How To Analyse Website Performance With Google Analytics

What is Google Analytics?

Google Analytics is a web analytics service offered by Google that lets you track and analyse your website traffic. It’s a free to use tool that you install by copying and pasting a few lines of code into your website. Then it starts to record visit and visitor data. What pages are visited, for how long, that kind of thing.

There are actually dozens upon dozens of metrics and dimensions available to help you analyse your website’s performance. And in the Google Analytics interface you have lots of handy, pre-packaged reports to make the task easier. They’re great but the problem is that each website is unique which those reports don’t necessarily take into account.

However, with a little understanding of the most important metrics and dimensions, you’ll be able to create your own dashboards and reports focused on your specific website inside either Google Analytics or outside of it using a different analytics tool that will help you make sense of your website’s traffic.

Intent

Before we jump into talking about the important metrics and dimensions you should be tracking, it’s important that you understand the concept of intent. That’s to say, visitor intent.

Most visitors, when they land on your website, are there for a specific purpose. Whether it’s to find out information, to look for a product or service, to sign up for something, schedule an appointment etc etc.

So, whether your website is achieving its goals or not doesn’t depend simply on the number of visits it receives. It depends on how well the intent of the visitor is served. Whether they managed to do what they came to your site to do.

You might think it’s difficult to figure out what the visitor’s intent was but it’s actually easier than you might think. I’ll talk more about how to analyse for intent later on.

OK, let’s start by looking at some of the important metrics available for measuring the success of your website. Firstly on their own as KPIs and secondly in conjunction with some relevant dimensions to help you dig deeper into the data. 

Metrics

First off we have visits and visitors that Google calls Sessions and Users. Good for measuring overall traffic.

But what might be more useful are the metrics New Users, so the number of users visiting for the first time, and Number of sessions per user, which is total sessions divided by total users.

This last one is good for measuring whether people tend to visit your site more than once. If you have a blog, for example, this might be a good KPI to measure because the more sessions per user means the more people come back to read more blog posts.

So all of these metrics are great for measuring overall traffic. But what about more detailed metrics for measuring what people do once they land on your site?

Well, you could measure Bounce Rate which is the rate of single page visits to your site. So, the user arrives and bounces straight off again from the same page. You might be thinking that this sounds like quite a good metric to track, and it is, but not in all cases.

For example, some websites only have one page, perhaps just one long sales page. The bounce rate here is redundant because there’s nowhere else for the user to go.

Another situation where bounce rate might not be very useful is when you’re driving users to a particular page in order for them to carry out a specific action, say, to sign up for a newsletter and they’re not directed to a different page afterwards.

Or perhaps the page you’re driving them to doesn’t have any links to other pages on it. In these cases the expected behaviour is that the user only visits one page so bounce rate is always going to be 100%, or thereabouts, for these pages.

Which is obviously going to impact the overall website bounce rate. So, as you can see, not so straightforward. My advice is to only measure bounce rate for specific pages where it makes sense to do so.

Talking about tracking page performance, other than bounce rate, there are different metrics you can use. Firstly, you have the basic Pageviews metric which is the number of times each page is loaded by users.

But I tend to prefer Unique Pageviews which only counts a page being loaded once in every session. Simply because I think it gives a more accurate picture of how much content on my website is being consumed.

Just because the same user loads the same page twice on the same session doesn’t necessarily mean that they spend time consuming the same content twice.

I’d prefer to know that the user has consumed the page at all rather than how many times. Besides, you have the time on page metric to help you understand how long people spend on a page.

What I also like to measure is the number of Pages per Session, also known as Page Depth. Again, this will only be relevant where your website has several pages but it can help you see how engaging your website is for users.

Another metric for measuring engagement is Average Time On Page. However, this is another one similar to bounce rate in that there’s a big caveat.

Basically, time on page is measured as the time between when the user lands on the page and then when they load the next page. However, like with single page websites, the user has nowhere else to go so can’t load another page on your website which means that time on page can’t actually be measured at all.

That’s right, time on page is only calculated for sessions of more than one page view. So do bear this in mind.

