Not sure why you're being voted down, as this was the first thing I thought. Admittedly, I'm not a statistician, so perhaps someone else can explain why this kind of thing is valuable?
Two big advantages as I can see (there are undoubtedly more):
1- As experimentalists do experiments they often arrive at these same conjectures, but after years of trial and error. Throughout history there were probably mountains of data that were recorded but discarded after they were hypothesized to have no scientific value. Having a program like this can essentially narrow the search for you. It can let you get back to doing science and do less data processing, which for the majority of scientists is a godsend.
2- You can now take historical sets of data and see if you can recover new aspects of structure. Perhaps there were important features in the data that were overlooked that can now be statistically analyzed. This can lead to new research avenues.
