Address-Based Matching

Overview

Address-based matching works when a payment system assigns a distinct destination to a record, such as one address per customer or one address per invoice. When a transaction reaches that address, the system can identify the likely owner without relying only on the payment amount.

The method is strong when the address is unique and the network is known. It becomes less precise when addresses are reused, multiple invoices share one address, or a destination requires a memo or tag. It also needs protection against transfers on the wrong network and unsupported tokens sent to a compatible-looking address.

A production matching engine typically checks destination address, network, asset, transaction hash, confirmation state, expected amount, and timing. Static addresses may be matched to a customer first and then to an order using amount or reference data.

Address-based matching reduces manual reconciliation, but it should not automatically mark an order paid before the transaction meets the merchant’s acceptance and confirmation rules.

A production matching engine should handle more than successful exact payments. It needs rules for underpayments, overpayments, multiple partial transfers, late payments, unsupported tokens, and duplicate observations from several data providers. Matching should remain idempotent so the same on-chain transaction cannot credit an order twice.

The engine also needs chain-awareness. A transaction first seen in the mempool may disappear, and a confirmed transaction can temporarily lose confirmations during a reorganization. Address matching identifies the probable business record, while confirmation policy determines when the record becomes settled. Static customer addresses are useful for recurring deposits, but the mapping history must remain available for late transactions.