Most financial transactions involving check deposits and cashing at automatic teller machines are often associated with high risk. Although conventional security measures for many automatic teller machines record financial transactions, the recorded events are typically viewed after suspicious or criminal activity has already occurred. Therefore, a financial institution may not know in advance whether requested financial transactions are of suspicious or fraudulent intent. The present teachings relate to a method of scoring customers prior to authorization in a manner so as to identify suspicious behavior or suspicious activity. The method also relates to marking suspicious individuals by storing related transaction information, such as biometric data, in a database for purposes of verifying identification information and scoring the validity of future financial transactions.