Email Testing Best Practices: How to Set Up a Split Test
If you’ve been tracking your email marketing metrics, you may have identified a few that you’d like to improve. Perhaps your open rate isn’t as high as you want it to be, or you suspect that you could boost clicks and conversions.
There’s only one way to know what can improve those metrics: experiment!
What is an Email Split Test?
A split test, or an A/B test, is a way of comparing a variable (something you want to change) against a control (its original form).
The test uses two audience segments. For an A/B test to provide actionable data, your audience, or sample, size should be large enough to have statistical significance. A good benchmark is a list of at least 1,000 subscribers.
If calculating stats isn’t your thing, though, it’s okay — many leading email service providers offer A/B testing tools to help you get the right sample size and understand the results. (Or you can just have Shawley Marketing manage email marketing for you!)
Following Best Practices for Email A/B Tests
First, identify the reason for doing your test:
More email opens?
Increased clicks?
Better conversions?
Next, decide on one variable you’d like to test. You can do more tests in the future — it’s important only to change one thing right now. Common items to try are:
Subject lines
Images
Body copy
Calls to action
Finally, decide how you’d like to change your variable. You may choose to:
Add or remove personalization
Change a word or sentence
Add, remove, or move an image
Change the placement of words in a CTA
Change the color of a button
Use paragraphs vs. bullet points
As an example, if you’d like to test personalization in your subject lines, you’d use the same subject for both, insert the personalization field into one:
Subject 1: Hey there! Check out these 20 no-bake recipes for summer.
Subject 2: Hey Julie! Check out these 20 no-bake recipes for summer.
Send each of your emails out to a similarly sized segment at the same time, and wait for the results to start rolling in.
Trying to figure out the best time to send your emails? Don’t send your A/B test emails at different times. Instead, try out some different send days and times separately. Once you’ve established the best time for you to send an email, move forward with testing your subject lines and other variables.
What Happens After Your First Email Split Test
If you have a large enough subscriber list to have statistically significant results, you can reasonably establish that the “winning” email variable is what you should utilize in the future. And if it didn’t give you the results you wanted, don’t stress! It doesn’t mean you failed.
There are countless ways you can test variables in your emails. If your audience doesn't respond to a new CTA or personalization, changes to your body copy or email design may be just the ticket.
Not sure where to start? Book a 90-minute consultation with Shawley Marketing! We’ll review your current strategy and develop an actionable list of things you can begin implementing right away.
