If you end up with an idea first (because you always find ideas in the shower and very rarely problems), try to take a step back and figure out what the problem is that your idea is going to solve. From there you can do your iteration.
I was thinking of creating a service where you can blast out an idea. If enough people signup and maybe categorize themselves, you can possibly get valuable feedback on whatever idea you just had. What do you think?
I believe in this casey you're supposed to show 3 products that could potentially solve the problem. Pick 3 axes that you think could be important. For giggles this one is going to be fast, cheap, or good (quality).
Draw the 3 axes and put each solution on it. You want a spread, so don't jumble all the solutions together. Are they together? Go back and change one of your axes such that they break apart (you don't necessarily need to change the ideas).
So now that you have your fast/cheap/good axis set or your makes toast/slices bread/gets you a drink automatically axis (to illustrate the point that this could be anything and applied to any set of solutions), and your ideas are sufficiently spread out, go show the people all 3 ideas. Tell them all 3 ideas then ask for feedback on any of them. You'll have 3 categories of feedback: love it, hate it, neutral. You want 1 or 2 because that means you've elicited a response which you can use to further your iteration. By giving them 3 solutions they can now evaluate your [great idea] in context with other things that can also solve their problem. They'll generally give you better feedback than "I'd use it" because they'll tell you "I wish #1 could be more like #3 in this way."
