There is an affliction that plagues scores of email marketers every day and breeds anxiety and neurosis as soon as the “send” button is pushed. This scourge is called “Sender’s Remorse” and symptoms include immediate hatred of your email’s subject line and intense regret from not choosing your second option. If not treated immediately symptoms can quickly escalate to repeated head-to-keyboard-related trauma and sporadic outbursts of pent-up frustration at the office intern, Brett.

Luckily for you (and Brett), there is an easy cure to this abhorrent affliction: Split testing.

What is Split Testing?

Basically, split testing, also known as A/B testing, is sending out slightly different variations of an email to a selected percentage of your contact base to see which performs better.

Split testing doesn’t have to just pertain to your subject line, however. You can split multiple elements of your emails including:

  • Content
  • From name/ from email
  • Design elements
  • Snippet text
  • Link placement
  • Time of day

The list goes on. You name it, you can test almost anything with split testing. It’s important to keep in mind that testing too many elements of your emails at the same time will result in muddled, unclear results so it would be advisable to use a light touch.

Split testing can be a very valuable tool in finding out what makes a client click or what might deter them from even opening an email.

For example, shorter, more obscure subject lines may make people open, or maybe a personal ‘from’ name might drive the click, by using split testing, you can gain a better understanding of how to engage with your valuable email readers.

Strategies

Experiment with dramatic changes

If you’re going to test, you might as well make it count. Making the tone of your changes dramatically different is an ideal way to test out different content and formats, and quickly gather data on whether the change is popular or a turn-off amongst a sub-set of your mailing list. Besides, changing one word of your subject line or one minor detail of your design could just as easily go unnoticed by your recipients and the data mean nothing, but if you go big you’re sure to have a noticeable difference.

For example, say you’re sending your recipients an email encouraging them to read some content you wrote. You can test between, “Learn more about our services”, and, “Find out more about our services”, and that would be totally valid in certain situations. If you have a very large audience with whom you are quite familiar with and are just trying to hone your wording choices for future reference, for example, this would be an appropriate tactic.

But, if you’re audience is not very large, you may want to go with two completely different tones and word choices in your subject line to get a firmer grasp on what your audience responds to.

Build an archive

When you split test a given email campaign, this shouldn’t be a one-and-done type of endeavor. Instead, you should be archiving the winners of every split test and establishing a best and worst practices list to guide all your email efforts moving forward. Think of your A/B tests like scientific experiments; with every hypothesis you test, you get a slightly clearer picture of how things work. Eventually your findings should compile into a powerful cheat sheet into the psyche of your audience so you can consistently deliver relevant, on brand emails that will perform at an increasingly higher level.

Perform Data-Driven Tests

 Performing split tests on arbitrary variables is all well and good and will gain you some general insight, but the problem is it will distinctly lack context. The most effective forms of split tests are those that directly result from research and solid data which helps form a testable hypothesis.

For instance, if you observe that a large segment of your audience consist of UK citizens, maybe try a split test where the spelling of certain words are changed to reflect the local orthography. In that way, you are actually answering a question that can definitely help you down the line, rather than addressing a random issue that may or may not help somewhere down the line.

Conclusion

Split testing provides lasting insight into improving your emails.

You can implement your findings on sender name or subject line in future emails. If you found that more people clicked on links at the top of an email, then place more links near the top in the future. If more people opened emails with a personal sender name, then have your emails come from a specific person.

Keep split testing, and you will get the most out of the emails you send.


We’re always here to help and advise you on how to get the most out of our app. If you have any questions or comments, please drop us a message at engage@stoneshot.com