We are running AB test with manual 50/50 traffic split between control and challenger, When we looked at the report for 5 days of data, reports shows that in challenger visitor numbers are 15% higher than control. We are doing manual traffic split to both the experience and % is equally divided 50%.
So I would like to understand technically how manual traffic allocation works in Adobe target. 15% difference in traffic allocation in manual AB test is huge for US to decide winning experience.
When you create an AB test with Manual Split you specify the percentage of entrants you want to see each experience. You can split the percentages evenly between all experiences, or specify higher or lower percentages for each experience, though the total percentage for all experiences must equal 100%.
If you are facing a 15% difference in traffic split on a 50/50 AB test, please log a support ticket via Enterprise Support. There is a dedicated support team which we can look into the activity in more detail.
Thanks for your question! You may find some useful information in this article: Adobe Target Help | How does Traffic Allocation works in an Activity in Adobe Target ? Hope this helps!