What is split testing?
Two roads diverged in a wood and I… let’s be honest, you didn’t take the one less travelled by. You deliberated for waaaaay too long over which one would get you there fastest with fewest hiccups. And then, spent the whole journey agonising over whether you’d made the right choice. You, Sir, need to know what is split testing.
This is the dilemma you face, too, when you’re about to release an email marketing campaign into the world. Should I run with this subject line or that one? Which (a) image should we use at the top of the page; (b) wording; (c) offer; and (d) which call-to-action?
And, of course, the moment you’ve committed and sent it out, you’re obsessively refreshing your analytics, worrying that you’d have got a better response with the alternative version you’d prepared.
Sound familiar? Then you need split testing.
In this post, we’ll help you answer these questions:
- What is split testing?
- Why should I bother with it?
- How do I split test my emails?
What Is Split Testing for Email Marketing?
In order to figure out which two versions of your email campaign will work better – without having to commit to either of them – you will have to use split testing, sometimes called A/B testing.
A small portion of the total email list will receive the two set up versions of your email campaign. Half of these initial recipients will see Version A, and half will see Version B.
The email marketing technology you’re using to run your split test then swiftly tracks important performance indicators like open rates and click-throughs from this initial batch, and figures out whether Version A or Version B is seeing the most success. The rest of your email lists will, then, receive the winner of the two based on the results.
Why Bother with Split Testing?
The awesome thing about trying out both versions simultaneously is that you take out all the risk of getting it wrong. It means you can try a completely new and experimental idea alongside a tried-and-tested one… without the creeping terror that it’s all going to blow up in your face and get you fired.
It also means you can get really forensic about the tiny details that change how well each campaign does. This works best when you’re only comparing one element at a time. For example, when the only difference between Version A and Version B is your subject line, and you’re comparing the impact this has on open rates. Or the only difference is your call-to-action, and you’re comparing click-throughs.
That’s because, if there’s too much difference between Version A and Version B, you still won’t be any wiser about what’s working and what isn’t. The more you can pinpoint the success factors, the more you can keep honing and improving your future email campaigns, too.
How Do I Do It?
If you’re using an awesome email marketing tool (*cough cough* like ours *cough cough*) to send out your email campaigns, setting up a split test is a doddle. Basically, you just need to create the email campaign and let them do the rest.
Provided you have access to top-notch, built-in analytics too, you can then dive right into the details to figure out exactly what’s making your campaign a success so you can tweak it next time or compare different factors, helping you get more and more benefit out of every A/B test you do.
Have we whetted your appetite, hungry for more information on how and why to A/B test your campaigns? Wikipedia has an example and geeky statistics here – don’t say we didn’t warn you… 6.5% = (40+25)/(500+500).
And, if you’re planning on A/B testing the effectiveness of your subject lines you should check out our article on the best email subject line.