How much detail does my prototype actually need?
You can build a prototype at any level of polish, and now the real judgment call is how much polish a question actually needs. This trips people because more fidelity always feels safer, as if a more finished thing gives a more trustworthy answer. Often it just costs more time to learn the same thing a rough version would have told you. Let the question set the fidelity, not your pride. If you are testing whether an idea makes sense, a rough sketch is enough, and a polished build is waste. If you are testing how something feels to use, then the feel has to be real, and roughness would give you a false answer. Ask what would actually change your decision, and build only to that line. When unsure, start rougher than feels comfortable. You can always add fidelity once a cheap version has earned it, but you cannot get back the days spent polishing something you should have thrown away. Match the effort to the question, and let most of your prototypes be lighter than your instinct wants them to be.