By Chris Woods

Some websites rely on multiple sessions before a user converts, whether that is checking out a high value product like an engagement ring or creating a trading account. Here, a session-level performance view does not tell the full story and taking a view of conversions in Google Analytics by user will be more representative in the context of the data. Read on to find out how this can be done.

Back to blog home

Goals, Sessions and Users

If you have Google Analytics, there's a very high chance that you have a set of goals configured to track and measure the macro and micro conversions on your website and through this, understand historical performance. Goals are set at the session-level and therefore the data view you have is of many users across many sessions.

Before we delve further, just to refresh the memory or for those that don't know, "Goal conversion rate" is calculated as follows:

(Goal completions / Sessions) * 100

This is the default behaviour of goals in Google Analytics and for some types of website, this is fine. You expect users to visit the site and convert within that same session, so tracking conversion behaviour as a percentage of overall sessions makes sense.

However, in many instances a session-level performance view does not tell the full story. Some websites rely on multiple sessions before a user converts, whether that is checking out a high value product like an engagement ring, or creating a trading account. Here, a single session conversion is atypical. We have a number of clients where, actually, taking a view of the user across sessions is more representative in the context of their data.

More Info and Methodology

If the above applies to you then you might consider creating user-level conversion rates through a set of calculated metrics, a feature which is available across all Google Analytics accounts. The only prerequisite is that you are using the Universal Analytics tracking method, not Classic.

For standard Google Analytics, you can create 5 calculated metrics; Google Analytics 360 customers can create up to 50.

Now, onto the methodology for configuring a user-level conversion rate:

  1. The first step is to figure out which goals. Without Google Analytics 360 you might need to be selective, depending on what goals you are tracking.
  2. Calculated metrics are applied at the view level so: navigate to the desired view -> Admin -> Calculated Metrics. You'll need edit access to the view.
  3. Create a new calculated metric and give it a sensible name.
  4. Select 'Percent' as the Formatting Type and then enter the following formula, replacing 'Goal Completions' with your own goal:                                                                                                                                                                                                           {{Goal Completions}} / {{Users}}                                                                                     
  5. Click create. Since the calculation is applied to pre-aggregated metrics it will be available for use immediately and can be applied to your historical data. It's really that simple.

As indicated in the interface, accepted operators include plus ("+"), minus ("-"), divided by ("/"), multiplied by ("*"), so you can achieve more complex calculations by combining goals and operators.

User-level Reporting

Of course, configuration is only one side of the coin. In order to leverage this data for reporting you will then need to build a custom report (like the basic example below: click to enlarge) or dashboard in Google Analytics. This will allow you to view the user-level conversion rate alongside your key acquisition and behaviour metrics. As always, avoid mixing incompatible dimensions and metrics.

Calculated metrics example

Calculated metrics are a supported feature of Google Data Studio so if you use that tool for reporting, you can also leverage any that you have created in Google Analytics. Alternatively you can replicate the groundwork in Google Analytics by creating a Custom field in Data Studio - the methodology will be very familiar.

I have demonstrated just one useful example of calculated metrics. Sessions per user, user-level ecommerce conversion rate, pages viewed per user, weighted conversion rates - these are all achievable and give you additional ways to enrich your data and reporting.

Lastly, feel free to check out the link at the end of this post if you're interested in an inventive GTM use case. Charlie has you covered.

Wrap Up

If you don't have any goals configured yet, you are not going to be winning any prizes. This should be your number one priority.

The vast majority will have a platform for reporting conversion rates at the session-level and if so, why not then try implementing the above and exploring additional calculated metrics. The differences between conversion rates at the session vs user-level can be significant and may completely change the way you view your data.

We're always looking for ways to enhance the data available for reporting and analysis. Poor data often results in poor insights so if you need any help, get in touch.

Image credit to Helloquence on Unsplash.

Share this article