That's pretty much in Jon! For me it is usually:
Image one object on a clear night for as long as possible, process the data as it's cloudy for the next week, get a finished result that I am reasonably happy with and keep telling myself I'll add more data to but never do
