In recrystallissation you are first trying to dissolve your compound in the minimum amount of boiling solvent right? So if the solvent is boiling away the volume is constantly reducing and hence your compound might start to come out of solution before you get to filter off insoluable impurities. One way to avoid this is to use a condensor, but this is usually only the case in industrial scale recrystallisation. In the lab we use a conical flask (rather than a beaker) because the shape acts to restrict the evaporation rate... and carry out the process quickly.
As for the percentage recovery -- have a guess yourself. You know what you start with, you can weight what you end with...