Playing the Numbers: Assigning EFT Testing Priorities

No organization can test everything. So how do you make intelligent decisions about which tests to include in your EFT regression testing? How do you create a test plan that makes the most of your test resources? This article offers one approach to categorizing and prioritizing your regression testing of electronic payments.

What Every EFT Tester Knows

While managers at financial institutions agree in principle that EFT software testing is necessary and important, these principles are not always reflected in the time and resources allotted to testing. More often than not, testing is constrained by limited budgets, or squeezed by delivery promises and aggressive deadlines. Unlimited testing is simply not an option. The only course of action, then, is to determine the best use of the limited resources of Quality Assurance/Quality Control (QA/QC) departments. The question is: how do you determine that?

Identifying Electronic Payment Events and Evaluating Probability

You likely already have a list of events that you want to test before moving a new feature or release into production. How do you begin to prioritize those EFT tests?

• First, you evaluate your list in terms of which events potentially have the most severe financial impact on your financial institution. Put simply, which events will "take money out of your pocket,” causing losses that directly affect the bottom line? Certainly, cases of ATMs dispensing cash in excess of account balances, or of incorrectly posting cash-back amounts associated with POS debits, are extreme examples of events that can have a severe impact on your organization.
  
  Some events do not result in direct losses, but will potentially decrease or limit your profitability. Events that limit ATM or POS usage, or affect switch availability, may fit into this category.
  
  
• Secondly, you determine probability, or the likelihood of these events occurring. By examining why and how such events occur, and incorporating this knowledge into development of your QA scripts and procedures, you can reduce the probability of these events.

Copyright © 1996-2008, Paragon Application Systems