Random Customer Slice is an attribute that allows you to randomly slice your customer base into 10 “slices”, each accounting for 10% of your customer base.
Every customer in your model is randomly assigned with a number between 1 and 10 (the number is not influenced by their activity or Lifecycle Stage).
A customer’s slice is permanent - once a customer is assigned the 1st Random Customer Slice, they will remain in this “slice” forever - they will never be part of a different random slice. This helps in maintaining the accuracy of tests over time.
Use this attribute to ensure your target group will always include the same portion of customers.
For example, when performing an IP warm-up, it is recommended to start with a small number of customers and increase sending volumes along the way. This can be achieved by targeting one customer slice and adding more slices as the warm-up process progresses.
Another example is setting up a Stream - the Random Customer Slice will be added by default, to ensure accurate testing. The target groups in each of your streams will always include customers from specific slices. This way, you can track the uplift of the entire stream and not only of specific campaigns.
Each time you create a Stream, Optimove will select different random customer slices for each treatment (or for the control). This is to make sure that, for example, the same customers don’t always fall into the control treatment.
