A paper published for the 2015 Network and Distributed System Security (NDSS) Symposium in February describes user interface modification techniques to address liar buyer fraud, and the results of experiments assessing the potential for these to reduce ecommerce fraud losses.
Liar Buyer Fraud, and How to Curb It authors Markus Jakobsson, Hossein Siadati and Mayank Dhiman describe "liar buyer" fraud, how traditional anti-fraud technology fails to curb this problem, and details the results of experiments of proposed alternative techniques to reduce the problem.
The authors explain that liar buyer fraudsters are generally not repeat fraudsters, but are otherwise honest people who are first-time offenders that act fraudulently as the result of temporary poor judgement. This manifests itself in claims that deliveries were not made. It is believed that at least a quarter, and as much as half, of direct fraud affecting some organisations is the result of liar buyer fraud.
The ideas considered by the authors for their research involve changes to the user interface that promote user honesty:
- Disclosure that the customer's computer/device has been recognised
- Disclosure of the customer's location (e.g. IP address, post code or location map)
- Production of statements by the delivery person
- Simplifying methods of goods return
- Forcing the customer to make a promise
- Attending to angry and upset customers carefully.
The research focused on the first two of these and found they have a significant reduction in customer's willingness to file false claims. The other options look promising and, perhaps with the exception of the third approach, could be undertaken by real-world retailers in A/B/N testing.
Posted on: 03 March 2015 at 07:48 hrs