A Payment Just Went to a Fraudster: The 48-Hour Funds Recovery Playbook
Recovery odds fall with every hour. This playbook covers what to ask your bank for, where to report, what evidence to lock down and when to notify insurers so the claim survives.
The First Hour Decides Whether the Money Comes Back
Someone in accounts payable has just realized the "new bank details" from a supplier were fake, or the CEO's urgent request came from a lookalike domain. The payment has left. What you do in the next hour matters more than anything you do next week, because funds transfer fraud recovery depends on reaching the money before the fraudster moves it on. Mule networks often split and forward incoming funds within hours, and sometimes within minutes on instant rails.
Work two tracks at once. One person handles the money: bank calls, recall requests, freeze requests. A second handles the record: evidence, reporting and insurance notice. In a small company that might be the CFO and the controller. In a larger one it's treasury plus the incident response lead. Don't let one person do both. The bank call slips while they're forwarding emails to IT.
Do these in the first 60 minutes, in this order:
- Call your bank's fraud or payments operations line. Don't email your relationship manager and wait. Ask by phone for an immediate recall and a request to the beneficiary bank to freeze the funds. Write down the time, the name of the person you spoke to and the case reference.
- Stop further payments to the same beneficiary account. Hold every pending payment where bank details changed recently (say, the last 90 days) until each is verified by phone, using contact details from your own records.
- Keep the circle small. That includes the supplier's email address, which may be compromised. If the attacker is reading a mailbox, a warning helps them move the funds faster.
- Open an incident log with timestamps. Your insurer, bank and police will all ask for the timeline, and you won't remember it accurately later.
Wire Recall Requests and Beneficiary-Bank Freezes: What to Ask For
A wire recall request asks the receiving bank to return funds. It is a request, not an order. The beneficiary bank generally needs the account holder's consent or a legal basis, such as a police request or court order, to debit the account. That's why speed and a police reference number matter so much. Your bank can't pull the money back on its own authority.
Be specific on the call. Ask for:
- A formal recall citing fraud, sent through the fastest channel the payment used. For cross-border payments this often means a SWIFT gpi recall through the Stop and Recall Service where both banks participate, which lets the recall follow the payment rather than chasing it bank by bank.
- A request that the beneficiary bank freeze or ring-fence any remaining balance while the recall is reviewed.
- The payment's tracking status: whether it has been credited, and to which bank, including any intermediaries.
- Your bank's internal fraud case number, and a direct contact on shift overnight if the payment crossed time zones.
Recall mechanisms and realistic windows differ by rail and country. The table shows general patterns, not guarantees; confirm specifics with your bank.
| Payment type | Recall route | Practical window | Notes |
|---|---|---|---|
| Cross-border SWIFT wire | gpi Stop and Recall, or a cancellation request through the bank's messaging channel | Hours to a day or two before funds are usually withdrawn | Intermediary banks can slow things down; ask for tracker status |
| US domestic wire (Fedwire) | Bank-to-bank recall; law-enforcement coordination for qualifying international transfers | Same day is best; the first days are critical | File with the FBI's IC3 promptly so a police reference exists |
| UK Faster Payments / CHAPS | Sending bank's fraud team contacts the receiving bank under industry procedures | Minutes to hours | Reimbursement rules for authorized push payment fraud mainly protect smaller customers; larger firms should not assume they apply |
| SEPA credit transfer / SEPA Instant | SEPA recall procedure on fraud grounds | Recall can be raised for a limited period, but money is often gone within hours | Verification of Payee checks apply to euro transfers in many cases; record whether a mismatch warning appeared |
| India RTGS / NEFT / IMPS | Bank fraud desk plus the national cyber fraud helpline (1930), which can prompt holds at the receiving bank | Report the same day | Hold requests depend heavily on how fast the report is filed |
| Brazil Pix | Special Return Mechanism (MED) through your bank | Raise immediately; the window is limited | Funds can be blocked in the receiving account pending analysis |
Red flag: if your bank says "we've sent a message, nothing more we can do," escalate. Ask whether it has phoned the beneficiary bank's fraud team directly. Those calls often decide whether funds are frozen.
