Hi Team. We have an experiment for a feature that we'd now like to roll out slowly to the remaining users. We dont want to remove the feature from existing users though. This part of the docunentation says that changing the percentage of the people seeing each treatment is a safe change to make given "they are simply increasing the number of users exposed to the experiment. There is no effect on existing users in the experiment.". However the next paragraph talks about having sticky bucketing enabled, implying that unless sticky bucketing is enabled, users may be reshuffled if the percentage is changed. Is there a risk that increasing the percentage of treatment groups will change users in the "feature on ( or B )" group allocations to feature off if sticky bucketing is not enabled?

Hi Chris, good morning 👋🏾 Since you're just increasing traffic this will be a safe change with or without sticky bucketing enabled. The hashing algorithm will make sure that the variation already assigned to users will remain consistent

Excellent. I just got to that conclusion too, so that's great to get confirmation. Thankyou @flaky-noon-11399

My absolute pleasure to assist 🌻