The Traffic allocation refers to the proportion of visitors that are put into each variation. Typically, we run 50/50 experiments, meaning that half the visitors should see the control, and the other half should see the treatment.
For more information, see: