

You know how Google search console has all of the clicks and all of the keyword search? Platforms like Keyword Hero allow you to type that data into your Google Analytics and then it allows you to create dashboards like this.

This is a dashboard that I love. It's basically all around goal completions by interesting pieces of data. And you can tweak this to be customized to exactly what you want it to do.

For us, it's important to understand unique visitors, it's important to understand what days of the week in a chronological way. Not just of the week, but we do a lot of promotions and partnerships with major brands. We really need to see the completion rate on a chronological way out of 100 or out of whatever our goals are. We've got a lot of static goals set up.

The age and sex demographics for this particular project are really important. For us, having goal completions by age is really important. That's important because we can translate this information into budgets for social media ad spend and things like that.

I always love having a channel grouping. We're all big SEOs here on this webinar. In many cases, when you put a channel grouping up, SEO is number one or a very close number one. You can see the channel grouping by goal completion, which channel is completing most of the goals, and then the keywords, and pages, and landing pages that are completing most of those.

These are piped in from the Keyboard Hero data, which by the way it's a little complicated to figure it out. If you want to reach out to me, I can show you how to set up having the not-provided data added to your Google Analytics platform. 

You can also go to Keyword Hero and they will do it. They have a freemium version that just does it automatically and it's great. Also, goal completions by most popular landing pages, page views. For us, it's interesting because we do a lot of very geo-targeted stuff by city.

This is dashboard number one, customer behavior dashboard, number one. All right. Let me go to the second one.

Tristam Jarman: Just in between, we've got a couple of questions coming in. I wonder if I can just drop them to you guys quickly.

Clayton Wood: Sure.

Measuring, Assessing and Optimizing High Bounce Rates

Tristam Jarman: It was just regarding, I think people have got a question around kind of the high bounce rate, high exit rate. Let's pick this one. "Can you get into detail of the process of kind of debugging higher bounce rates on certain pages?"And then we've got, "Can you clarify if we should be finding top landing pages that have a big bounce rate or top exit pages and then optimize?"

Clayton Wood: Sure. I was saying before, and I think it's important to understand that different types of pages behave in different types of ways. Let's use B2B SaaS as an example. 

You might have customer stories as a type of page, you might have a product type of pages, you might have industry type of pages. If you can group those pages together and look at them historically and their bounce rates over a longer period of time, you'll start to see that they all follow similar trends with each other.

We've got different categories with different baseline bounce rates. The first step is understanding what those baseline bounce rates are by just looking at the history of that same type of page over a period of time, let's say six months. And then, if you see one that stands out with an extremely high bounce rate, that's a red flag. Right? 

If it doesn't meet the norm of all the other blogs, let's say, then you're not doing something right. Sometimes you can look at that page versus another similar page visually and see what's wrong with it. Like with a blog, it might be, "Oh, look, everybody's bouncing because the blog is only 200 words." That's pretty obvious.

When you're looking at the different pages side-by-side and trying to figure out why one bounce rate is higher than the other and you can't tell visually, then you really need to get to measuring it. And the best way to measure it, in my opinion, is a heat map. I like Hotjar, but there's a bunch of heat maps out there, Crazy Egg, and some others. 

What you want to do is install the heat map and start tracking where the traffic is clicking. Many heat maps will have mouse movement recordings and some heat maps will have squirrel tracking, which means that it will show you how far down most of the people got on the page.

By looking at that heat map data, you can piece together whether your call to action isn't obvious enough, or maybe the call to action is too far down the page. Or maybe even sometimes you start to realize that this page's content isn't the intent of what the person that landed on this page wanted to see. 

Heat maps I think are probably a good first step for that question. Jason, do you have any additional insight here?

Jason Barnard: I think people tend to forget that when a user comes in from Google, it doesn't know your brand. They're completely out of context. On that blog page, you need to say who we are what we do. With that proposition, value proposition, that might be even further to the site.

I think that for me in terms of bounce rate, that's the biggest crime that I see committed by at least my clients is they don't explain who they are and what they do. 

Clayton Wood: They might not need a heat map, but I guarantee you, if you don't have a heat map and after this you go and install one and let it run for a couple of weeks, I guarantee you you'll learn something about what your audience is doing and why. That's why the point I was trying to make was sometimes you might not see much on the heat map and you'll realize, "Oh, I don't have the right content on this page."

Jason Barnard: Yeah. My point earlier on about people forget to explain who they are and what they do, and how to do upsell and call to action. And you're saying there are some people who overdo it and that there is a nice middle ground type.

Two More Dashboards for Uncovering Customer Behavior

Clayton Wood: Heat maps. Okay. Do you guys want to see the second dashboard?

Tristam Jarman: Yeah. I think we've got 11 minutes left, so let's crack through it.

