Memento provides next-generation technology and solutions that enable financial institutions to rethink and improve the way they combat fraud and manage compliance. Memento customers realize unmatched business value and rapid ROI.

bank fraud forum

Alert Throughput: Managing or Drowning?

November 17, 2009 by David Hood
0 comment(s)

Paul McCormack's recent post on the High Cost of Fraud Operations got me thinking about how banks and credit unions struggle to keep up with fraud. Specifically how to manage the large number of alerts that is created daily. One of the important factors for loss prevention is the maximum number of alerts a team can review in a given time period. It's not difficult to get a sense for the throughput of any given loss prevention department.

 

The implications are obvious. If the bank or CU in question has fewer or an equal number of alerts than their throughput, they can effectively manage the current number of alerts (get through them all in a given time period). If the # of alerts is greater than the throughput, the institution is forced to shift employees to concentrate on the "fraud of the day". This firefighting mentality often results in less than desirable outcomes for the bank or credit union such as missed fraud or employee burnout.

Going back to our formula, what are the variables we can change to improve the throughput. According to Paul, the # of employees is fixed, so let's leave that constant. That leaves the average # of alerts an employee can handle in a given day. To improve the productivity of an employee, the bank or credit union must provide tools that aid the job of quickly dispositioning an alerted item. Certainly a worthwhile goal.

But banks and credit unions can also take steps to make sure the total number of alerts produced is roughly equal to their throughput. To do this, they can reduce the number of total alerts by tweaking the detection system. One strategy is to increase thresholds to reduce the total number of alerts but this has a negative impact on the amount of fraud caught. Another strategy is to reduce false positives, which if done correctly does not impact fraud detection. By reducing false positives, analysts maintain their productivity but have fewer alerts to review. This allows banks and credit unions to align the current throughput of their loss prevention teams with the total number of alerts and to get into a more manageable mode of fraud prevention.

What is your current throughput for loss prevention? Does it match the number of alerts you are getting on a daily and weekly basis?


Make a Comment

* = Required
*
*
*
*