I was listening to an interview with Paul Feig, director of the movie Brides Maids, and the topic of test screenings was brought up. The basic idea is that once a movie reaches a certain state of production, they show it to an audience and gauge the reaction. The fascinating thing to me was that they will actually record the physical response from the audience, and then later use that information to determine which parts of the movie need more work.
The lone designer in me can absolutely see where a process like that might garner some criticism. Creativity and expression should flow from a mind, and when we rely on mass opinion to inform our art, we can end up with something that is diluted and ultimately appealing to nobody.
Alternatively, a comedy only functions when it is producing laughs, and a process like this is empirical evidence of success, and provides a guide for what needs to be worked on.
In the world of design, we have many similar procedures, where we present works in progress and then gauge client reaction. Sometimes this produces meaningful input that ultimately improves the project, and other times we can be forced to go down avenues that we do not agree with.
There are examples of great success and failure on both sides of this debate, and I feel that relying too much on either end of the spectrum is ultimately short sighted. One cannot lead without knowing the followers.