What a new payee checks. Against what came before.
- Not read here
The face
Not asked.
- Checked
The number and SIM
The payee is added from the bound SIM, outside any cool-off after a change.
- Checked
The phone
The bound phone, with no screen-sharing or remote-control app running.
- Bound here
The payee
Scored before it is saved: how old the account is.
- Checked
The session
An active call or a shared screen while the payee is added.
Where fraud shows. And the decision.
What DeepID reads. One call, one decision.
The decision. Cool off, in the example.
A new device, SIM or payee
Why
First-time payee, account 2 days old and linked to 14 other accounts, added while the customer was on a call.
The customer sees
“You can send up to ₹10,000 now. You get the full limit in 12 hours.”
A good customer
A known payee is untouched. A new one may come with a short note, or a lower limit for a few hours.
Your FRM receivesCOOL_OFF · first-time payee · linked to 14 accounts · active call
What to measure. Agreed before the pilot.
Where they start- MulesFound before the first payment.
- Cool-offsPayees started on a lower limit.
- LossMoney that left to a mule.
- FrictionGood payees slowed down.
Use cases. Step by step.
Rules it supports.
India, Malaysia and the Philippines
A payee and mule graph across accounts and devices, with recipient risk scored before the payment is sent.
The rules, market by marketSingapore, Malaysia and the Philippines
Cool-off and hold decisions after a new device, SIM or payee, or when a payment looks coerced, set per market and journey.
The rules, market by marketIndia, Singapore, Malaysia, Australia and Saudi Arabia
Device, behaviour, scam and payee signals read at login, new payee and payment, returned as one decision with its reasons.
The rules, market by market
Start with sign-up, login or payments.
Pick one, test it on your own traffic before you commit, and add the others on the same SDK when you’re ready.