Landing Page Optimization Using Google Analytics

Hey there,

I’m back with more reviews on courses from the CRO minidegree by CXL Institute.

In this lesson, we take a look at Quantitative Landing Page Optimization research using Google Analytics.

Without further ado, let's begin.

Week #8

Course Name: Landing Page Optimization Contd…

Instructor: Michael Aagaard

Total Time: 4h 22 minutes

Google Analytics takes us under the hood of our landing pages and shows us things we could never have seen just by looking at a landing page.

For example: Where are people dropping off? On the landing page or further on in the funnel?

What devices are people using? Mobile phones? If the majority of them (e.g. 90%) are using mobile phones then we don’t really need to worry about optimizing for the desktop.

We’re going to use this lesson to gain important insights from Google Analytics in less than an hour, using insightful standard reports.

We’ll make use of Quantitative Analysis to answer “What and where” questions like:

Users are leaving the landing page without filling the form. So we sort of know what’s going on and where. We obviously need to know the “Why” but we’ll come to that in the next section on Qualitative Research.

For now, let’s get started with Quantitative Research by taking a look a Custom LPO report

The Custom report is one big report that consists of multiple smaller reports.

It gives you insights like:

Bounce rate -?a good indicator of the overall health of the landing page

Transactions and conversion rate - ?it is important to know what your conversion rate is based on and you don’t want to be misled by small samples.

The custom report we’re going to look at is optimized for e-commerce landing pages.

Pick a landing page, then pick a device category (desktop, mobile, tablet). Once you pick a device you’ll see where the traffic is coming from.

Let’s click on landing page/Device/Source:

This will show you the second page which is the page you go to after landing on the landing page. This gives you a good idea if people are doing what you want them to do and if not, then what are they doing?

When you click on Second page/Exit page, it shows you the pages they exit from which gives you a good idea of user behavior.

The next one is Gender and Age:

?You start with device category. Click on the landing page it will show you the gender split and when you click on gender, it will show you the age groups and a breakdown of the conversion rate. This is important because it gives you a good idea of who is visiting your landing page. You can also check to see if the specific target audience you had in mind is part of the people visiting your landing page.

Device/browser:

With this one, we'll start with landing page, device category, operating system, browser and browser version.

This is mainly to check if an operating system or browser is broken. There could be bugs affecting performance and some other smaller issues that we could resolve quickly.

Let’s now look at a live demo of how to go through this report.

We'll be using, Google's Google Analytics demo account.

Take any one page and analyze it.

You’ll see the number of sessions, the bounce rate, and the E-commerce conversion rate.

The /nest-usa page has a relatively low bounce rate as compared to the other pages on the list s well as the industry average. The number of sessions is quite high which means it is getting a lot of traffic which is a good sign.

Clicking on the page takes you to the next level which is the device category.

You’ll see that 96.5% of sessions are on the desktop.

Consequently, most of the transactions (97.9%) and happen on the desktop. Conversions are also very high on the desktop as compared to the mobile or tablet and so it makes sense to focus on optimizing the page more for the desktop rather than for any other device.

Clicking on “desktop” shows you where the traffic is coming from. We see that referral traffic is quite high while there’s not much direct traffic.

This is a good place to see which channels or sources are underperforming and based on that insight, take a call as to whether they should be going to that landing page.

As you can see, this gives us a good overview of how the landing page is performing, how it’s doing across devices, and how it's performing on different sources.

Let’s now head over the second tab (Second Page/Exit Page). We start by looking at the device category.

We start with Desktop because it is the most important category for this landing page. Clicking on Desktop takes us to the second page that users land on after they land on the landing page.

Here we see that users go to the sign-in page after the landing page which might be exactly what you want them to do. Your CTA on the landing page takes them to the sign-in page.

Clicking on the sign-in page shows you the page they exit the website from.

The landing page and the sign-in page combined give you a good indication of user behavior and whether users are taking the action you desire them to take.

Let’s move on to Gender/Age:

We start again with Desktop. We see that most of the visitors are male (78%) but there’s not much of a difference in conversion rates between men and women.

Clicking on “male” takes on one level deeper and shows us statistics related to age groups.

A quick look shows us that the highest conversion rates come from the 55-64 years age group but don’t be hasty to take any decisions based on that.

The sample size is very small and doesn’t warrant any change in decision-making from our end.

Looking more closely, we find that most of the traffic comes from the age groups 35-44 and 25-34 with pretty solid conversion rates as well.

Tip: One statistic for you to keep an eye out for could be if the conversion rates are low for your main source of traffic.

For e.g.: if the conversion rates for the age group 35-44 were way lower than the conversion rates for the other age groups, you might want to do some testing to find out why that was the case.

Time to head over to Device/Browser:

We start with Desktop and clicking on it shows us the operating systems the landing page visitors are using.

We see that most of them use a Mac and that Macs account for the bulk of traffic. The conversion rates from Macs are also high, probably for this reason.

Clicking on Macintosh takes us to the list of browsers where we see the list being almost entirely dominated by Chrome.

Tip: If you had happened to see an instance where the bulk of traffic was coming from a certain browser (e.g. Safari) but the conversion rate was low, that could point to a bug with the browser, which you could then fix immediately.

Clicking on ‘Chrome’ takes you to the browser versions and there’s not much of interest here. However, this too gives you an opportunity to see whether a particular browser version is underperforming.

What we’ve seen thus far are some important quantitative insights that can help us with research into landing page optimization.

Here’s the list of things we looked into. Use this to get started with your landing page optimization.

• Overall landing page performance
• Overall device performance
• Overall browser performance
• Traffic, conversion rate, bounce rate, transactions
• Source, second page, exit page, gender, age
• Potential bugs

Now, let’s look at one more report that’s used in connection with new clients and that’s the Google Ad Campaigns report.

Go to:

Acquisition -> Google Ads -> Campaigns

And you’ll see a list of all the Google Ads. What we’re looking for here are campaigns where the cost is higher than the revenue.

Like in the snapshot the amount out in was $3200 but the return was 0$. This is a good way of finding out which landing pages are not working. You can then choose to fix or remove the landing page from the campaign.

To find out which landing page it is, click on 'secondary dimension' and add ‘Destination URL’. It will then show you which landing page it is.

Manual Step-Drop Analysis

There are a number of steps in the funnel that lead you to the landing page. It is important to know at which step there is a drop is users so that we can focus on optimizing that particular step in the process.

Let’s try and understand this with the help of the following diagrams.

The mistake many people (me included) make, is trying to optimize the landing page if the page. does not convert.

The problem could actually be in the steps following the landing page.

In the following snapshot, there is an 81% drop-off in users from the landing page to the Sign-up page. This means that there is a problem with the landing page.

But in the snapshot below, the problem is with the sign-up page. So instead of optimizing the landing page, it would be better to fix the sign-up page first.

The Users/Page tab allows us to add in the URL of every step in the landing page experience. The number of users at every step is then added in and this gives you the count of the number of users dropping off between steps.

Recap:

• We learned how to get the most important insights from Google Analytics in less than an hour.
• We got to know of the most insightful standard reports in Google Analytics to use in our LPO research
• We learned how to create a manual step-drop analysis so we could pinpoint where users were dropping off from our conversion funnel

That's all I have for this week. I'll be back with another review next week.

Keep an eye out for that!

Sincerely,

Ryan