A/B test half a list | Community
Skip to main content
March 18, 2013
Solved

A/B test half a list

  • March 18, 2013
  • 2 replies
  • 1058 views
What is the best way to A/B test half my email list, so that I can send the "better" email to the other half?
This post is no longer active and is closed to new replies. Need help? Start a new post to ask your question.
Best answer by
In the smart list of the smart campaign you are using to send the email, make sure your entire list is included, then add a filter for random sample, and set it to 50.

Under flow, add a step for send email, and make the default "Email A". Add a choice > If random sample is 50, send "Email B".

25% of your list will then be getting "Email A" and another 25% will be getting "Email B".

Wait until you have some results, then send the winning email to the other 50%.

When you setup the smart campaign for the first test, I would also recommend creating a static list called "was sent email A/B" or something like that, and then add a flow step that adds each lead that receives an email from your test to that list.

Then you can just clone the smart campaign, remove the random sample filter, add in a filter that says "member of list NOT IN was sent email a/b", and you're good to go with the other 50%. This should save you some time.

Hope this helps!

2 replies

Accepted solution
March 18, 2013
In the smart list of the smart campaign you are using to send the email, make sure your entire list is included, then add a filter for random sample, and set it to 50.

Under flow, add a step for send email, and make the default "Email A". Add a choice > If random sample is 50, send "Email B".

25% of your list will then be getting "Email A" and another 25% will be getting "Email B".

Wait until you have some results, then send the winning email to the other 50%.

When you setup the smart campaign for the first test, I would also recommend creating a static list called "was sent email A/B" or something like that, and then add a flow step that adds each lead that receives an email from your test to that list.

Then you can just clone the smart campaign, remove the random sample filter, add in a filter that says "member of list NOT IN was sent email a/b", and you're good to go with the other 50%. This should save you some time.

Hope this helps!
March 19, 2013
Thank you so much Gary! This made perfect sense, and was very easy to do.