While my initial reaction is similar to yours, keep in mind that what's being proposed by the article is the concept that no one sets goals well enough that they should be used the way they are right now. This is a premise I don't agree with, mostly because I think their data is really shallow, but the point is that they are proposing a reality in which you cannot find a decent dataset of people setting goals well within their company.