The Importance of A/B Testing in the Obama Campaign’s Email Marketing Strategy
The 2012 Presidential campaign has been a great case study for email marketers. Both the Republican and Democratic parties know the importance of email marketing and used it extensively. The Obama campaign was particularly aggressive, sending numerous fundraising emails that proved to be very effective. In fact, much of the $690 million the campaign ended up raising came from those emails.
What was it that made the difference?
In a recent Bloomberg’s Businessweek article, Amelia Showalter, Director of Digital Analytics gave the answer: “[…] we did extensive A-B testing not just on the subject lines and the amount of money we would ask people for, but on the messages themselves and even the formatting”.
A/B testing is a very popular and simple marketing tactic that consists of testing different variations of the same content to see which one yields the greatest results. The focus is on one and only one variable: different variations of it are used and the best-performing one is the winner.
Applied to email messages, this means testing different subject lines, links, images, layouts, or any other variable that you believe may affect a recipient’s engagement level. For example, the question could be: which of three different subject lines will yield the highest open rate? The Obama campaign, for instance, surprisingly found that the best-performing subject line was a simple Hey.
The Obama email marketing team had 20 writers on staff, sometimes testing up to 18 variations of the same subject line. The key takeaway is clear: keep testing!
With a professional email marketing platform such as MailUp, A/B testing is quick and easy to implement: (click image to enlarge)
Why perform A/B tests? Because predicting what works is nearly impossible without testing. In the words of the Obama campaign staff: “[..] we were so bad at predicting what would win that it only reinforced the need to constantly keep testing.”
Happy A/B testing, then!
By the way: for a complete guide on A/B testing, view the MailUp User Guide at help.mailup.com.