My last post on corporate cultures got me thinking a little more about businesses. Years ago I use to read at least major parts of business management books, the ones that are always feature the hot trend of the year, usually published by some prof a the Harvard Business School. The ones where they look at a company or a bunch of companies and then generate data of some sort to prove their idea of why the companies succeeded or failed. The last conclusion then is that everyone should be doing what was discovered.
Completing ignoring all the issues associated with a lack of a control study, the big problem is that they never play their results forward. If they really believe that X is the secret to unlocking value in a company, then find some company and install X in it and see what happens. It's no different that what we do in science and engineering: if you think you have the answer, then run some experiments to check it.
Personally, I think that running a business is about the most complicated human undertaking ever attempted. As such, you can always find data to support anything you want to believe. For comparison, consider the 210 reasons for the Fall of the Roman Empire", all of which are supported by data. The real test is when you have to work with your results.
People always complain that weather forecasters have such an easy job since they can be wrong over half the time and still keep working. I'd say business profs have it easier yet since they never even try to see if they are right or wrong. They just need to keep cranking out new books.