That most beautiful, fresh, lively, and spontaneous, yet the most frustrating, unforgiving, and nerve-wracking medium of watercolor is arguably the most difficult, not the easiest, medium to master (or even to become semi-competent for that matter).
Approaches and techniques also seem to have their paradoxes: the look of spontaneity is often the result of much pre-planning and yet of quick handling of the medium; oftentimes what brings out the forms into light is to paint the negative spaces instead; and, unlike in oil painting, to create wonderful highlights, one must leave the paper untouched!
In general, watercolor is a medium through which the painting develops like the old Polaroid photographs, working from light to dark. In this way it is much similar to pencil drawing, but with color added. Being by necessity a value study, watercolor can be helpful in learning to see values correctly (Value is the quality of light or dark of a thing, a shape, or a space), even if liberties are taken with color.
Let's assume we are working from a photo reference to create a watercolor painting, and let's also assume that we know we have a good composition--that it looks great from across the room and without detail, because regardless of detail, it has a wonderful arrangement of lights and darks throughout. It can be very helpful to have a grayscale image of the same photo, as removing all colors, it gives only the values from which we determine the order of execution. In my approach I plan out my steps based upon the layering of values. First thing is to determine where the lightest lights are--the highlights and the highest values. Generally it's a good idea to delineate these shapes extremely lightly so that lines may be erased later. I find that masking these areas usually leave blotchy or poorly defined shapes, so generally I don't use it.
From there I look for the next value down, and with this value I create the main shapes of the composition. From this point on, with an eye on the grayscale image, I work step by step down the value scale to achieve the finished painting. In theory.
The rest all comes with practice, goofs, outright failures and, now and then, some "happy accidents" as Bob would say...
A few pointers:
Beware of going down in value too soon. Any edge in the finished painting will be determined by the darkest value of the area. Usually, if an edge is touched by another wash, it will soften or even become obliterated.
Once the steps are planned out, the brushwork ought to be done fairly quickly, without too much thinking.
Beware of "chasing edges", a very common habit especially seen in larger washes. Depending upon all factors involved, once a wash is laid down, and imperfections become apparent, it is best to leave them alone and let it dry. Going back up into a wash to "fix" a slight edge generally makes a worse edge. It is best to practice making washes on another piece of the same kind of paper until confidence is felt.
Similarly, always have that extra piece of the same paper at the side to test a color, a brushstroke or an effect.
Some haiku on the subject:
Think of it like chess:
planning many moves ahead;
where the medium itself
is your opponent.
Always an attempt,
so try it again.