Example Scenario: Identifying Valuable Customers

When it comes to focusing your marketing efforts, often it mean understanding who is your most valuable customer. Industry research shows that recurring customers spend more than new customers. Let’s start our analysis there.

Let’s say you’re an online retailer for a new line of solar flash lights. You want to understand the differences between new and returning customers. First, you want to create a new view based upon the standard dimension for New-Returning User to see how each group stacks up. Start with a smaller date range at first until you figured out the view definition, then expand from there.




As you can see, the total number of new user sessions last month was higher than those for returning customers. But, let’s look at who is actually converting. To do this, we need to add a measure to the view for conversions. If it exists, you can search for it in the Available measures side panel, but if it does not exist you can create a new Standard or Calculated measure pretty easily. We have a conversion rate measure already built, so we simply drag it from the Available measures panel onto the view.


Clearly the returning users are converting at a rate more than twice that of new users. That’s certainly good to know, but what is the business value of these events?

Since we have data on sales, let’s add the Revenue measure to the view. The Revenue measure is based upon a custom parameter collected by the tag. If you’re doing online retail, you probably are collecting this also. In Explore all standard and custom parameters are available, but if you don’t see a specific custom parameter in the list, then simply go to Parameter Names within Account Settings and add it to Explore. You need to have permissions to add parameters, so talk to your administrator if you don’t.


Wow. As a group, returning users spent almost twice as much as new ones during the month. What does this mean for individual purchases? You can drill in even further and add the metric for Average Order Size to the view to better understand their value.

Note: Average Order Size is a custom metric. If you don’t have it already, just click on Measures and then the Create New Measures button. You’ll have two options, Standard or Calculated. For example: Average Order Size is a Calculated metric based on Units Purchased divided by Purchases.



Once the calculated metric is created, you can add this to your view as a new measure to understand order values.

What can we learn from our best customers? Using ad hoc data exploration we can find out more about this valuable customers to understand them better, and keep them coming back for more. By understanding more about this group of returning users, perhaps we can better market to new customers and turn them into returning and loyal customers.

One way to do that is to look into where these customers are coming from. Let’s do a quick segment on returning customers. You can do this very quickly by clicking on the “wheel” next to “returning users” and selecting the funnel icon that represents a segment.



Now, let’s look at geography by adding a drilldown for country.




Looks like we have a lot of high converting customers from Guam and Haiti. However, the customers from Guam appear to have higher order values than from other regions. Looks like we should consider running campaigns to find new customers in the area.

Let’s make sure we can deliver the right experience. What devices are they using? Add another drilldown for Device Type.


Looks like three-quarters of customers from Guam are using a desktop device to view our site. We can use this information to better tailor the experience for them—and perhaps explore other aspects of these high-value customers to help increase their ranks.

Now you can see how easy it is to dig deeper into your data to learn more about your customers and understand their value.