This paper analyses the demographic, socio-economics and banking specific determinants that influence the risk of fraud in a portfolio of credit cards. The data are from recent account archives for cards issued throughout Italy. A logit framework is employed that incorporates cards at a risk of fraud as the dependent variable and a set of explanatory variables (e.g. gender, location, credit line, number of transactions in euros and in non euros currency). The empirical results provide useful indicators on the factors that are responsible for potential risk of fraud.