A method is provided for flagging email messages sent to a user containing inquiries directed to the user comprises defining a natural language model for a set of inquiring phrasal forms in a first data store; defining a list of terms used to identify a first user having an email address managed by a host system in a second data store; accessing the host system to retrieve an email message sent to the email address; parsing a textual content of a body of the email message to generate one or more natural language tokens each corresponding to a text string in the body; accessing the first data store to identify each of the one or more natural language tokens that matches with an inquiring phrasal form; accessing the second data store to determine if any of the text strings corresponding to the one or more natural language tokens that match with an inquiring phrasal form includes a term from the list of terms; and flagging the email message if any of the text strings in the message body corresponding to the one or more natural language tokens that match with an inquiring phrasal form includes a term from the list of terms.