Customer Problem It is estimated that the nation's banks experience over $10 billion per year in attempted fraudulent check deposits. The daily challenge for a large bank is to identify a few thousand risky deposited checks out of hundreds of thousands of normal ones. It is extremely important to create highly effective rules that detect fraudulent checks with minimum impact to good customers. Due to constraints in the bank’s data infrastructure and analytic capabilities, the existing check deposit fraud detection rules have not been adjusted for many years and therefore are ineffective. This leads to higher fraud losses and large numbers of good customers impacted by false alerts. When holds are falsely placed on normal check deposits, experience shows that customers may leave the bank. Customer Solution ZSA built a predictive model that evaluates the risk of each check. The model assigns each check a score. The score takes into consideration approximately 19 variables about the account and its transactional history. Analysts using a custom dashboard can easily set up a score threshold to generate any desired number of daily alerts. Our solution prevents $10 million fraud loss annually.