March 4, 2008, 12:00 AM

E-retailers finding more effective tools at fighting fraud, study says

Risk-scoring models, or fraud screens, designed to block specific types of fraud experienced by individual retailers were cited most often as the most effective risk management tools in a study by CyberSource Corp.

Risk-scoring models, or fraud screens, designed to block specific types of fraud experienced by individual retailers were cited most often as the most effective risk management tools in a study by CyberSource Corp., a provider of online risk-management technology and services.

The risk-scoring models, configured using an individual e-retailer’s historical data related to online orders, are designed to block, and alert merchants to, the types of orders that have been known to be fraudulent on their web sites. Because criminals often revise their strategies to compensate for retailers’ risk management efforts, however, the risk-scoring models also need to be continuously updated, CyberSource says. “Since fraudsters learn over time and vary their strategies, we typically find that most risk-scoring models need regular tuning with new analysis and data in order to maximize their effectiveness,” CyberSource says in its 9thy Annual Online Fraud Report, which is based on a survey of 318 online merchants last fall.

The survey found that risk-scoring models scored the highest when respondents were asked to name the most effective risk-management tools. 57% of respondents said risk-scoring models were among their three most effective tools; 50% cited “out-of-wallet” applications, which seek to authenticate online buyers by asking them for information not likely found in a stolen wallet, such as a mother’s maiden name; and 48% cited IP geolocation applications, which identify the geographic location from where an online buyer is placing an order.

The ratings on effectiveness, however, did not directly coincide with the rates of deployment for each tool. Risk-scoring models ranked third among the most deployed, at 39%; IP geolocation ranked fifth at 37%; and out-of-wallet applications ranked sixteenth at 6%.

The most widely deployed risk management tool, address verification, has an 80% usage rate, but less than a third of e-retailers consider it among their most effective tools, the study says. The second most widely deployed application, the card verification number, has a 74% usage rate but considered among the three most effective tools by 24% of merchants.

Following are risk management tools listed in the study, with the percentage of respondents that had deployed the tools at the time of the study and the percentage that planned to deploy them within a year of the study:

  • Address verification services, 80%, 4%
  • Card verification number, 74%, 8%
  • Postal address verification services, 39%, 8%
  • Retailer-specific risk-scoring models, 39%, 3%
  • IP geolocation information, 37%, 9%
  • Negative lists, 36%, 6%
  • Order velocity file, 35%, 7%
  • Automated decision/order screening, 34%, 7%
  • Customer behavior analysis, 29%, 8%
  • Fraud screen/general industry models, 25%, 7%
  • Verified by Visa/MasterCard SecureCode, 25%, 15%
  • External passive verification sources, 22%, 3%
  • Chargeback management, 20%, 7%
  • Positive lists, 17%, 5%
  • Electronic case management systems, 11%, 4%
  • Out-of-wallet/in-wallet challenges, 6%, 2%
  • Customer device identification, 4%, 8%
  • Other, 9%, 3%

    Following are the risk management tools ranked by the percentage of retailers who cited them in the CyberSource study among the three most effective tools:

  • Retailer-specific risk-scoring models, 57%
  • Out-of-wallet/in-wallet challenges, 50%
  • IP Geolocation, 48%
  • Negative lists, 43%
  • Automated decision, 43%
  • Customer behavior analysis, 40%
  • Address verification service, 29%
  • Customer device identification, 25%
  • Card verification number, 24%
  • Verified by Visa/MasterCard SecureCode, 23%
  • Order velocity monitoring, 22%
  • External passive verification sources, 18%
  • Postal address validation service, 16%
  • Positive lists, 11%
  • Fraud screen/general industry models, 6%

  • comments powered by Disqus




    From The IR Blog


    Paul Dobbins / E-Commerce

    6 tips for maintaining sales during the post-holiday lull

    Use that stellar email list you built during the holidays to market new products or ...


    Jeff Sass / E-Commerce

    How brands use domains beyond dot-com to attract shoppers

    Amazon, for example, posts holiday ads to