X

Welcome to the Oracle Modern Marketing Blog:
The latest in marketing strategy, technology, and innovation.

4 Easy Tips for Your A/B Tests


Unbalanced

 


 

 


A/B testing is just what it sounds like ... you randomly deliver one of two options of your online content to your audience to see which option performs better. At its core, A/B testing is actionable data: It gives you metrics by which to create better content. 

 


A/B testing is not just something for large companies either; small business should be A/B testing, as well. All content marketing strategies should be testing their approach constantly. One of the benefits of a content marketing plan is how simple it is to test your approaches. So when deciding to A/B test, here are four things you should know.

 


Test One Variable at a Time

 


You need to test one variable at a time. Changing the subject line of an email, altering an image or the layout of a CTA button will tell you which method was more effective. If you layer on multiple changes in your A/B test, you won’t be able to say exactly which element is more effective.

 


Test These Common Variables

 


There is no end to the number of changes you can A/B test on your site or email campaigns, but here is a quick list of things you should be testing:

 


  •     CTA wording


  •     CTA color


  •     CTA placement


  •     Number of CTAs on a page


  •     Form requirements


  •     Headers


  •     Images


  •     Copy -- length or style (first person/third person)


  •     A video that auto-plays vs. one that is click-to-play


  •     Special/promotional/free offerings (10-Day Free Trial vs. Moneyback Guarantee)


  •     Location of social icons


  •  
 


Never Stop Testing ...

 


A/B testing isn’t a single event; it’s an ongoing procedure that is driven by the need for more effective messaging. Over time, you will learn to see trends in what works best for your audience, so you can make better decisions the first time around, but you never stop testing.

 


The only constant is change. Just because an image tested better 16 months ago doesn’t necessarily mean that it will always be the champ. And maybe there is another image that can test even better than the first A/B test winner.

 


… But Be Smart

 


When you begin A/B testing, you may want to do an even 50/50 split between your audience. But when you have a winner, you don’t want to risk losing the information you have gained from the test. So when you test again against another variable, you may want to do an 80/20 or 90/10 split. If it looks like the lower-number B is converting at a higher percentage, then let more people see it.


Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.