I am not at the point in which I can use data and analysis as Jacob Marley suggests so I am taste profiling instead. I have several test ciders going now and I am about to implement two 1 gallon batches in which I will use the same press, the same yeast, and left in the same conditions. I am going to pull one of the lees early and let the other sit on it for at least 6 months. I will bottle them after 6 months and let them sit for a week. After that I am going to bottle age both batches for up to a year while opening one or two on the way.
The rest of my current test batches have a variety of variables. I really like the idea of trying to use the lees to extend some tastes but its unclear to me as a beginner what the results will be.
I do have one result from an unintended lees experiment. I have been saving yeasts and lees in a 1 gallon jug. I do not wash the yeast, I simply store it with a bung and airlock. I always leave a layer of cider on top of the yeast cake. I have re used this yeast cake a few times but before I do I always smell and taste the cider layer on top to make sure it hasn't gone bad. The taste has been out of this world. Smooth and very rich. It was after tasting this saved yeast cake cider top layer that I thought about extending my fermentations without racking.