Ever wondered if making a change to your email content or subject line will improve your results or make them worse? Make your new year resolution 5 of 12 to split test your next campaign. Then you’ll know for sure.
Split testing (you may hear this referred to as A/B testing) allows you to test different variations of your campaigns on sample segments of your audience. You’ll be able to see which variation performs best and even send the winning variation out to the remaining subscribers automatically.
There are 3 main variations – that is testing open rate, click rate and goal completion.
The first, open rate split testing, tests factors that are encouraging (or preventing) your campaigns form being opened. The most common application of this is subject line. Many studies show that subject line is highly significant in whether people open, skip or simply delete incoming emails. But you can also split test your ‘from name’. This too can have an influence on whether your emails are opened or not. Remember not to mess with your ‘from address’ – especially if you’ve taken the trouble to encourage people to whitelist this.
The second, that is click rate split testing, can be used to experiment on anything in the body of the email (of course this assumes that the email has already been opened). Here there are many possibilities from subtle variations like the style of wording of your call to action, your content itself, right through to the entire design, even the effects of plain text style versus highly designed content.
Similarly, the third approach, goal completion split testing also requires the email to be opened. From there you can choose any goal – it could be to click a link, download some information, or complete an online purchase.
Great! How do I do it?
Setting up a split test is really easy. First create the different versions of the email that you want to test. Then simply schedule the test using the ‘Split Test’ option in our Send section. Here are some things to bear in mind.
You can test up to 5 campaigns at once, although testing more than 2 variations can make it harder to interpret the results.
When deciding on the scale of your test, remember that you can either choose a percentage or number of subscribers to send each test version to.
You can run a split test over a few hours or up to 7 days. Either way make sure you have given your guinea pigs long enough to open and/or interact with your campaign. It varies, but if they haven’t opened in 7 days then, sorry, they’re probably not going to.
One final step. After your split test you can either use the winner as your preferred strategy for the future, or if you have only sent the variations to a sample of your audience then you can select to have the winning variation automatically sent to the remainder of your list.
So, your new year resolution number 5 is to try a split test on your next campaign.
It’s really easy to do and even small improvements in opens, clicks or goal completions can have a significant effect on the success of your campaign.
Here’s a summary of a few actions to take:
1. Read some more on split testing on our website.
2. Take a moment to view our split testing tutorial. We think you’ll find it useful.
3. Review what objective is most important to you (open, clicks or goals), decide on what is likely to affect this, and just go for it. There’s nothing to lose and plenty to gain.
I’ll be back with new year resolution 6 of 12 soon. In the meantime if you have any questions regarding split testing, give us a call, we’ll be pleased to help.