Full Disclosure: I’m Not Split Testing This
Let me put this disclaimer out there: in the words that follow, you will read about three types of email marketers. I could cleverly mix up the order, but I am not split testing today – well, I am but not here.
Rather, I’m presenting some thoughts on how to make your split testing more effective so you can become a savvy email marketer. More than that, though, let me explain why you need to have proper intent when split testing. Without it, you will end up without any actionable results – and it does not matter in the actions are taken today or later down the road.
Now, I’ve written about 5 Techniques To Avoid The Split Testing Blues and Split Test Optimization, but I believe this goes a step further and uncovers a very common rookie mistake, which we will get into right now.
Split Testing Without a Reason
The first email marketer that we meet simply does not split test, or if they are split testing, they are just running a classic A-B 50-50 test. The problem here is that they are making little use of the information – if any at all. And here’s another point, when I talk about split testing 50-50, it can be ⅓, ⅓, ⅓ or ¼, ¼, ¼, ¼. This concept applies when one is split testing utilizing their entire list (or segment), as opposed to using a portion of the list.
Alas, this marketer sends out campaign after campaign but they are not optimizing the current campaign (which in reality is about to be a past campaign) or future campaigns. Let’s take a moment to unpack that last line because I think it’s very important that we not only utilize split tests for the now but also when it comes to future campaigns.
Split Testing for the Future
You can use today’s email marketing results to determine future campaigns, particularly for your autoresponders. These are the campaigns that you are sending in your automation funnels. For example, when someone signs up to your list, it’s a pretty good idea to “onboard” them – not just one welcome email, but send out a whole repertoire featuring your best content. Naturally, do not bombard them and be careful to send them at the proper cadence.
But how do you know what are your best email marketing subject lines? You cannot simply look at your last 30 days’ worth of data. The problem with that is there are a lot of variables that go into these results, like weekday vs. weekend, morning vs. evening, a day of strong IP deliverability vs. weaker IP status, etc.
Consequently, the only way to get a clear-cut answer is when you are split testing your email list 50-50, right down the middle – same time, same place, yet different results proving subject line A’s superiority over subject line B (or did B outperform A?).
When you get scientific data like this then it is a great idea to take these findings and apply them to your welcome email campaigns.
And since we are not interested in the results for today, the best way to split test this is an even split down the middle into the largest possible segments. So, go ahead and test various content types or subject lines.
Split Testing for Today
The last type of marketer is split testing top optimize their current email. The important thing is that you are split testing the same concepts, like subject lines.
In this scenario, you can offer up 3 or 4 subject lines to 30% of your list. Then, let the winning version go out to the remaining 70% of your list. It’s amazing what a little tweak – say adding the subscriber’s first name – at the beginning of your subject line might do to your email marketing campaign.
Of course, you have to plan this because you do need to give some time to let data accrue. For tests that are examining the subject line, I like to let these “brew” for around 2 or 3 hours, no more than that.
If you are wondering how to do this, ActiveCampaign has the best tools to accomplish this.
If you have content changes – different orders of offers in a newsletter, different images, different buttons, etc. – you need to let this run a little longer because here your target group gets watered down since only the people opening your campaign are effectively the jury. In this case, I usually let it wait for at least 9 hours.
Which type of split testing do you prefer? Either way, just make sure you are not split testing anymore without a rhyme or a reason because if you can deliver the right message to the right person at the right time, you’ll be more than one step ahead of the competition.
