- Develop a good logline. It doesn't need to be perfect, but you need to know who the protagonist is, what he or she wants, what the obstacles are, and what the stakes are. If you can create one that makes people say "wow," so much the better.
- Write down everything you know. What do you already know about your characters? What scenes are in your head? What might the ending be? Which unwritten scene has stolen your heart?
- Make the promises are inherent in your work so far explicit. These are implied in your premise, the genre you've chosen, and some of what you know. In this step, consider these in terms of what the audience will expect and especially in what you will need to provide to satisfy their expectations.
- Brainstorm the possibilities. I begin this with a list of powerful questions and multiple answers to each one. My lists of possibilities go on until I surprise myself and engage deep interest. Then I write down some things that might happen in this world that is starting to take shape. Again, I keep writing things until I have elements I can't wait to write and parts I want to find answers to as I work. I like at this point to have 40-60 scenes I can report out in full sentences, but it's possible to create a useful outline with seven, if they are the right seven.
- Beat it out. I usually use Blake Snyder's Save the Cat! beats (which can be downloaded in a form for free), but I often add in explicit genre beats (like "meet cute" in romance). What the beats do is take the many scenes I have and begin to order them in a useful way. (I usually put each beat into Scrivener so I can reorder them and create a good template for drafting.) They may be defined enough at this point to be checked for story logic, but I don't require that.
What do you use your outline for? Obviously, it can be your roadmap for drafting your story. But it can also be the bones of a synopsis (which you'll have to write sooner or later). It can even help you to distill your story for a pitch or a query.
I always allow myself to skip scenes in the outline. I let them come in there own time, and sometimes I make them prove they belong. I also am not reluctant to rework the outline of a work-in-progress (keeping the original in a safe place, of course).
I never feel obliged to stick with my initial outline--though I do commit to finishing the work, even if that means returning to the outline to get find my way to "The End." For me, straying keeps me engaged, serves the muse, and, even when scenes I write get cut, adds dimensions to the story.