How to A/B Test Email Campaigns

AB Testing

We’ve talked about the value of video in email, how twitter in email increases views and shares, and how instagram in email can increase engagement.  But what about those tiny, little email details- i.e. subject line, from name, send time- that have a major impact on engagement?  You want to send the best email possible, so doing an A/B test to test these elements before your big send is important.

Do you remember learning about the scientific method back in elementary school?  Our teachers taught us how to run an experiment on a control group (the group that stays constant) versus the variable group (the change).  Well the same goes foe A/B testing.

What is A/B Testing Exactly?

A/B testing in its most basic form is a method of market testing in which you send two versions of your email copy or communication to two test audiences from your subscriber list and track which version receives either the highest unique open rate or highest click-through rate.  Of course one communication will have a varying characteristic such as a different subject line, different image, font, or color.

At StoryPorts we prefer using ExactTarget‘s A/B testing feature where you can play with five variables.

Here are the five elements we test to optimize our email sends:

1. Subject Lines

See what really catches your subscribers’ attention.  How personal can you get?  Is it short and sweet, or more informative and to the point?  The subject line is the second element your subscribers will see when they scroll through their inbox, sending the best subject line is key.

2. From Names

The from name is the first thing subscribers will see in their inbox. Maybe they are more likely to open an email  from a specific person or department, rather than your company.

3. Send dates and times

Depending on your industry, your audience may open their emails first thing in the morning, or later in the day.  Timing is key, if you send at a bad time your risking getting your email buried in someone inbox.

4. Variable content areas within your email

This is an ideal way to test the best location for your call to action.  Is it at the top of the email or towards the end, maybe in the middle?  Maybe a button or a link will work better?  ET lets you test all of these.

5. Two entirely different emails

Yes! This is possible too.  This is how you can test different designs and layouts, to see what works best for your audience.

ExactTarget recommends sending your test emails to 5% of your subscribers per condition if you have more than 50,000 subscribers and to 10% if you have less.

Although it takes time to create an A/B test it is far more economical in the long run to spend 30 minutes creating an A/B test and running it to find the best options versus blindly sending emails out with fingers crossed.  Hope this helps!  Get to testing!

×