Manual Review Best Practices:
Get More Data for a More Informed Decision

Thus far, our Best Practices Series has discussed how you can use the data provided by the minFraud service for better decision making during manual review.

But actionable data from minFraud starts with the inputs you include with each query.

The minFraud service requires that each query include the IP address associated with the transaction at a minimum; as best practices, MaxMind recommends you send as many data points as possible.

The more data points you provide, the better the riskScore and the more information you make available to your fraud analysts as part of the manual review process.

Prudent online merchants balance the need for collecting as much information as possible against the shopping experience of the customer providing it. If the order entry process is too onerous, customers may give up and abandon their purchase. Experts advise that minimizing data entry is one of the best ways to improve conversions.

Customers provide an email address with just about any online transaction. Including this as part of a minFraud query provides significant lift to the riskScore. In order to assess email reputation and associated risk, the minFraud service determines if the email has a history (good or bad) amongst the merchants of the minFraud Network. The minFraud service can also look for signs of high transaction velocity around that email address which may indicate higher risk due to account takeover or other factors. This is then taken into account for the riskScore calculation for the transaction.

Email address is good data to include because it is almost always available to you. For other data points, it’s sometimes just not possible to capture information you would like to have. For example, most customers are unwilling to provide credit card information when signing up for a free trial. Additionally, if you are selling digital goods, you may not be able to get a shipping address.

Expert Tips

As you assess the data you have available to you, we encourage you to think creatively about your business specifics and how you might use them to identify fraud patterns. For example, consider one targeted solution we recommended to a customer who sells electronic concert tickets: pre-populate the shipping address on the order form with the address of the concert venue. You can then flag orders for review where the geolocation of the IP address, the billing address and the location of concert venue are disparate enough to cause concern.

In a similar fashion, those selling air or bus tickets can send minFraud the departure airport or bus terminal as the shipping address. A ticket purchased using a billing address in Florida with a departure airport in Chicago fails the common sense test from a consumer point of view, and could be pulled for review.

Think about fraud scenarios common to your business and review the inputs available to you. Prioritize the data you want to collect, and weigh the cost of increased customer friction relative to the benefits of enhanced fraud detection. Provide these inputs to the minFraud service for the best possible riskScore and help your fraud analysts have the data they need to make informed decisions.