My definition of "real world" problems are ones that people are willing to pay for. If nobody is willing to give you money for whatever it is that you are doing, then it probably isn't solving someone's actual problem.
The problem with this approach is that people pay much more money for limited gains they can exploit than for general gains that benefit everyone equally. Advancing academic knowledge provides gains distributed over a very large number of people, but the current system often does not efficiently allocate resources to this problem.