A scammer calls a customer and tells them their savings are at risk. To protect the money, they must move it to a “safe account.” The customer opens their banking app on their usual phone, logs in, and sends the transfer.
In an authorized push payment (APP) scam, the person sending the money has been deceived about who they're paying or why. To the bank, everything on the sender’s side can look legitimate because it really is the customer making the payment.
That’s why the bank needs to understand who’s getting the money as well as who’s sending it.
The receiving account may have links to earlier scams or suspicious activity. Evidence of a connection between the sender and recipient can add another clue.
Looking at these alongside the sender’s activity gives the bank more information about the payment itself.
If other customers have reported scams involving that supposed “safe account,” the bank has a reason to question the transfer. The customer may be genuine and approving willingly, while the receiving account has links to earlier fraud reports.
Prior fraud reports are one example. Suspicious patterns in how money comes in and goes out, links to accounts associated with fraud, or suspected mule activity can also raise concerns.
A customer’s successful login won't reveal those issues at the receiving end.
Finding a concern about the recipient also changes what the bank needs to resolve. Asking the customer to approve the transfer again may only confirm that they're following the caller’s instructions.
Scammers know the warnings banks use and can explain them away before they appear. The fake bank employee may tell the customer to expect an alert and ignore it. Another approval doesn't resolve the concern about where the money is going.
The account’s history is only part of what the bank can learn from the receiving end. It can also assess evidence of a connection between the two people. That can be useful even when the bank has never seen them pay each other before.
Friends go to dinner. One pays the bill, and the others send their share. It may be their first payment through this bank, even though they've known each other for years. The bank sees a new recipient, but the customer is paying someone they already know.
But evidence of that real-world relationship can give the bank a strong signal for distinguishing legitimate payments from potential scams.
In Incognia's analysis of more than 300 million transactions, we found that:
To look for those connections beyond payment records, Incognia’s Physical Social Graph uses the places devices have visited over time, together with persistent device intelligence. The assessment requires usable history about both parties.
Our Trusted Locations technology identifies places visited frequently over time. Repeated activity at some of the same precise locations can suggest a connection. Indoor location precision helps distinguish activity in one apartment from activity in a neighboring apartment. That helps separate repeated visits to one precise place from unrelated activity elsewhere in the same building.
Of course, people send money to strangers all the time. A customer buying from a seller they've never met may have no observed physical-world relationship with them. That doesn't make the payment a scam.
A real-world relationship doesn't automatically make a payment safe, but it adds a strong signal that banks wouldn't get from looking at the sender alone.
A payment can come from the genuine customer, on their usual device, and still send money to a scammer. Looking at the receiving end adds information about where that money is going: prior fraud reports, suspicious account activity, and evidence of a physical-world relationship between the parties.
Combining those signals with sender-side checks gives banks more to assess before the money moves. It can help them identify scam risk that the customer’s identity, device, and approval alone may miss, and decide when to let a payment through or intervene.