How To Implement A/B Testing?

A/B Testing can only be done in the Campaigns section on the dashboard, and only on an active campaign.

1. Choose the campaign you have currently activated.
   
   Figure 1: Choosing an Active Campaign
   
   
2.  Click on the edit icon
   
   Figure 2: Edit icon
   
   
3. Choose which Active conditional Popout you would want to do A/B testing on. In the example below, we have chosen the Popout that triggers after hours (timezone: 6pm - 6am). Note that the status is green/active.
   
   Figure 3: Choosing An Active Popout
   
   
4. Next, click on Edit and choose 'Configure Conditions'
   
   

   

   Figure 4: Configure Conditions
   
   

5. You will be brought to this screen where you will be shown under what condition this Popout is meant to appear.
   
   
   Figure 5: Edit Special Conditions (left half) & Popout Selection screen (right half)
   
   Tip: Steps to set the condition for your Popout is already outlined in steps 2 to 8 under the Add Special Conditions under 07 - How To Build A Campaign
   
   
6. The right half of the above screen is where you will be setting up the A/B Test. First, you will have to toggle the 'Add A/B Test' button.
   
   Figure 6: Toggle A/B Test on
   
   
7. The following will happen when you toggle that button. 
   • A Variation Popout row will appear underneath where you get to select which other Popout you would like to show when the condition on the left side is met
   • Your top Popout will be labelled A and the Variation Popout is labelled B
   • You get to split how much traffic goes between A and B
   • You get to specify when the test ends 
     
     Figure 7: A/B Testing Options
     
     
8. In the Variation Popout section, you get to choose which other Popout you would like to test by clicking on the dropdown button.
   
   
   Figure 8: Selecting your variation Popout
   
   
9. Once you have selected your variation Popout (B), you then get to split the traffic between A and B to see how much conversion one gets over the other. The industry benchmark for A-B testing is to split the traffic 50-50. Which is the default setting. Still, ConversionCow gives you the option to split the traffic however much you want between A and B by moving the white doubled arrow marker in the middle, either right or left.
   
   
   Figure 9: Splitting the traffic between A & B
   
   
10. How the 50-50 split might work is that they will direct even traffic to A and odd traffic to B. 
    
    
11. You can end the test by specifying the end condition. It's by allocating a specific number of events to be reached.
    
    
    Figure 10: Test End Condition 
    
    
12. These are the following test end conditions/events allowable
    
    • User Opens - when a user clicks on the Popout tab to open the Slideout Panel
    • User Clicks - when a user clicks on a button, email or phone within the Slideout Panel
    • Form Submissions - when a user hits submit on the form or books an appointment on Calendly
      
      
13. To the right of the condition, you get to specify the number of the event that you would want to happen before the test ends.
    
    Tip: Ideally you would want the number to be fairly large so you get a good sample of the population before deciding on the more effective Popout for you.
    
    
14. Once the number of event specified has been reached, this would automatically turn off the Popout to allow for you to assess the results before deciding on switching out to the winning Popout only. At this point, you would head back to the Campaign dashboard.
    
    
    Figure 11: Back to the Campaign dashboard to assess results of A/B Testing
    
    
15. Review the results from the A/B Testing
    
    
    Figure 12: Results of A/B Testing show Popup A as the clear winner
    
    
16. You would then need to go back to Configure Conditions under the Edit Dropdown and choose to either keep the original Popout, or switch over to the original Popout to the B version you tested. 
    
    
    Figure 13: Reselecting the original Popout
    
    
17. Finally, you can toggle off the A/B Testing. 
    
    
    Figure 14: Toggle off