Clayton Wood: Yeah. Let's get through it. This second one looks like this. Some more interesting stuff on it. Again, my favorite part of this whole thing is the search terms on the right-hand side. 

This is coupled with having that Keyword Hero view on your Google Analytics and allows you to see how many people search for it, how many transactions if you're on an e-commerce page, what that looked like, average order value, where those transactions are coming from, revenue, all sorts of interesting stuff. And we won't stay on that for too long.

Let's just take a look at the third one. Now, this is my favorite one. Again, you need to have the not provided keywords unlocked or added to the view in your Google Analytics. 

But this one I like best because it shows specifically long tail, and how many words in the phrase, and their goal completion, and their bounce rate, and their sessions by the different short-term or long-term keywords. This gives a really great idea of what's working best. Is it these long-tail bottom of the funnel transactional type of keywords? Or is it the short ones that are two words in the keywords?

I don't think that I've seen in Google Analytics just sort of naturally ever, but it really gives a good idea of whether the keywords that you think you want to rank for are actually driving revenue.

Those are the three dashboards. I know we rush through them, but this deck and those dashboards are available at the bottom of the slide or SEMrush posts the recording of these webinars on their website for all to see and you'll be able to get those right there.

If you have any questions about them or about how to install the other piece with the not-provided keywords, get in touch with me. I think we've got a little bit of time to answer something now. Tristam?

Organic and Paid Traffic: Do You Need Separate Landing Pages?

Tristam Jarman: Yeah, we do have one question. Just going to kind of right back to the start, Neil was asking, "Dying to know if it's better to drive paid traffic to a separate page to organic traffic. The behavior generally being really different in most cases." What are your thought, chaps?

Jason Barnard: Well, for me, paid traffic, in search at least, the keyword is the same, the search is the same, intent is the same. There isn't any reason to have a different landing page. 

From my point of view, I do a lot of DSA dynamic search ads, and dynamic search ads, I come from the SEO world, so what I will say is if my SEO is what Google understood what my page is about, it's understood what the intent is, and if my organic is converting, I should be able to switch that over to DSA and the two landing pages would, therefore, be the same.

My point of view is the intent isn't different, therefore, the pages should be the same. And if they're not, you should start thinking about why. And in some cases they will be different...but I can't see an enormous number of reasons why that wouldn't be the case.

Clayton Wood: Yeah. I see some scenarios where somebody will have a paid landing page that they've only driven traffic to and will start accidentally ranking on that page for the keyword organically that they want to rank for. And then asked me, "Oh, well, if this is not set up for SEO, should we just create a new one? 

I would have to say, "Look, if Google's liking this page and it's possible for us to not screw with the conversion of it, then yeah, it's possible to leave it for both purposes, paid and organic." What do you think, Tristam? Do you do mostly organic or paid stuff?

Tristam Jarman: I do maintain mainly organic. My business partner heads our PPC. And I know when we do work together with particular clients a lot of the times we do send both sets of traffic to the same page.

But I think it's kind of circumstantial based really. You really got to take each case as it comes. You're doing lead generation or something like that, yeah. I don't have a definitive answer, but I would say take it case-by-case.

Jason Barnard: Well, I think a lot of companies do have set the landing pages simply because it's two different teams. The SEO people, the AdWords people are stepping on their toes and vice-versa. We get this impression in the industry that everybody's got these two different landing pages and that there must be a reason for it, especially when we're looking at that competition. The fact your competitions are doing something doesn't mean it's necessarily right. 

Clayton Wood: Absolutely. I agree. By the way, plug for SEMrush, one of the best tools is the paid ad research section of SEMrush where you can look at the competition and just spy on their landing pages and see what they're doing. 

Tristam Jarman: Okay, we only got a couple of minutes left.

Closing Thoughts on the Impact of Customer Behavior

I think yet we've got one minute left so let's just wrap it up. If I could take one top tip from each of you from today, what would it be?

Clayton Wood: I'd say that customer behavior does influence SEO, so pay attention to it. This is what Google cares about, so it should be what we care about.

Tristam Jarman: Jason?

Jason Barnard: I would add to that. Even if it doesn't affect your SEO, it affects your bottom line because they're buying more from you.

Tristam Jarman: Definitely. I guess, from me, yeah, always keep the kind of customer user in mind and work for them because they're the ones they're going to make you money at the end of the day. If you haven't installed heat maps do it, it might change your world. It's so awesome.

I'll just round off today. Don't forget we've got the Global Marketing Day on the 30th of October with these two lovely chaps, however, my fingers are pointing at you, 30th of October. 

Yeah. Thanks for everyone joining in today. It's been another fantastic SEMrush live webinar. Thank you, Clayton. Thank you, Jason. And I think we're out. See you later.

Jason Barnard: Thank you, guys.

All levels

Check out other webinars from this series