For example, create two identical welcome emails, but send one at the time you typically send your welcome emails and one at the time reflected in your hypothesis. Following the hypothesis example above: if you typically send your welcome emails two days after the user joins, send your control email at this time. Your test group email could be sent 10 minutes after the new user joins to test the effectiveness against your baseline results from your control group.
The only thing different between the two emails should USA Phone Number List be the time you sent them. If you were to test more than one element, it is called multivariate testing. For example, a multivariate test would be if you were testing both the time the email is sent and different subject line. You should only use multivariate testing when you are testing combinations of different elements. And it’s best to implement multivariate testing only after testing each individual element.
For example, after you test and find the most effective time to send your email, you can then combine it with winning subject lines to measure the combined impact. If you attempt to test all aspects of an email at the same time, it can be difficult to determine which is contributing positively or negatively to the overall outcome.
1 post • Page 1 of 1