If you think you can just chuck a website design up on the internet, and expect the money to come pour in, I have a message for you: stop dreaming. Having a website is like having a dog: you have to train it to do what you want.
It’s easy to see a dog’s performance, you simply give the command, and if it rolls over, you give it a treat. With websites, it’s not so easy, because there are limited ways to judge its effectiveness.
If you are an eCommerce website, you can tell when you make a sale. If your website generates leads, you can tell when you get a new lead. But how do you tell what your most popular page is? How do you tell where most of your traffic comes from? How do you figure out what to improve next on your site?
The answer is Google Analytics- a free tool that tracks the behaviour of each website visitor. Using this, you can discover new insights into how your website functions.
So let’s set you up on Google Analytics!
Website design eats up time- if you’re pressed for it, you can download this article as a guide and refer back to it later. Click here to download the article!
You will be presented with the “New Account” page, which looks like this:
Fill out the form as follows:
Set your Data Sharing Settings, according to what you are comfortable with, then click “Get Tracking ID.”
You will have to accept more Terms of Service on the next page. Be sure you set the region drop-down box to be your country.
You will be brought to the Tracking ID page.
Your tracking ID looks like this:
If you’re using Shopify, you can paste this into your store by going to Online Store > Preferences.
Then just paste your tracking ID in the Google Analytics Account box:
If you’re not on Shopify, then simply copy the Global Site Tag (gtag.js) and then paste it in the <head> tag of your website’s code.
Once installed, refresh the page and press the “send test traffic” button.
It will make a pop-up of your website, and then it should show you that test traffic has successfully made it to your website:
Awesome, this proves everything is working! However, we want to make it so that internal traffic from our organization is not included in our reporting- otherwise, we will be making inaccurate judgments on our website’s performance.
Click the gear in the lower left-hand corner to get to the Admin page.
Google Analytics nests information- an Account (business) contains multiple Properties (different websites), and a property can contain multiple views (filter configurations.)
You don’t want to create filters on your first view- keep at least one view for un-tampered-with data, just in case you need it.
In the righthand column, click “Create View.”
Name it, set its time zone, then press “Create View.”
Back in the admin settings, click “Filters” in the view column.
This is what the filter section looks like.
Press “Add Filter”
You will be brought to the “Add Filter to View” page, where you will make a filter.
Name your filter something along the lines of “Exclude Internal Traffic.”
Now in the dropdowns, we are going to tell Google Analytics to filter traffic from your office’s IP address. So in the first dropdown, choose “exclude.”
In the second dropdown, choose “traffic from the IP addresses.”
In the third dropdown, set the expression to “that are equal to.”
Now all you need to do is paste your IP address into the field below:
Don’t know what your IP address is? Just Google “What’s my IP address?” and Google will tell you.
When you’re finished, press “Save.”
If you’re an eCommerce, you’ll also want to activate eCommerce Tracking. To do that, from the Admin panel, click “Ecommerce Settings” in the right-hand column.
Flip the “enable Ecommerce” switch.
Another switch will appear, “Enable Enhanced E-commerce Reporting.” Flip that switch, too.
Another step will appear, “Checkout labelling.” Ignore this if you’re using Shopify, as it doesn’t work for Shopify. If you aren’t using Shopify, and you’d like to use this feature, here is a guide that can show you how to set up Checkout Labelling.
Great! You’re all set up to start using Google Analytics!
Google Analytics divides data into five reports: Realtime, Audience, Acquisition, Behaviour, and Conversions.
For the purposes of this introduction, we’ll ignore the Realtime report for now, (all it does is show you the present state of traffic on your site), and move to the Audience report, which you can access by clicking “Audience” in the left-hand toolbar.”
This will open a drop-down with many different options. Click “Overview” for now.
This brings you to the overview report, which looks like this:
This is how all reports on Google Analytics are structured. It might be overwhelming at first, but once you know what to look for, you will take to it like a fish to water, or a hand to a glove, or a fish with hands to a glove specially-designed for fish with hands.
The first thing we look at is the date range, in the upper-right-hand corner. By default, this will be set to the last seven days. You can look at whatever date range you like, and even compare date ranges against each other. Configure this by clicking on the date range.
Let’s set our date range to be the last thirty days.
Then open the dropdown and choose “last thirty days.” Google Analytics will update to show you the data in this date range. Gnarly!
Now, let’s look at the line graph. This gives you a good idea of the trends of your traffic’s behaviour. You can change the periods displayed on the X-axis to be hourly, daily, weekly, or monthly. So if we set the data to hourly, we can determine what time of day our audience is most likely to come to the website.
You can hover over any part of the graph to get information about that particular data point. Looking at the peaks, we can see that traffic is highest between 7PM and 9PM. This is the power of Google Analytics. If we were doing a digital marketing campaign meant to bring traffic to the website, we could now try increasing ad spend between 7PM and 9PM.
It’s not just traffic you can look at- using the drop-down, you can look at whatever metric you wish:
And by clicking “select a metric,” you can even compare two metrics against each other! So for example, you can compare Average Session Duration versus Pages per Session. We can see they are closely correlated:
However, below the big graph are all available metrics, presented in miniature:
These are all self-explanatory, buf if you don’t understand exactly how the information is gathered, you can hover over any metric, and a tooltip will appear with an explanation. These become really useful when you use the “compare dates” feature, but first let’s look below these, where you’ll find a table, which shows you information about your user’s demographics and devices they use:
In the left-hand pane of this section, you can choose what metric to look at:
So for example, we can see that most of our visitors come from Auckland.
