Imagine we're evaluating a fresh user interface (UI) aimed at boosting conversion rates. We conduct an experiment by presenting this new UI to a randomly selected group of users.
The experimental UI shows a 5% improvement in the key performance indicator. What would you predict will occur once the new UI is rolled out to all users? Will the metric increase by about 5%, more, or less?
Note: Assume there is no initial excitement effect.