@victor  In 24.2 the old “control group by customer property” setting is gone; Impact Analyzer only lets you define control groups as a percentage split, so you cannot directly say “all customers where X=Y are control” in that screen anymore. The supported way now is to use segments or strategy logic on top of Impact Analyzer. You can first build a customer Segment filtered by your property and then run the experiment only for that segment so that, in practice, you are testing on exactly the subset you want. Another option is to keep the %-based control group in Impact Analyzer, but in your NBA strategy add extra conditions on the customer property so that only certain customers are eligible for treatments and others effectively act as a fixed control. This gives you similar behavior, but it is modeled in strategy/segments instead of in the experiment definition. At the moment there is no out-of-the-box checkbox or field in Impact Analyzer to define the control group purely on a customer property in 24.2. If that older behavior is critical for you, the only “clean” route is to combine segments/strategies with the new experiments and log an enhancement request with Pega Support.