Useful. But back to the graphs.
In the date range selector, set the date range to the last 30 days, then click the “Compare To” checkbox. It will highlight the previous thirty days in orange, as well as the last thirty days in blue:
When you press apply, you’ll notice some changes. First, the graph changes to compare the two date ranges:
And also the miniature graphs will display not the amount of users, but the rate of change:
Oh no! We see that there has been a drop in almost everything! By looking at the graph, we can see that traffic has been equal across both months, except for a few days in June with abnormal levels of traffic:
Where did that extra traffic come from? To find out, we’ll have to go to…
In the toolbar, click “Acquisition.”
This brings you to the Acquisition report, which looks like this:
This breaks down your Audience according to what brought them to your site, e.g. Organic Search, Social Media, or Paid Search. The pie chart shows you the channels that bring you the most traffic.
If you want to get more specific, you can change the dropdown to display according to source (e.g. the individual website). Next to the pie chart are graphs showing your users and conversions over time:
And below that is a table containing bar graphs:
Do you know your ABC’s? This table will tell them to you: Acquisition, Behaviour, and Conversions.
Each column of this table can be modified so you can look at different aspects of your traffic’s behaviour:
But anyway, what about our conundrum- why was traffic much better last month than this month? Let’s set our date range back to comparing the last 30 days with the previous period: From the audience report, we say that there was 16.92% fewer users on this website in June-July versus May-June; or about 8,000 fewer visitors total:
In the bar graph, we can now see, by channel, what channels were responsible for the drop in traffic:
For example, we can see that there was 30% fewer traffic. We can click into “Other” to see what channels this represents. This brings us to the Explorer report, which looks like this:
There is the line graph again, but below is a data table displaying every source, and the metrics that match it. At the top of each column is a sum of the data represented in that column. You can click on these columns to sort their data in ascending or descending order, indicated by the flipping arrow.
Looking at the data table, we can see there was a 77.5% drop in traffic from “Stuff”
If we click on “Stuff”, we can isolate it in the chart:
In the line graph, we can see that the peaks of this graph correlate with the graph we saw in the Audience overview.
In fact, this client had purchased sponsored content on Stuff.co.nz on these three days, which explains why the traffic was higher back then.
So that’s how you use the Acquisition report, to learn about how people are coming to your website. But what do people do once they’ve arrived at your website?
To find this out, we’re going to use the…
At first glance, the Behaviour report looks very similar to the Audience report.
The available metrics are different:
You can find more information about the behaviour report by either clicking “View Full Report” in the bottom-right-hand corner, orby clicking “Site Content > All Pages.”We see the data table again:
So when we compare the date ranges, we can begin to make inferences about the performance of this website.
This displays absolutely every URL ever visited on your website, which can be problematic because it includes automatically generated URLs.
So let’s limit the pages we’re looking at to those that had over 1,000 views in the last month, by clicking “advanced” at the top of the data table.
Then change “Page” to “Page Views.”
Then add minimum pageview amount in the box.
Then press “apply.”
The Search Box will now say “Advanced Filter ON”, and you can press “edit” if you ever want to change the filter, or the “X” if you want to delete it.
Something to be aware of is your Bounce Rate: the rate at which visitors come to a page, but then leave without interacting with it. Click the top of the Bounce Rate column to sort according to the highest Bounce Rate:
We can see that the page collections/sale has the highest bounce rate out of all these pages.
Since it also gets over 20,000 page views, it makes sense to make efforts to reduce bounce rate.
RIght now we’re looking at the bounces for all users. However, we can compare two different segments of our audience. At the top of the page, click “Add Segment.”
Segments are portions of your audience, separated according to a certain characteristic.
So let’s click “Organic Traffic” – users who came via a search engine.
Looking at the bounce rate now, we can see that it’s much lower versus the total:
If we add segments for Facebook Traffic and Referral Traffic, we can see that the bounce rate is high when visitors come in via an ad.
So the way visitors come to the page has an effect on the bounce rate. These ads should be reviewed to ensure that the offer they are promising is reflected on the page.
To find this out, click “/collections/sale” in the data table.
Now, we need to add a secondary dimension, which can be done by clicking the dropdown:
Now type “Source/Medium” into the dropdown menu. And select it.
Now we can see traffic to /collections/sale, separated by where they’ve come from:
Now if we filter out all sources with less than 1,000 page views per month, and then sort by ascending bounce rate, we can see that NZ Herald, Stuff, and Outbrain all have bounce rates of about 90%.
Whereas Google CPC ads and Facebook ads have a bounce rate of 78% or 61%, respectively.
So it might be a good idea to invest the budget from NZHerald, Stuff, and Outbrain into Google CPC or Facebook Display ads.
Conversions are the most powerful part of Google Analytics- you can determine what actions are the most important to you, and then start tracking how often these are completed.
Conversions are so important that we’ve already covered them in another article, which you can read here!
This guide gives you a good introduction on how to start using Google Analytics. Hopefully this demystifies this application, which can be obtuse initially. We plan on releasing more articles on how to use specific parts of Google Analytics sometime in the future, so keep an eye out for further articles!
We work with clients on a month-by-month basis to provide custom analysis & data-driven strategies, to optimise and increase online conversion rates for eCommerce & business websites.
If you are serious about growing your business, click the button below.