Anyone who has browsed Google Analytics has come across a collection of variables in the reports. Google Analytics reports consist of a grouping of metrics and dimensions. To create a customized report in Google Analytics or Google Data Studio, you are free to combine metrics and dimensions freely. But be careful! You may be creating a useless report. In this post I will show the difference between dimensions and metrics and what to look for in order to combine them in a report.
What are dimensions in Google Analytics
All the data you see in a Google Analytics report is a dimension or a metric. THE Google explains dimension as follows:
Dimensions are attributes of your data.
In summary, dimensions are a description, a characteristic, a feature or aspect of your data. It’s a qualitative variable of how a metric happened on your site. Some examples of dimensions:
- Source / Medium
Note that the dimensions appear in the form of words, not numbers. Of course, we have dimensions such as time and date for example that are expressed in numbers but are still a characteristic of what is happening on the website (sessions per hour of the day for example).
You can see the dimensions in the first column of your reports:
Reports also give you the option to select other dimensions, and there are several. Do a test, go to the Acquisition> Source / Media report and click on the Secondary dimension button:
When you click to display in alphabetical order you will see the variety of dimensions that can be part of a report. It is a good way to get acquainted with dimensions.
You can for example choose “Device category” or “Type of users” to better understand what devices your users are using or whether they are returning to your site.
But before you do the tests in Google Analytics, what other question are you waiting for an answer to?
What are metrics in Google Analytics
Metrics are the numbers you see in each dimension. They show what the user has done within the site, in numbers. See for example the Behavior report> All pages:
In the example above, Page is the dimension and pageviews, entries, etc. are the metrics.
Metrics need dimensions to be contextualized otherwise they are just numbers.
By default, Google Analytics does not allow you to add a secondary metric, because not all metrics are collected for all dimensions. It may seem confusing but it gets less complicated if you know how Google Analytics collects data.
What is scope in Google Analytics?
Have you ever wondered why Google Analytics displays some dimensions and metrics but others are left out of reports? This is because Google Analytics does not want to combine them, this could show wrong results and lead to incorrect analysis and decisions about the data collected. But why this happen? Why does Google Analytics avoid some combinations of dimensions and metrics? This is related to how Google Analytics processes data: scope. Each dimension and metric can have only one type of scope, which can be:
Whenever a user does something on your site, they will send data to Google Analytics, a hit. Each action is stored. The hit scope is the lowest level of data storage. A page is a dimension at the hit level, as is the language and title of the page. Pageviews, time on page, load time and total events are examples of hit-level metrics.
Examples of hit scope:
|Hostname||Time on page|
|Event Category||Total events|
Read too: "What are pageviews in Google Analytics“
The scope of the session is more time based and is a higher level than the hit level. A session consists of occurrences that happen in just one session for the same user. Examples of dimensions at a session level are source / medium, landing page and device category. Examples of session-level metrics are sessions, bounce rate, exits, goals, and page views per session.
Examples of session scope:
|Source / Medium||Abandonment rate|
|Campaign||Average session duration|
Read too: “What are sessions in Google Analytics?”
The user scope is the highest level at which data is organized. Users can have more sessions and a session can have more hits. Examples of dimensions that belong to the user's scope are user type, days since the last session and gender. Examples of user-level metrics are users, new sessions, and percentage of new sessions.
Examples of user scope:
|Session count||New users|
The scope of the product type relates to all information about a product.
To learn more about scope, watch the video below from Measureschool:
Combining dimensions and metrics
Session level dimensions and metrics contain information about specific sessions. If you decide to combine pages with sessions in a customized report because you want to know how many sessions a page has received during a period, you may find something different than what you expect:
This report will show you something like:
If you think that page 1 has been viewed 591,336 times you are mistaken. What you are seeing is how many sessions started on page 1, because this is the first hit of the session.
There are a few frequent mistakes that are made when combining dimensions and metrics which are explained in more detail by Bounteous.
You cannot combine dimensions and metrics that do not share the same scope. But how do you know which metrics and dimensions can be combined in a custom report in GA? Google offers a metrics and dimensions explorer. On this page you will find all dimensions and metrics. It may not be a very intuitive tool at first, but you get used to it.
Start by checking the two fields above and everything is easier. Click on a dimension or metric box and those that do not share the same scope will already be displayed in gray. It is still not an error-proof tool. If you choose “sessions”, “pageviews” are not grayed out (and are not of the same scope). Therefore, it is up to you to know if the metric and dimension you are choosing for your report share the same scope.
When creating custom reports, segments, or using custom dimensions and metrics, think first about what you want to measure. Think about what level or scope your dimensions and metrics are at. And think if it all makes sense. In general, if you want to add the “sessions” metric to a custom report, stay true to the scope at the session level. And don't combine variables at the hit level with variables at the session level.
Keep reading: “Nomenclature in Google Analytics and Tag Manager”
Keep reading: "Marketing metrics: partner metrics"