Do you regularly perform A/B split tests?
An email marketing strategy plan is not complete without including A/B split testing. However, correctly running tests can be a challenge, and is it worth it?
At Kickdynamic we still believe in the A/B split test and see it performing well for our customers. However, to make the A/B split test really work you have to be accurately testing different elements of your email program.
New to split testing? There are two main ways you can run a split test:
- Total split test:
This is where you split the total list size into the test groups (two for an A/B, three for an A/B/C, you get the picture) and then analyse the results after the campaign has finished sending.
- Live split test:
This is where you test a percentage of the list size first to see what is winning, and then send the winning content to the remaining members on the list.
So what are the two main A/B tests should you regularly run?
A/B subject line test
The A/B subject line test is a classic and should be performed regularly. It involves sending two different subject lines to a percentage of your list, seeing which subject line generates the most opens (and clicks) and sending the email with the winning subject line to the remainder of the list.
This is a great way to test a more risky, different subject line; a subject line that is much shorter or one that contains special characters for example. You can test this with a smaller set of your list to gauge the reaction.
I recommend the live version for split testing your subject line. I have found that there is no real pattern to generating better opens or clicks and it is constantly changing. Therefore with a live test you are seeing what works during a campaign and adjusting accordingly.
A/B content test
The A/B content test is another classic but can be harder to setup and run. It involves sending email with a different content area (main image for example) to a percentage of your list, seeing which content generates the most clicks and conversions and then sending the email with the winning content to the remainder of your list.
This is a great way to test the design of the main images in your email content or your call-to-action placing and colour. It can be harder to setup because it does require building two different HTML emails.
A new way to think about split tests….
A Kickdynamic we thought long and hard about how to improve A/B content testing for our customers (and for ourselves). It was something we had performed 1000s of times before and always thought there must be a better way to do it.
Live image testing using Image Optimizer
Using Image Optimizer you can now test multiple images in a single email with just one HTML TAG. You simply upload to Image Optimizer the images you want to test and we will randomly display the images to a percentage of your list size, determine which image generates the most opens and clicks, and then automatically display that winning image to the remainder the recipients.
This is all performed with just one email content and one HTML TAG – therefore reducing the email build-time and increasing the number of times you can run a content A/B split test. There is also no delay as Image Optimizer is performing the split test and the time of open, not during the send. You don’t have to wait for the email to be delivered to see your how your test is performing.
Want a free valuation of your split tests strategy? Then give me a shout.