5 Techniques to Avoid the Split Testing Blues
If the title for this blog could go on and on without having to worry about space constraints, it would probably read something along the lines of this:
The Absolute Worst Part of A/B Split Testing in Email Marketing and How to Avoid Ever Falling into this Trap with 5 Great Techniques to Avoid the Split Testing Blues.
I get it, that’s a mouthful for a title, so please excuse me if—while winnowing the headline—I opted for the less dramatic phrase, but to my defense, I utilized the second half in the sub-headline.
The truth is, that I hold both aspects of prime importance, but when you’re writing, you don’t always have the luxury of split testing. Sometimes you just got to go with your gut. However, as marketers, there really are plenty of opportunities to split test. And without a doubt, the worst part of a split test is when you come into the office the following morning and you find the results to dead even. Now as the adage goes, “Fool me once, shame on you; fool me twice, shame on me.”
This can be applied to split testing because once or twice I can understand. When you get to the point where everything you are split testing comes back even, however, you have to ask yourself why. If this keeps up, then it might make sense to hang up the gloves and carry on with your regularly scheduled program. After all, split testing takes time and resources to prepare and analyze.
My Top Tips to Spruce Up Your Split Testing Efforts
So here are some tips to make your split testing more efficient and (hopefully) beneficial.
Of course, when analyzing split tests, you need to remember you’re dealing with percentage points. I wrote about this when I asked, Do A/B Subject Line Split Tests End with the Open Rate?
- Now that we have that cleared, the golden rule of split testing is having a clear plan and understand what the differences are, e.g., are we going long copy versus short, are we going to test hero image versus minimal image, etc. Of course, you need a solid naming structure; I say it’s the unsung hero of email marketing.
- Equally important is understanding what you are aiming for. That’s right, set your targets before you begin to split-test. Don’t fall into the trap of, “well, this looks good.” Before you launch your campaign, have clear goals, e.g., “we want a 5% increase in click-through rates.”
- You also want to make sure your differences are stark enough to make a difference. This is why split testing the color of your button may not be enough to move the needle, but a different call to action or button style might give version B the oomf to supplant your control group. Yes, this is where I see many people fail. So, one way to get around this is to have two people come up with the split test.
- Sometimes just go with your gut! That’s right. There comes a point where split testing might be holding you back. You can only do so much, so perhaps scaling back on the split testing is what you need. Moreover, if you employ this method and are just going with one version, always go with your gut feeling—because common sense is gold.
- Don’t overextend yourself. If you only have the capacity for one split test per month, do that and do it properly. Make sure you know the answer to the why and what questions, as in “Why are you split testing? What exactly do you hope to achieve?”
