This is a really outstanding explanation by Mike.m with respect to the problem-solving strategy, I’d like to re-emphasize those first three steps:
1) ID the conclusion
2) ID the support (premises)
3) Think of why the support doesn’t guarantee the conclusion.
Perhaps the most important thing you can do on any LR question that has one is identify the conclusion. Why? Because the premises have to be taken as true, but the conclusion doesn’t. EVERY flaw question comes down to this: “How is it that the conclusion might be false, even if all these premises are true?” Being familiar with commonly tested flaws is tremendously important, but even if it’s one you’re unfamiliar with, that’s what the question comes down to – How can the conclusion be wrong, even if the premises are all true?