What is a control group?
A control group is a set of people or markets who don't receive the marketing treatment being tested. Comparing their results with those of the exposed group shows what the marketing actually changed.
Last reviewed .
Why you need a control group
Sales rise and fall for many reasons, including seasons, promotions, and news. If you only look at people who saw your ad, you can't separate the ad's effect from everything else happening at the time.
A control group experiences the same conditions without the ad, so the difference between the groups points to the ad's effect.
Building a good control group
The control group should be comparable to the test group apart from the treatment. Random assignment is the preferred approach, including random assignment of geographic regions or matched geo pairs. When randomization is infeasible, carefully matched markets and explicit assumptions can support a quasi-experiment.
The groups also need to be large enough to detect the effect you expect, and they should stay separate for the whole test.
Common questions
What is the difference between a control group and a test group?
The test group receives the marketing treatment, such as seeing the ads. The control group doesn't. You compare the two to measure the effect.
Is a holdout the same as a control group?
They're closely related. A holdout is a control group made up of part of your audience or market that is deliberately withheld from the campaign.
Can a control group see other marketing?
Yes, and that's fine as long as both groups are equally exposed to it. The test isolates only the one thing you changed.
How big should a control group be?
It depends on your baseline rate and the size of the effect you want to detect. Smaller effects need larger groups, and a sample size calculation can estimate the amount.
Want help putting this to work?
Tell us where your marketing stands today, and we'll show you what to do first.