Richard Borcherds, a field medal winner, stated that he approaches new topics by doing a lot of examples before trying to understand the theorems and such. There is no "right" way to do math, contrary to what OP claims.