This week Oracle announced the release of a cross-channel fraud management capability for bank's that draws on the database company's intregrated real-time capabilities. This is a step in the strategic direction we call for in the Javelin Customer Driven Architecture model that we released in a report a couple of months ago, but these systems will only be succesful if 1) they are designed to involve "the other ID fraud victim" (the consumer or small business) and 2) they have some capability to incorporate a prioritized and factual view of fraud by type (channel, product, etc). A few years ago I participated in interviews with a sizeable number of risk and fraud executives within the the largest banks in US, Europe, Asia and elsewhere, and I was shocked to see the degree of confusion between the unique goals of *integrating* fraud management and *prioritization* of fraud managment (by type of fraud, channel, etc). You see, if an organization integrates fraud managmeent across all their lines of business but still lacks a prioritized model then they've only succeeded in integrating a collection of unorganized efforts. This would be akin to getting John, Paul, George and Ringo to play together back in, say 1960, but finding that they don't have the basis for agreeing on what exactly they should be playing and how their playlist needs to change over time. I need to also say that among US financial institutions, Comerica, CapitalOne and USAA were among the leaders in both integration and prioritization of fraud management efforts. As banks integrate fraud management practices across lines of business and channels, it becoomes even more important to have a factual and near-real-time model for setting priorities by type of fraud.