1st question: is it marketable? No point taking it any further if not. There may be patents on similar ideas that make that impossible. It may be because there's only one person in the world who wants one. It may be because everyone wants one, and everyone makes one, and you can't develop something different enough to make it stand out. Or that the price has been forced down to where there's no margin except for the multinational who's making them in China for pennies..
That's the first step. Develop your idea with a view to selling it, because, in its' initial form, it is not likely to be something you can sell.
Next you want to turn it into something you can make, and figure out how much you can make it for, research your market and see if it'll stand the price you have to pay. Repeat until something promising drops out (or doesn't).
Next you need to refine the design into something more detailed. Draw up a requirement spec. What is it, what features will it have, what are its' important features and selling points?
Now break it down into the tasks you need to carry out to make it. Specify each so that you can give a bit of paper to someone and tell them to make it. Figure out how long each of these steps is going to take, what it'll cost, what the major challenges are going to be. Do a project plan so you know what the critical path through the project is, and refine this so you can get it to market in a timescale and cost that will work. By this point, you will have had to go back to the beginning a few times and refine it, take it back to the market research stage and see if it'll still work as a product, and so on.
You might want to build a prototype and perhaps use that to test the design, and test the market. You're into the phase that TheBoy described now - except that it doesn't need to be as good as it can be, just good enough.
Not the end of the story yet, but my dinner's ready
