Apr 27, 2022
Until today, banks had to choose the “better be safe than sorry” approach for payment approvals resulting in declining legitimate transactions to reduce the risk of losing their clients’ money and credibility.
We say “had to” because, based on today’s payment authorization flows, banks simply don’t have any incentive or enough data points to analyze the authenticity of each transaction in real-time. The real problem isn’t the declined transaction. It’s the damage that comes to the bank’s reputation, and the cardholders’ dissatisfaction can take a long time to recover. But it doesn’t have to be the case.
Many other industries already know that knowledge is power. Sufficient data is vital when it comes to automating delicate real-time actions like approving transactions. If banks had more data such as individual consumer behavior and specific purchase history, they could have something to rely on when deciding to authorize or decline purchases.
Issuer banks decline transactions for a variety of reasons. Some are related to internal policy, while others are related to outdated NSF (insufficient funds) data or fraud. If a human had been reviewing these transactions and could have the time to check details or pick up the phone and talk to merchants or customers in real-time more transactions would likely be approved. Check out the following example.
Say a customer tried to buy shoes online. His transaction is declined. All he sees is “payment rejected.” He doesn’t have another card, doesn’t think to try again, and can’t pay cash. The customer is disappointed, the merchant lost a sale, and the bank lost revenue.
Let’s assume this was a technical outage issue, and the request timed out. The bank’s risk model isn’t picking up any red flags. The customer and card verifications were valid.
What happens when the 2 major players in digital payments decide to work together? Let’s play this out.
A customer tries to buy shoes online. He hit “pay now,” and the screen is now loading.
The issuing bank is about to decline the transaction but instead decides to call the merchant. They have the following exchange:
The bank tells the merchant, “Hi there. I’ve received this payment request from you. You should know I’m about to decline this transaction.”
The merchant responds with, “wait, this is a valid transaction. This is the 3rd time this customer is buying from me. It’s the same IP address and billing address. I don’t know what went wrong, but I want this sale. I’m willing to pay you for the risk you’re assuming so I can complete this sale.”
Bank: “How much will you pay me?”
Merchant: “2% of the transaction”
Bank:” Approved. Let’s get this man his shoes.”
The green checkmark reflection can be seen on the customers’ glasses as a smile spreads across his face. The merchant received a new sale email notification, and the bank received its share. What a happy ending.
The technology that enables issuing banks and merchants to have this exchange seamlessly is available. It requires no human interaction and is heavily based on shared goals.
Solutions that leverage the joint interests of all parties create win-win situations that are a no-brainer. Would a merchant pay $2 to approve a $100 transaction? Probably. That’s the logic behind this idea.
Operating by this scheme requires data, communication, and collaboration. AI-operated solutions can process and communicate volumes of data that can have the above conversation millions of times a day. The only setup required is playing with the numbers.
The best part is this process takes less than a second from the moment the customer clicks “buy now.”
Until today banks had to be the ones to protect clients, assume financial risk, and compensate the consumer if an illegitimate transaction happened to be approved. Starting today, banks can share the financial risk with merchants.