Checkpoints: Actors

Have you found all the actors? That is, have you accounted for and modeled all roles in the system's environment? Although you should check this, you cannot be sure until you have found and described all the use cases.
Is each actor involved with at least one use case? Remove any actors not mentioned in the use-case descriptions, or any actors without communicates-associations with a use case. However, an actor mentioned in a use-case description is likely to have a communicates-association with that particular use case.
Can you name at least two people who would be able to perform as a particular actor? If not, check if the role the actor models is part of another one. If so, you should merge the actor with another actor.
Do any actors play similar roles in relation to the system? If so, you should merge them into a single actor. The communicates-associations and use-case descriptions show how the actors and the system interrelate.
Do two actors play the same role in relation to a use case? If so, you should use actor-generalizations to model their shared behavior.
Will a particular actor use the system in several (completely different) ways or does he have several (completely different) purposes for using the use case? If so, you should probably have more than one actor.
Do the actors have intuitive and descriptive names? Can both users and customers understand the names? It is important that actor names correspond to their roles. If not, change them.

Rational Unified Process