Google has been automatically upgrading the interface of all Google Analytics Accounts to Universal Analytics. We discuss one of the features, Custom Dimensions and Metrics, that replaces what we know as "Custom Variables" in Classic Google Analytics.
Custom Dimensions and Metrics replaced what we know to be "Custom Variables" in Classic Google Analytics and have a much simpler implementation than a creation in the Admin interface of GA, and a tiny snippet of code to employ. Important Note: when upgrading to Universal Analytics, your Custom Variables will need to be migrated to Custom Dimensions.
Sounds great! But what are Custom Dimensions and Metrics?
Metrics are literally a count of data, such as Sessions or Pageviews. Dimensions, on the other hand, specify what you would like to measure, such as what browser your users are coming from.
- Columns are your Dimensions. Rows are your Metrics.
- Browser is your Dimension. Sessions is your Metric.
- Page is your Dimension. Pageviews is your Metric.
With Custom Dimensions and Metrics, we can bring outside data into analytics and create our very own personalized versions suited to each distinct site alone. This allows room for broader customized reporting and also provides further context to relate our data to.
Firstly, let us cover a couple of fun facts about these handy but powerful little gizmos:
- 20 Custom Dimensions and 20 Custom Metrics are available for Standard Universal Analytics in each Property. (FYI, with Google Analytics Premium you get 200!)
- Every Dimension and Metric must be assigned a Scope. This is its classification and must be one of following: Hit, Session, User or Product (For Metrics, only the Hit and Product Scope apply).
- Each DimensionÎ¾and Metric is also assigned a specific index number (1, 2, 3, etc.) This is important for implementing them on your site.
- Once these have been created, they cannot be deleted. However, they can be activated or deactivated through the Admin interface.
You can name each Dimension and Metric as you choose. Then select a value or range of values, which are set according to what your Dimension or Metric is about. For example, a Custom Dimension named User Type could have the values "New User" and "Returning User."
You can send Custom Dimensions and Metrics with a Pageview or Event hit type, using the code:
On the other hand, if you want to send these for all the hits that occur on a page, you would use the following:
(Note the difference in the code here with the 'set' and 'send' commands.)
The variables in green can be altered appropriately to match your data so that it's simpler for you to know which values have been allocated to each dimension. It's best to ask your web developer to implement this, just to avoid any errors occurring.
Let's look at a few examples.
Our site has various "actions" that can be carried on its pages, such as print, email etc.
Custom Dimension Scope: Hit
Custom Dimensions Values: Email, Phone, Print, Bookmark
With this, we would be able to tie together our reports to see
What URLs are users bookmarking more?
Which pages are users printing compared to the pages that are being bookmarked?
We have a retail site and want to measure users who are purchasing for the first time against those carrying out repeat purchases. We can create the following:
Custom Dimension Scope: User
Custom Dimension Values: First Time Purchase, Repeat Purchase
Applying this to our reports, we would be able to see the following
In which months do first time purchases occur the most?
How many of these are becoming repeated purchases over the time?
With this data available to us, we could drive further into our data using GA's default dimensions and metrics and see which device our users are purchasing from Mobile, Desktop or Tablet.
These are just a few examples to get you started, but in the next part of this series, we'll cover the nitty gritty of how to assign the Scope to Custom Dimensions and Metrics, what slip-ups to avoid, and how Jellyfish's Analytics Team has made use of these to form stronger and even more powerful insights.
Stay tuned for Part 2.