An improved subset-difference method is provided. The improved method uses the value of a current content key to help generate the requisite difference keys. The requisite difference keys are then used to encrypt the next content key which will be delivered only to users who are supposed to remain in the group. Users who have the current content key are then able to generate the requisite difference keys which they can then use to decrypt the next content key. Using the decrypted next content key, the users are then able to continue to receive contents. Since previously revoked users do not have the current content key, they are unable to determine the next content key and thus are prevented from receiving future contents.