Hello - is there a way to run 2 experiments in an ...
# ask-questions
c
Hello - is there a way to run 2 experiments in an exclusive way? Say I have 2 experiments that we want to run at relatively low exposure (say 20%), and we want both to be totally independent from each other, so that no user is exposed to both experiments simultaneously. Is this possible via some growthbook config? Or we’d have to roll our own logic to determine that at runtime?
h
Definitely. The easiest way is to use namespaces. See the docs here: https://docs.growthbook.io/features/rules#namespaces
c
oh that’s helpful thanks @happy-autumn-40938. And to be clear - we would use the same namespace for both experiments right?
follow-up question: then say in the future, one of the experiments we decide to roll out to everyone with a Forced rule. In that case we would still be able to effectively roll out to everyone without a problem, and not only to the allocated part of the namespace, right? Because the namespace is at the experiment-level not at the Feature-level.
h
If you want the experiments to never overlap, then you would use separate namespaces for each experiment. If you roll out using a forced rule instead of an experiment, that would apply across all traffic and would not be namespace-specific.
👍 1