The same problem actually occurs for session duration and time on page that I mentioned before which is calculated as the time between the first and last time the tracking code is fired for each user.

So if the user loads the page they land on and then leaves without doing anything else that triggers the tracking code, the session duration won’t be calculated. So, again, use with caution.

There are more advanced ways that you can better track things like time on page and session duration. It’s done by setting up things called Events that trigger the tracking code when something specific happens.

One I like to use is called Scroll Depth that you set to fire when the user scrolls past specific points on a page. But this is a bit more advanced than the basics we’re focusing on here. Just know that there are workarounds.

Dimensions

Ok, so now that we’ve explored some of the key metrics you might want to track, now let’s talk about some dimensions you can use to break them down and really get a more accurate picture of website performance.

One of the most important things you’ll want to track is where your website visitors are coming from. Also known as the traffic Source. And you also have the Medium dimension that tells you whether the visitor arrived via organic search, paid search, referral, email etc.

But, if you don’t want to get that detailed about the individual Sources and Mediums, Google kindly provides another dimension called Default Channel Grouping which gives you handy combinations of different Sources and Mediums. It’s a great place to start. And if you want to dig deeper you can.

You might also want to track where your visitors live and Google Analytics has lots of geographic dimensions to choose from. From Continent all the way down to the longitude and latitude of the user’s city based on their IP address.

Depending on your business or website, you might not think that where the user lives is that important. And perhaps it isn’t. However, you can use this data to filter out visits from countries that might look suspicious.

For example, you might run a store in a specific town but you’re getting way too many visits from a place like Kathmandu and they’re loading every page on your website. In cases like these there’s usually something dodgy going on. So what you can do is filter out users from Kathmandu from your analytics to give you a more accurate picture of your user data.

I’ll talk some more about filtering and segmentation in a minute but before I do let’s talk about page tracking dimensions which are some of the most useful. After all, it’s pretty important to know what pages users visit and their journey through your website.

First up we have the basic Page dimension which is the full url that comes after your domain name. But depending how your website has been set up you may prefer to use Page Title which is the name given to the page. This might be easier to read.

But the Page Title always remains the same whereas the page url might contain some helpful tracking information so something to bear in mind.

One dimension that, for me, is crucial to track and analyse is called Landing Page. It’s the first page in a session, where the user landed on your site. With Landing Page, it’s important to measure bounce rate. Especially in cases where you might expect the user to go from that page to another one.

You then have the Second Page dimension that you can use in conjunction with Landing page to see where people go to (if anywhere) once they land on your site. I talked earlier about intent and this can give you an idea of what a user’s intent is.

Especially when looking at your Home Page. If the About page is the Second Page, they’re probably looking for information about your business. If the second page is your contact page then they’re there to contact you etc.

Those are just a couple of examples. Obviously every website is different but it gives you some ideas for measuring the performance of specific pages.

As well as Landing Page and Second Page, which are set for each session, you also have Next Page which will tell you which page was visited next when you use it in conjunction with, for example, the Second Page or indeed the regular Page dimensions.

One final page dimension you might want to track is Exit Page. This tells you what the last last page of a user’s session was. Perhaps you’re driving people to a specific page to complete a task that will redirect them to, say, a thank you page. You could then see how many people landed on the first page and exited on the second.

Goals and Events

This kind of thing could also be tracked by setting up goals and events in Google Analytics. That’s out of the scope of this video but if you’re interested in finding out how to do this then let me know in the comments below and maybe I can do a new post on it.

Segmentation

The last thing I want to talk about is also something that you set up via the Google Analytics interface and that's Segments.

Segments are like filters that you can use to create subsets of users or sessions. Earlier I gave the example of unexplained visits from an unexpected place. In this case you could use a segment to exclude users or sessions from that country so that you’re only analysing “good” traffic. So, Segments are really useful tools.

There are quite a few predefined Segments that Google sets up by default that you can use or you can create your own. I definitely think you should at least play around with them and see whether you can learn anything from these subsets of your data.