For me I know the fundamentals well enough that I can kind of winnow it down to a couple movements that would be good to start looking around in, but I'm not sure if it's a strategy you can use when you're just breaking in. You first have to know the history, then you can determine how original your ideas are.
The problem is that you often can't really find this through Google, unless you know the name we've given it (or happen to phrase your question in almost the same way someone else did on StackExchange). That's why it is so valuable to have a good human search engine to talke to (say, a professor who knows their field really well).