In this article, we will walk through the steps to setup a treatment/control split for an experimental audience. If you are curious what this treatment/control split does, how to pick the appropriate split sizing, or tips for maximizing your chances of achieving statistical significance in your campaign evaluation, we highly encourage you to check out our GrowthLoop Campaign Evaluation Methodology Explained article.
Step 1: Build your Audience
Creating your experimental audience is easy with GrowthLoop Audience Builder! Simply apply the Flex Filters necessary to narrow down your customer segment to just those fitting the desired criteria. Step-by-step instructions for how to do this can be found here if needed.
Step 2: Select your Treatment/Control Split
Before you save your audience, take a look at the Audience Report on the right hand side of the screen. Here you can see a live view of how many customers fit the criteria defined by the selected Flex Filters. You can also move the slider to adjust how many customers fall within the treatment group (who receive the marketing or sales intervention), and those who fall within the control group (who do NOT receive the intervention for this specific campaign). The default split shown is not necessarily a suggestion, but rather the default split established in the settings of your Audience Builder dataset, and can be changed at any time.
There are several trade-offs to keep in mind when selecting the Treatment / Control Split percentage:
There is often the temptation to bias the sample towards the treatment group in order to maximize the number of users receiving the marketing or sales intervention. Generally speaking, however, the best experimental comparisons between treatment and control groups will be achieved when the size of the groups is similar.
That said, the overall size of the audience matters a great deal when using statistical significance testing to evaluate the effectiveness of a campaign. Larger treatment group splits can be warranted if the overall size of the audience is suitably large.
When in doubt, GrowthLoop can perform what is called a Statistical Power Analysis - an analysis used to determine the minimum sample sizes and effect size (% incremental benefit in the treatment group) required to achieve statistical significance in an experiment.
Common Approaches to Treatment/Control Split:
When evaluating "evergreen" campaigns - campaigns that will be active on an ongoing basis and apply to any new users who qualify - many marketers will opt to split their campaign into two distinct phases.
In the first phase (the assessment phase), a robust control group will be used to ensure a proper determination of the campaign's effectiveness can be achieved.
In the second phase, assuming the assessment phase resulted in a sufficiently impactful incremental performance boost, the size of the control group will be reduced to ~20% (depending on audience size) to maximize the impact of the campaign while still retaining the ability to track the campaign's performance directionally.
Step 3: Save your Audience and Begin Export to Destination(s)
Once you are happy with the treatment/control split, all that is left to do is to save your audience and begin exporting it to your desired marketing or sales destinations! Campaign Evaluation will kick off automatically and will analyze the behavior patterns of those customers who fall in the treatment group and compare to those in the control group.
If you have further questions, please check out our Help Center or reach out at anytime to solutions@GrowthLoop.com!