Pre-settlement payment verification is a control that checks a payment, who is sending it, who is being paid, how much, and why, and confirms that a real, authorized person approved it, before the money is released and settles. It is the difference between stopping a fraudulent payment and reporting one, because on modern rails a settled payment cannot be pulled back. This is the model RankShield Financial is built on: verify the payment and the approval at the last controllable moment, then seal a record of the decision that anyone can check. The need for it comes from how final payments have become. The Clearing House’s RTP network is credit-push and final once it settles, and moved a record $8.62 billion in a single day in 20261, while instant rails now clear in seconds. When money moves that fast and that permanently, any control that runs after settlement is a report, not a defense. This guide defines pre-settlement payment verification, contrasts it with the fraud scoring most systems rely on, lays out the framework it checks against, and explains what it does and does not do, including the honest boundary that it never takes custody of your funds. The short version is that verification before release is the one control that still changes the outcome on an irreversible rail.
What pre-settlement payment verification is
Pre-settlement payment verification is the practice of confirming that a payment is legitimate and authorized before it is released, rather than scoring it for risk after it has settled. It answers a specific set of questions at the last moment you still control: is this the right payee, is this the amount and purpose you intend, and did a real, authorized person approve this exact payment. If the answers hold, the payment goes; if they do not, it is held. The name is literal. The check happens pre-settlement, in the window before the money becomes irreversible.
That window is the whole point, because it is closing. Once a payment settles on a wire or an instant rail, there is no clawback, and even ACH offers only a narrow return window on unauthorized debits and none on a credit you originated. A control that runs after settlement can raise an alarm, help a recovery attempt, or feed a report, but it cannot change what already happened. Pre-settlement verification moves the check to the one place it still alters the outcome. It is not a new rail or a new bank; it is a verification step inserted into the authorization path you already use.
It is worth being precise about what pre-settlement verification is not, because the category is new enough to be confused with older tools. It is not fraud detection that scores a transaction and hopes to flag the bad ones; it is not a bank control that matches a payment to a file you already submitted; and it is not insurance that pays out after a loss. It is a check, run at the point of release, that confirms a payment is what you intended and was approved by someone allowed to approve it. Everything else in this guide follows from that definition.
Verify before versus score after: a verdict you can check, not a score you must trust
The dominant approach to payment fraud is scoring: a model rates a transaction’s risk from patterns like device, velocity, and anomaly, and flags the ones that look wrong. Scoring is genuinely useful against intruder-style fraud, where behavior is abnormal, but it has two structural limits for authorized-payment fraud. It runs on statistics rather than facts, so a payment a deceived but genuine employee pushed looks normal and scores low. And it produces a private probability you have to trust, not a checkable result. You cannot independently confirm why a score was what it was.
Pre-settlement verification is different in kind. It checks facts, the payee against a trusted record, the approval against an authorized signer, rather than estimating risk, and it produces a verdict rather than a probability. That verdict is a concrete, recorded statement: this payee, this amount, this purpose, approved by this person, verified before release. The distinction that matters most is trust. A fraud score is something you have to believe; a verification verdict is something you, and an auditor, insurer, or partner, can check. That is the line RankShield draws, a verdict you can check rather than a score you must trust, and you can see how the verdict is produced. For a direct comparison with a scoring-based incumbent, see RankShield versus Accertify.
The framework: payer, payee, amount, purpose, and an authorized approval
Verification is only as good as what it checks, so pre-settlement verification is defined by a specific framework rather than a vague promise of safety. The version RankShield verifies against has five elements, four that describe the intent of the payment and one that proves a human stood behind it. Each element maps to a category of fraud it is designed to stop, which is what makes the framework a control rather than a slogan.
Read as a whole, the five elements answer the only question that matters at release: is this the payment you actually intended, approved by someone actually allowed to approve it. Fraud scoring never asks the last part, because it cannot see intent or authority; it only sees the shape of the transaction. This framework is the linkable, checkable core of pre-settlement verification, and it is the same on a wire, an ACH file, or an instant payment, because intent and approval do not change with the rail.
| Element | The question it answers | What it is designed to stop |
|---|---|---|
| Payer | Who is sending this payment, and are they who they claim to be | Compromised or impersonated originators |
| Payee | Who is being paid, matched against a record you already trust | Vendor swaps and lookalike accounts |
| Amount | How much is being paid, against what is expected | Inflated, duplicate, or altered amounts |
| Purpose | Why this payment is being made | Payments with no legitimate business reason |
| Approval | Proof a real, authorized person approved this exact payment | A deceived approver or a single-person release |
Does it hold my money? No, and why that matters
A common and fair question is whether inserting a verification step means someone else is now holding your money. The answer is no. Pre-settlement payment verification, as RankShield implements it, is a verification and attestation layer, not a wallet, a processor, or an escrow. It never takes custody of your funds. Your bank and the payment rails still move the money exactly as they do today; the verification sits in the authorization path and returns a verdict of release or hold. When a payment is held, it is held in your own system, under your control, not swept into a third party.
This boundary is deliberate, and it is worth being honest about because plenty of products blur it. Taking custody would make RankShield a money transmitter, add regulatory weight, and create exactly the single point of control a verification layer should avoid. The role is narrower and, we think, more defensible: verify the payment and the approval, produce a checkable verdict, and seal a tamper-evident record of the decision, without ever touching the funds. The money stays yours the entire time, and the value added is certainty about the payment, not possession of it.
A verdict you can check: verifiable attestation, not a private database
What turns a verification from an internal opinion into something durable is attestation. When a payment is verified, RankShield seals a signed, tamper-evident record of the decision: the intent that was checked, the approval that was proven, and the verdict that was reached. Because the record is cryptographically signed and independently verifiable, it is not just a log entry you have to take on faith. An examiner, an insurer, an auditor, or a counterparty can confirm that the verification happened and what it concluded, without trusting RankShield’s word for it. That is the practical meaning of a verdict you can check.
Two honest qualifications belong here. First, the value of a shared verification signal grows as more members participate, so it compounds over time rather than arriving fully formed; we describe it that way rather than claiming a network scale we have not yet reached. Second, the signing is quantum-safe by construction, built to resist future quantum attacks on signatures, which is a design choice, not a guarantee that anything is unbreakable. Pre-settlement verification does not promise to make fraud impossible. It promises to move the decisive check to before settlement and to make its result something you can prove later, which is a meaningful and honest improvement over a score you simply trust. You can read more about the attestation model.
What rails pre-settlement verification covers, and how it fits your stack
Pre-settlement verification applies to any rail where a payment can be released, because the point of control is the moment of release, not the mechanics of the rail. That includes wire, where the payment is final once sent; ACH, where an originated credit has no return right and Nacha’s 2026 fraud-monitoring rules now require screening for payments made under false pretenses2; and instant rails like RTP and FedNow, where settlement happens in seconds. The check is the same on each: verify the intent and the approval before the release instruction goes to the bank.
In practice it sits in the authorization path between the person or system that initiates a payment and the rail that settles it, and it returns a release-or-hold verdict along with a sealed record. It does not replace your bank controls, your fraud scoring, or your ACH filters; it closes the gap those leave open, which is the authorized payment that looks normal to every one of them. With the FBI reporting $3.046 billion in business email compromise losses in 2025, 86 percent moving by wire or ACH3, that gap is where the money goes, and verifying before settlement is the control aimed squarely at it. If you want to add that check to your own payment flow, you can request access.
