I would argue that just as important as social media, measurement and goal setting are when it comes to marketing so is A/B testing.
With marketing, it is so easy to make assumptions about your audience. Whether you are creating a landing page, content, an ad on Facebook or an email, we all think we know our own audience. We go full force into a campaign and then are shocked when we don’t get the response we expected.
So what happened?
It’s because you assumed instead of using A/B testing to improve the process. You can’t stick to one idea and think that jamming it down your audience enough times, through every form of marketing, will make it work. Most of the time you will get something wrong about your audience, in fact, you will probably get more things wrong than right, especially the first time around. It happens to all of us! But the ones who recover are the marketers who use A/B testing. You can’t get discouraged by your failures but get motivated to make improvements and turn things around!
Now that you know why you need to A/B test we are going to help you avoid three common mistakes marketers make.
Click Here To Learn More About Converge
1. You Are Testing More Than One Thing
The tests that were always the most challenging growing up are the ones that covered multiple topics. Whether it was in Algebra, History, Literature or even Music. Anytime a test covered just one subject, even if it was every little detail about that time in history, I was confident. It’s when you had to cover multiple eras, books, or composers when everything fell apart, and it all jumbled together. And the same can be said about A/B testing.
The good and bad thing about A/B testing is you can test anything and everything but that does not mean you should do it all at once. Where people fail at A/B testing is they run tests on multiple features at once. And the problem with that is you won’t know exactly what worked and what didn’t work. As marketers we want (and need the answers) fast so this part can be the hard part of A/B testing, you really have to plan for it.
Let’s say you have a landing page and you have an original copy and then for an A/B test, you change the color of your CTA button, the text of your CTA button and the position of it. While yes, you might be making changes to just the CTA button you are actually changing three different elements that could all impact if and why someone clicks the button.
Instead, stick with your original CTA button and then perform an A/B test where you change the positioning of it on the page but keep everything else on the landing page the same. That way you will know exactly if moving it has an affect on people clicking it and you won’t have to assume like you would have if you had tested all three elements at once.
Click Here To Subscribe To Our Blog
2. Focusing On Conversions
Most likely the piece of marketing content you are testing is trying to see if you can increase the number of conversions on a piece of digital marketing material. Whether it’s ads being clicked, e-books being downloaded or emails being opened. You are trying to get your audience to take an action.
So it makes sense that you would focus on a conversion metric- whatever that might be- downloads, email addresses, emails opened.
However, who are you now converting?
Are they people who frequent your website already?
Are they new visitors?
If so, are they lower or higher quality leads?
It’s important that you don’t just celebrate a high conversion rate and think you work is done. Instead, have more defined goals going into the A/B test and dig into your audience.
If repeat visitors are the ones converting on the A/B test that doesn’t mean as much as first-time visitors. Your frequent page visitors could be immune to these changes and used to seeing them, while first-time visitors converting might actually mean something is working. And if they are first-time visitors do they fit your target audience? Or are they just a random person. You want to make sure your results are meaningful and are actually impacting the goal you set for the beginning of the test(s).
3. Ending It Too Soon
An A/B test cannot be run overnight or in a day and then be expected to show you some impactful results. That is why you have to really plan it out! There is a number of things that can happen that will impact your results so we suggest, at the very least, running it for 7 days. We strongly recommend 14 because there could be factors affecting a certain day of the week the first time around that might not be there the next week.
Another reason marketers end their A/B tests too soon is that one of the tests appears to be the clear winner early on. But the reality is, no one can clearly be the winner after twenty-four hours so stick with it! Your sample size is too small and therefore and test results are inconclusive. You don’t know what can happen the next 6-13 days if you keep it going. If you end it too early you could actually lose conversions in the long run and end up going with the result that won’t impact your end goal as much.
Wrap It Up
What mistakes have you made with A/B testing before?
We hope we encouraged you to put an A/B testing strategy in place (for all your marketing content) as it can be incredibly valuable and lead you to find some powerful information. We all assume we know what will work best for our audience but what happens when you put it to the test? When you test different elements you can start to understand why something is underperforming- (maybe it’s the placement on the page or the copy) and avoid these mistakes in the future. You won’t only benefit from an increased numbed of conversions but you will also gain some knowledge that can save you from underwhelming results next time!
Interested in running more tests? Take a look at our FREE e-book, 30 Social Tests to see how you can use content on your social accounts in more impactful ways!