cancel
Showing results for 
Search instead for 
Did you mean: 

A/b Testing question

Highlighted
Occasional Contributor

A/b Testing question

Hello everyone,

 

I have a question about A/B testing. lets say I have a list of 10,000 people. What percent of the 10,000 do I test. Should i do a small amount like 1,000 people for each test or should I just split right down the middle.

 

Thank you

1 REPLY 1
Moderator

Re: A/b Testing question

Hi @RalphC049

 

That's a great question! There isn't one definitive answer as A/B testing can vary depending on what you're trying to accomplish. What is it that you're looking to do or test? Is it testing for one specific campaign, subject lines, or design? This FAQ has some information about A/B testing in case you'd like to review that as well. The two most common types of A/B testing that we see are when are lists split 50/50, or you can test on two smaller sections of a list. For example, let's say that you're testing which subject line gets the most opens- you can send a campaign with one subject line to 2,500 and another subject line to a different 2,500 contacts. Depending on which does better you can then send the campaign to your remaining 5,000 contacts with the subject line that performed the best. I hope that helps, please let us know how it goes!

Natalie B.
Social Media Marketing Manager