On the other hand, you do not need to run a scientific study to make every UX decision. There is a lot of "information" (in an abstract sense) in a person's int

On the other hand, you do not need to run a scientific study to make every UX decision. There is a lot of "information" (in an abstract sense) in a person's intuition, experience and common sense. Using a small usability group to validate is quite sensible.