Reporting Payment Fraud to Police in Every Relevant Country
Reporting payment fraud to police isn't a formality. A crime reference number is often what lets a beneficiary bank justify freezing an account. Most crime policies also expect evidence that you reported the loss to law enforcement. File in the country where your company is based, and ask your bank whether to report in the beneficiary's country too. The receiving bank answers to its own police, not yours.
| Country / region | Typical reporting channel for business payment fraud |
|---|---|
| United States | FBI Internet Crime Complaint Center (IC3), plus the local FBI field office for large international wires |
| United Kingdom | The national fraud reporting service run by the City of London Police (Action Fraud, being replaced by a new Report Fraud service; check which is live) |
| EU member states | National or regional police cybercrime units; Europol coordinates cases but doesn't generally take reports from businesses |
| India | National Cyber Crime Reporting Portal and the 1930 helpline, then a formal complaint with local cyber police |
| Singapore | Singapore Police Force reporting channels; its anti-scam unit works with banks on fund freezes |
| Australia | ReportCyber, run through the Australian Cyber Security Centre |
| Canada | Local police plus the Canadian Anti-Fraud Centre |
| United Arab Emirates | Emirate-level police e-crime platforms, such as Dubai Police or Abu Dhabi Police channels |
| Nigeria, Kenya, South Africa | National police cybercrime or commercial crime units; in Nigeria, the EFCC also handles financial fraud |
Have the essentials ready: transaction reference, amount, currency, date and time, beneficiary name, account number or IBAN, beneficiary bank and SWIFT/BIC, and the fraudulent emails with full headers. Give the police reference to your bank at once so it can pass it to the beneficiary bank.
Data-protection duties may run on a separate clock. If the attacker reached a mailbox containing personal data, regimes such as the GDPR (which sets a 72-hour regulator notification window where its threshold is met), India's DPDP framework and national breach rules may apply. Ask privacy counsel to make that call. Don't let finance assume it doesn't apply.
Evidence to Lock Down Before Anyone Cleans Up
The instinct to reset passwords, delete phishing emails and wipe a laptop is understandable. It also destroys what your insurer, forensic team and police need. Contain the threat, but preserve the evidence first.
Preserve these within the first 12 hours:
- The original emails in native format (.eml or .msg with full headers), not forwards or screenshots. Forwarding strips the header data that shows whether a message was spoofed or sent from a compromised account.
- Mailbox audit logs and sign-in logs from Microsoft 365, Google Workspace or your mail platform. Retention can be short on some license tiers, so export now. Look for inbox rules that auto-forward or hide messages; attackers often create them.
- Payment system records: who created, approved and released the payment, and when and by whom vendor master data changed.
- Call records and notes of any verification callback, including which number was dialed and where it came from.
- Bank correspondence: confirmations, recall acknowledgements and tracker status.
Then contain: reset credentials and revoke sessions for affected accounts, remove malicious inbox rules (after documenting them), and enforce MFA where it was missing. If your cyber policy has an incident response hotline, call it before hiring a forensic firm. Many policies only cover vendors from the insurer's panel or ones it has approved.
Worked example (hypothetical): say a 60-person distributor pays $180,000 to a fraudulent account after a supplier's mailbox is compromised. If IT deletes the phishing thread and the attacker's inbox rule before export, the company may be unable to show whether its own systems or the supplier's were breached. That distinction can decide which policy responds, and whether the supplier shares responsibility.
Social Engineering Claim Notification Under Crime and Cyber Policies
Legitimate losses get disputed because notice was late or went to the wrong policy. Coverage for fraudulent transfers can sit in several places: a commercial crime or fidelity policy (often with a social engineering fraud endorsement), a cyber policy (under funds transfer fraud, social engineering or cyber crime insuring agreements), and occasionally other forms for specific scenarios. Labels and definitions vary by insurer and country.
Notify every policy that could respond, in writing, through your broker, within 24 to 48 hours. Crime policies are commonly written on a discovery basis and require notice "as soon as practicable" after discovery, followed by a proof of loss within a set period, often a few months. Check your own wording. Missing these conditions is a common reason claims are contested.
