Good decisions rarely arrive as flashes of certainty. More often, they come from reducing the noise around a question until the real trade-off is visible.nnStart by naming the decision in one sentence. Then separate facts from assumptions, and assumptions from preferences. This simple separation prevents a confident opinion from disguising itself as evidence.nnNext, define what would make the decision successful three months from now. A useful measure is specific enough to guide action but modest enough to be checked. Finally, decide what you will revisit and when. A decision with a review date is easier to make because it does not pretend to be permanent.