Case Study: Using Historical Transaction Data to Reduce Chargebacks

The new year has arrived. With transaction history from a busy holiday season on hand, this is a great time to take a look your historical transactions with a fresh and critical eye.

Reviewing your chargeback data to identify fraud patterns is a good way to get started. In this month’s blog post, we provide a case study of an online penny auction business, which improved their bottom line by doing just that.

Penny auctions, defined by their industry as “entertainment shopping,” sell merchandise through a live, online, auction process. Customers buy bids to participate. Auction winners hope to win the auction and pay less than retail price. Losers risk spending money to bid and ending up empty-handed.

To mitigate risk and encourage auction participation, some penny auction sites offer a “Buy it now” option. This provides all auction participants who didn’t win the option to buy the item at the retail price, less the cost of the bids placed.

After their first holiday season, the penny auction’s Director of Operations at that time, Jenn Sessler, decided to take action against chargebacks. A recent operational review had shown chargebacks as a significant expense, and Jenn wanted to see what could be done to reduce them.

Her first step was to find an internal report that could help her identify common scenarios leading to chargebacks. She decided to work with their daily shipping report. “I was looking to see what we had shipped that we shouldn’t have,” she said. “What questionable transactions had our fraud system missed?”

The company’s daily shipping report supplied their warehouse with a list of customer orders placed in the last 24 hours. The report provided not only a record of what was to be shipped, but also what the customer had paid for the item, how they had purchased it (through an auction or via the Buy it now feature), as well as the retail price of the item up for bid. She combined this information with the company’s separate chargeback report and looked for patterns.

Jenn observed, “During my investigation, I learned that, for those using the Buy It Now option, where the Price Paid, (the retail price, less the credit for cost of the bids placed by the customer in the auction they lost), was more than 95% of the retail price, I was getting chargebacks.”

Having identified this pattern, Jenn took action. Prior to providing the warehouse with the shipping list, Jenn looked for Buy It Now transactions with a price paid of more than 95% of the retail price. Jenn used common Excel features (sorting and formulas) to quickly scan the daily shipping report for these transactions. On finding one, she reversed the authorization and canceled the transaction. To prevent further fraud from this party, she blocked the email address. If she found a number of similar transactions associated with the same IP address, she blocked the IP address as well.

How could Jenn be sure that the transactions she was canceling were fraudulent transactions? Jenn anticipated that legitimate customers would complain; however, she never heard from anyone whose transactions were canceled for this reason.

Jenn manually canceled transactions for four weeks to further confirm her theory and process. Once she confirmed her theory was correct, Jenn created a custom rule so that the company’s fraud system automatically rejected these transactions without any manual review cycles.

Case Study Takeaway

Take existing order reports and compare them to known fraud or chargeback data. Look for patterns specific to your business. Combining your data from different sources gives you insight into your own customers’ anticipated behaviors for good transactions and fraudsters’ anticipated behavior for bad transactions. Test your hypothesis and use your findings to refine and automate your fraud detection practices – increase automation and decrease manual review, all while maintaining a positive customer experiences for your legitimate customers.

Jenn Sessler is currently MaxMind, Inc.’s Director of Business Development.