as well as the comments below, I would add 1. Get a visual prototype going as early as possible - even if its screen mockups. Your interpretation of what the c

as well as the comments below, I would add 1. Get a visual prototype going as early as possible - even if its screen mockups. Your interpretation of what the client wants and what they tell you they want are often completely different, and many times they don't really know at all. Once they can see something, then requirements get a lot better. 2. Once you have a better idea on what you will be building, work out in your head how you will build it. Think about all the parts and do the bits that you haven't done before, or don't know how to do and build them FIRST. This will avoid showstoppers down the road - you may not be experienced enough to know if it's something you can't / shouldn't do, or if extra help / libraries are needed to complete the job. So once you have a clear idea on what you are building, and have done test functions to do the hard bits - then it's time to turn on the music and build the rest.