This could just as easily be written as, "Why doesn't enterprise software work?" If you're just paving cowpaths, you're not getting the best out of any technology improvement.

Agreed, it's just incredible how many times I've seen it. The person implementing is so afraid to push the boundaries that they end up with the same basic platform with an updated UI at best.

This is frequently the case with IT driven projects. Someone sees a new technology and has to promise the users, "You can still do what you always did" to get approval. This is also what happens when you replace things for technical reasons. (We have to upgrade Sybase, the old system doesn't support the new version of Sybase, so we'll implement something new...)

The best cases of improving enterprise business is when the project is driven by a respected line manager, who can challenge people to make tough decisions. These are frequently managers with CS or technology backgrounds who have migrated over. It can't be a, "Put this person on the project because they're not useful elsewhere." It has to be, "Put this person on the project, because it's the only way to guarantee it'll succeed."