Key findings of the survey
- Cost of fraud: The LexisNexis Fraud Multiplier measures how many times more a fraudulent transaction costs over the amount of the lost transaction value. The Fraud Multiplier is greatest for businesses in Malaysia (3.57), compared to those in the three other countries. Indonesian businesses come second at 3.52, followed by Philippines and Singapore at 3.46 and 3.45 respectively.
- Impact on revenues: Regionally, the average overall costs of fraud is 1.75% of revenues. By country, fraud cost is highest as a percentage of revenue for Philippine businesses (2.03%), followed by Malaysia (1.93%), Indonesia (1.66%), and Singapore (1.57%). Fraud costs are biggest as a percentage of revenues for e-commerce businesses (2.50%), followed by retail businesses (1.64%) and financial services firms (1.56%).
- Growing mobile commerce enhances risk of fraud: Mobile commerce, or m-commerce, is growing quickly throughout all four countries surveyed. Mobile web browsers have historically been considered less secure, and fraudsters are now increasingly targeting mobile apps at a global level through click flooding and botnet attacks. Managing risk in mobile commerce, however, remains secondary to providing a streamlined user experience.
- Businesses are not optimally fighting fraud: Few businesses in the region track fraudulent transactions by both channel and payment method in an effort to understand weak points within their anti-fraud strategy. They are inadequately tracking fraudulent transactions by both channel and payment method to understand weak points within their anti-fraud strategy.
- Verification of customer identity is a challenge: Verifying customer identities emerged as the top challenge in all countries for businesses serving customers through online and mobile channels.Factors differ among countries and between channels, including the volume of malicious botnet orders placed simultaneously, the rise of synthetic identities, the need for balancing speed of approval against customer friction, and the accessibility (or the lack thereof) of tracking tools.
Three main factors were found to contribute to increased fraud risk, including:
- Mobile—rise of botnet attacks and consumer risk behavioural patterns such as using open WiFi networks
- Cross Border—Uncertainties regarding transaction origination and new payment methods
- Digital—Speed of transactions, bots testing stolen cards and use of synthetic identities (comprised of real and/or fake personal information—making them extremely difficult to detect with traditional identity verification / authentication solutions).
Identifying and reducing fraud:
- Less than 50% (Singapore) of transactions (42% in the Philippines, and <40% for Indonesia/Malaysia) are flagged by an automated system. Of those flagged, the majority (67% for the Philippines, 64% for Singapore, 60% for Malaysia and 40% for Indonesia) are sent for manual review of which 32% (Indonesia/the Philippines), 35% (Malaysia) and 37% (Singapore) are declined in error. The current fraud processes used by businesses are resulting in lost revenue opportunity and a negative customer experience.
- The use of more advanced, multi-layered fraud mitigation solutions can help. Optimal results can be achieved by layering the right combination of solutions to meet threats from specific types of channels and transactions.
- Due to increasing fraud complexity multi-layered solutions are necessary to address the various and concurrent risk. Fraud tools need to authenticate both digital and physical criteria as well as identity and transaction risk.
All the survey respondents believe that selling digital goods increases the risk of fraud but that the channel is too important to ignore. Even though m-commerce transactions carry a higher level of fraud risk, all businesses are increasingly accepting these transactions primarily due to customer convenience, the need to remain competitive, and to help grow business. Therefore, reducing fraud would not only help to drive revenues but also increase customer loyalty.