A social engineering claim notification should include:
- Date and time the loss was discovered, and by whom
- Amount, currency and payment details
- A short factual description: how the instruction arrived, what verification was done, who approved it
- Recall and freeze actions taken, plus the police reference
- A statement that the investigation is ongoing and you reserve the right to supplement
Keep it factual. Don't speculate about fault or write that staff "ignored procedure." Insurers will look closely at conditions such as callback verification to a pre-established number, dual authorization, or controls described in your application. If your application said you call back on every bank-detail change, expect questions about whether that happened.
Three things that can jeopardize a claim:
- Agreeing a settlement or release with the supplier, customer or bank without insurer consent.
- Hiring lawyers or forensic investigators outside the panel without approval.
- Recovering partial funds and not telling the insurer. Recoveries reduce the claim, and policies say how they're allocated.
Sublimits are a further issue, because social engineering cover is frequently sublimited. Typical 2026 market ranges run from roughly $50,000 to $250,000 for small and mid-size buyers, with $1 million or more available to larger buyers at higher premiums. Terms vary by contract and region, so confirm with your broker. Know your number before you promise anyone a recovery.
Hours 24 to 48: Proof of Loss, Counterparties and When to Escalate
By the second day the money is frozen, partly returned or gone. Work shifts from emergency calls to building the record and deciding how hard to pursue recovery.
Build the loss file. Pull together the timeline, payment records, evidence package, bank and police references, and a copy of every notice sent. Your broker can tell you which proof of loss form the insurer uses. Start drafting now while memories are fresh, even if the deadline is months away.
Deal with the counterparty. If a supplier's email was compromised, you may still owe the invoice, and they may dispute liability. Contact them through a verified phone number, share only what's needed to confirm the compromise, and take no position on who bears the loss until counsel and your insurer have weighed in. Outcomes vary by jurisdiction and contract terms.
Decide on legal escalation. When funds are frozen but the beneficiary bank won't release them without an order, a local lawyer in the beneficiary's country can seek a freezing or disclosure order. Typical 2026 costs run from roughly $5,000 to $30,000 or more depending on jurisdiction and complexity; get quotes. The larger the frozen balance relative to your sublimit and retention, the stronger the case for court action. Ask the insurer first, since it may fund or direct recovery efforts.
Close the control gap. Put a temporary rule in place: no bank-detail change takes effect without a callback to a number already on file and a second approver. Keep it until a proper review is done.
Decision rule for the next incident: if a suspected fraudulent payment exceeds your social engineering sublimit, or crossed a border, call your bank's fraud line and your broker within the same hour. The first gives you the best chance of getting the money back. The second keeps your claim intact if you don't.
Frequently asked questions
How quickly do we need to request a wire recall for it to work?
As fast as you can. Same day is the realistic target, and within the first hour is better. Fraud networks often move incoming funds within hours, and on instant payment rails it can happen within minutes. Requests after a couple of days still sometimes recover funds, but the odds drop sharply.
Can our bank force the beneficiary bank to return the money?
Usually not. A recall is a request, and the receiving bank generally needs the account holder's consent, a law-enforcement request or a court order to debit the account. That's why a prompt police report and direct bank-to-bank fraud team contact matter so much.
Is a fraudulent transfer covered by our cyber policy or our crime policy?
It depends on the wording. Many companies have overlapping cover under a crime policy's social engineering endorsement and a cyber policy's funds transfer fraud or cyber crime section, and the cause of the loss affects which one responds. Notify both through your broker and let the insurers work out allocation. Don't pick one yourself.
Will our claim be denied if staff didn't follow callback procedures?
Not automatically, but it's a real risk. Some social engineering endorsements make verification a condition of coverage, and insurers compare what happened with the controls described in your application. Give accurate facts, avoid admissions in the notice, and have your broker or coverage counsel review the policy conditions.
Do we need to report payment fraud to police if the bank already knows?
Yes. Banks often need a police reference to support a freeze, insurers usually expect one, and data-protection or regulatory obligations may apply separately if systems or personal data were compromised. Report in your home country and ask your bank whether to also report in the beneficiary's country.
Written by the Guard Policy Cloud editorial team. Prices, features and rules change often - confirm details with the vendor or a licensed professional before you decide.