Segmenting Google Analytics Data To Give It Meaning
Segmenting your data gives it context, which gives it more meaning, allowing you to gain meaningful insights and take meaningful action.
Segmenting is just a fancy way of describing the action of separating your data into segments that help you make sense of your data and remove the overwhelm. Once your data is segmented you can see patterns and it becomes easier to explain your users behaviour.
The guys and gals at Google have tried to make your life easy by providing some already defined segments. Take a look and familiarise yourself with what is already there.
You can access the segments from any view you are in, near the top of your screen you will see “All Users” and “Add Segment”.
All users means you are looking at data from all your users right now. Click on “Add Segment” then click on “System” to access the pre defined segments.
Select “Made a Purchase”. Now you will see the data from people who have made a purchase. You can now investigate that data further by looking at where they came from, what did they do before making a purchase, are they new or returning customers, where are they from, are they male / female, etc..
The limitation of these pre defined segments is they are not customised for your business. I have a client who needs a report on their analytics, their market is specifically in the UK. Since we can’t really control who comes to a website, the traffic can come from all over the world. It’s not a good idea to base any business decisions on the behaviour of people from Holland (for example) if the products you are selling are not intended for them. The only thing to do is remove them from your data by applying a custom segment that shows only the people from your target countries. That’s the data we are really interested in.
To do this you add a new segment as before but this time click on custom instead of system. You will see there are a lot of conditions you can use according to your particular business.
For this example, you went set the location at the bottom where it says location.
Ok, so now you have had a very quick introduction to segments, below is a list of some e-commerce segments that you will find useful for your store.
- Sessions with Transactions
- Sessions with Transactions per Device
- Sessions with no transaction that include item being added to cart
- Sessions with no transaction, add to cart action and item being removed from cart
- Sessions with no transaction, product page seen and pageviews > 10
- Days to transaction < 1 (or equal to x, greater than x)
- Sessions to transaction = 1 (or 2, 3, 4, 5, greater than x)
- Sessions with transaction by browser type
- Sessions with transaction by user type (new / returning)
- Sessions with transaction and one product page seen
- Sessions with transaction and coupon code used
- Sessions with transaction and landing page = product page
- Landing page = product page and immediate next step is add to cart action
- Landing page = product page, immediate next step is add to cart action, no transactions
- Landing page = product page and within same session add to cart action
- Landing page = home page, immediate next step is sign in
- Landing page = home page, immediate next step is search for product
To set up these segments you need to practice using conditions and sequences. The good thing is the segments do not alter your data irreversibly like filters do so you can experiment and you can do no damage. just remember what segmentation you are using!
Pro tip: Depending on how many visits your website gets, your data may be sampled. That means only a certain percentage is being reported. Segmentation means you are working with smaller data sets and this can have an effect of the quality of data Google Analytics is reporting. This can cause sampling problems.
You can read more about it here. Lunametrics – sampling problem
If you need help setting up your segments you can always get in touch with me for help.