When Fraud Detection Technology Does the Wrong Thing . . .
Posted by Andre Edelbrock on Wed, Jan 27, 2010
Last week, the BBC reported that bank anti-fraud systems are blocking donations to Haitian relief following the earthquake that has brought the small, impoverished country literally to its knees.
Although it is a well-known technique of fraudsters to use stolen cards at charitable sites to make "micro-donations" to test whether a card is still active and usable, using an indiscriminate block to refuse legitimate donations in a time of extreme need is surely an immoral and insensitive use of technology, and a severe unintended consequence of a security measure taken to protect the banks' own interests. I'm sure they did not mean to stop money from getting to the relief effort, but the reality is, we can do a lot better.
Anti-fraud solutions for card-not-present fraud are infinitely more sophisticated than this today. It's relatively easy to identify risk both statistically and behaviorally (preferably in combination), and in extreme situations override rules can be programmed quickly. It might be an extra expense for the bank to contact the card customer and do a live fraud check, but especially in this case where the world is trying to reach out with an empathetic hand, that's exactly what they need to do, because automatic transaction blocks to the Red Cross are going to leave banks with yet another big raspberry on their collective faces.
So, let's call for some common sense. It's bad enough when a sledgehammer rule costs you a bit of profit by falsely rejecting a legitimate customer; it's devastating when it could cost lives and prevent help from getting to where it's needed.
Please, if you haven't contributed yet, consider making a donation to Haitian relief. This link connects you directly to the Red Cross, and lists several legitimate charities that are participating in the direct immediate support that is desperately needed.
http://www.redcross.ca/article.asp?id=000043&tid=016