NovA refund has its own banking timetable
Why a property refund should be tracked as a new payment process, with its own trigger, instruction, banking status and proof of receipt.
This article reflects the named expert’s practical perspective. See NovAsia’s editorial policy for how material is prepared and reviewed.
A payment that reached the seller in two days does not create a two-day promise for the money coming back. That symmetry feels intuitive, but the banking process is not a video played in reverse. A refund first needs a commercial or contractual reason. Someone on the recipient side must authorise the amount. A new payment instruction is then created under that party's bank rules, and the returning funds begin their own journey to the buyer.
That distinction matters most when the conversation becomes vague. “The refund has been sent” can describe several different events: approval by the seller, submission to the seller's bank, debit from the seller's account, movement through correspondent banks, arrival at the buyer's bank, or final credit to the buyer. When a buyer is waiting for a substantial sum, those stages should not be collapsed into one status simply because everyone expects the money to come back eventually.
Start the refund clock at the event that actually happened
I would first establish what has been confirmed. If the parties have only agreed that USD 12,000 is refundable, there may be no banking event yet. If the seller has issued a transfer instruction, the banking process has started, but that still does not prove the buyer has been credited. If the sending bank can provide a transaction reference or status, that tells us more. The endpoint is the buyer's actual receipt of the agreed amount under the agreed conditions.
Swift's material on cross-border payments is useful here because it separates the in-flight movement between institutions from the final credit to the beneficiary. Swift also describes the UETR, a unique end-to-end reference used in its payment messaging environment to improve tracking and status visibility. A buyer does not need to become a payments technician to benefit from this idea. The practical lesson is that an international transfer has identifiable stages, and a refund can be followed through them instead of being described only as “sent” or “not received”.
The refund amount itself may also need careful reading. Imagine that the original property payment was USD 20,000 and the parties later agree to return USD 7,500. The refund file should state why USD 7,500 is due, in what currency the obligation is expressed, which account is to receive it, and what happens to bank charges. If someone proposes returning an equivalent amount in another currency, that is a new financial term, not a harmless administrative substitution. The exchange basis and costs can change the result, so they need explicit agreement.
I would be equally cautious about assumptions concerning the destination account. Some institutions or transaction documents may require funds to be returned to the original source; others may have different procedures. The correct answer comes from the actual bank and deal documents. A new account should never be inserted merely because it is more convenient without confirming that the change is acceptable and explainable.
Track the refund as its own file, not as a footnote to the original transfer
The cleanest record has a short chronology. First comes the event that created the refund obligation: cancellation, overpayment, failed condition or another documented reason. Next comes the agreed amount and currency. Then the bank instruction, transaction reference and any official status. Finally, the buyer's receipt and whatever confirmation closes the commercial obligation.
This structure also tells everyone when escalation is justified. If a bank has provided a processing window and the transfer remains within it, repeatedly asking the property manager for the same receipt may add no useful information. If the expected status has stopped progressing, the right next step may be a bank trace or formal query. The exact procedure belongs to the institutions handling the payment, but the decision becomes easier when the last confirmed stage is visible.
A refund can also expose an awkward difference between banking completion and transaction completion. Funds may have reached the buyer while the seller's accounting record still shows the matter open, or the seller may mark the refund as completed before the buyer has actually been credited. I want both sides of that final step visible. The banking record proves what happened to the money; the transaction record proves what the parties regard as settled.
The useful expectation, then, is not “the return should take as long as the original payment”. It is “the refund has a new start, a new instruction and a new chain of confirmations”. Once that is accepted, delays become easier to locate and promises become less likely to outrun the evidence.
Sources
- Swift — *How long does a Swift payment take?*: distinction between interbank movement and the final credit to the beneficiary; accessed 6 October 2026.
- Swift — *What is a Unique End-to-end Transaction Reference (UETR)?*: end-to-end payment identification and status tracking; accessed 6 October 2026.