Consider I have taken a random sample for forming control group in month of January with sample size 5k and treatment group size being 95k.
I am Sending offers to treatment group which are intend to increase the purchasing capacity of the customers.
At every month end I am computing the effectiveness of treatment by computing:
(Average purchases of target group - Average purchases of control group) * Target Group size
Now Consider every month there people leaving from the target and control group also, people are getting added every month. So, I have two questions:
- How do I make changes to the Control group to be able to compute uplift accurately?
- What should be the validity (expiry) of control group/ when should I refresh my control group?