While fraud and data hacks get a lot of attention, the inverse problem of lost sales due to false declines is equally costly yet garners far fewer headlines.
“Many merchants are seeing on average around 20 percent … of all of their transactions falsely identified as fraud and declined, [sometimes even higher], so these are non-trivial percentages,” Hicks said.
While large, those losses can also often be hard to understand. As an example, Hicks highlighted a subscription box customer who pays for their box once a month, without issue, for 12 months before having their card declined on month 13, even if nothing about the card has changed. There is perhaps a small chance that the consumer’s card balance was too high, and they were declined for insufficient funds — but in the vast majority of cases, it’s just a bad decline.
There is a complex interplay of back-end features that simply need to evolve to meet the ever-changing needs of the time, said Hicks.
The Many Ways Things Can Go Wrong
Due to the card networks’ requirements, issuers have a second to approve or decline a transaction, something that banks backed by older, less agile legacy infrastructures can only accomplish by using rules-based systems that simply get it wrong a lot of the time because they are too rigid and work with too little useful data.
Fraud is increasingly getting more complex with more rules layered into the system, creating today’s situation in which $600 billion worth of consumer transactions are “stuck between the cracks” in an incredibly fragmented system.
“These decisions to approve or decline are made at the issuer level, and there are 6,000 independent issuers in America,” said Hicks. “That’s a recipe for chaos, which is another one of the drivers, and it’s why we see what we do today.”
As Hicks noted, transactions often fail for understandable, but unfortunate reasons that have nothing to do with the cardholder. Subscription service users, for example, will see their transactions declined if they are suddenly processed late at night, which is frequently identified as potential fraud in many rules-based systems. And even if the consumer had signed on at noon, many subscription services batch all of their transactions to process between 2 a.m. and 5 a.m. when server traffic is lower. That means a lot of those transactions are bounced.
It’s just one of many reasons that a consumer can get flagged as potentially fraudulent. If the merchant has been hit by a fraud ring of late, for example, that will also raise the three-digit risk score for every transaction and potentially cause the consumer to fail.
“It has nothing to do with the relationship between this cardholder and this merchant — it’s other external factors, other things going on within the merchant and the issuer, other things going on with the cardholder and other merchants they’ve transacted with,” Hicks explained. “Everything is connected, woven together into this complex system.”
It’s a system that simply needs upgrading because the losses due to false declines are only going to rise, he said.
Catching Lost Value
False declines, although not as flashy as losses to fraud, are every bit as costly to the merchant that lost out on a sale (and also alienated a consumer in the process). The good news is that the problem is manageable, Hicks said.
Ultimately, data is the solution — both in terms of having more of it and using better sets of it by creating partnerships that see merchants talking directly with the issuers to create “sort of a shared ecosystem,” enabling them to share information instead of working in rule-governed silos.
“We all want the same thing: less fraud and less friction,” said Hicks. “We want merchants and card holders to be able to transact together when they’ve decided they want to do business together, and the rest of the friction should come out.”