*Hearts iconoclast already*
This is actually a really good question, and I have to admit that I'm drawing on my paper-writing/journalism experience to answer it, because forming a good philosophical question is, to my mind, just the same as forming a good question of any other sort.
(Oh, iconoclast, you forgot to add that we have to add plenty of caveats, on the 'off' chance we're wrong about something...
So I suppose it depends on from angle one is coming from. I'll assume that we're 'advising' someone who has a vague idea that they have a question about something philosophical, but nothing much more than that.
I'd say the process goes something like this:
1) Inkling that something is not as one had presumed.
2) meditate on said inkling, determining what its nature is.
3) Pin down what happened/what one heard that caused one to question one's position (helps in the information gathering process).
4) Ask oneself, "What is it that I'm confused/concerned about?" and answer using as many precise words as possible
5) At this point it's a matter of turning it around making the question as precise as one can for one's purposes. For example, it's one thing to ask "Is there a God?" (unanswerable and not much of a conversation starter, really...to vague) and quite another to ask "If there is a God, and that god is an omnipotent, benevolent being, what would have to be true about the world and that being?" (gives plenty of parameters for discussion and gives us something firm to work with with regards to actually finding some answers. Maybe not the answer to life, the world and everything, but maybe we'll have a better idea at the end of the discussion when a benevolent, omnipotent being is a being we ourselves can believe may exist).
To summarize, I guess, the goal in forming a philosophical question is to make it as precise as possible, giving it parameters to work within and something firm to discuss.