Cybersecurity

Invoice fraud and CEO fraud: how to protect your payments

How business email compromise works, the warning signs of invoice and CEO fraud, the payment controls that stop it, and what to do if money has already left.

An email arrives from a supplier you have paid for years: same logo, same signature, same tone. Their bank has changed, it says, so please use the new account details for this month’s invoice. Or a message from your managing director asks for an urgent, confidential transfer before the end of the day. Both are classic forms of business email compromise: fraud that needs no virus, only a believable email and a payment process with one weak step.

This guide, for finance teams and office managers, explains how the scams work, the warning signs, the controls that stop them, and what to do if money has already gone.

This article offers practical guidance as of October 2026. It is not legal advice: for questions of liability or recovery, consult your lawyer and your bank.

How business email compromise works

Business email compromise (BEC) covers any fraud where someone pretends, by email, to be a person you trust in order to have money sent to them. The two most common versions target your payments directly.

Invoice fraud: fake supplier bank details

The fraudster writes as one of your real suppliers and announces new bank details, or sends an invoice that looks genuine but carries a different account number. Sometimes the supplier’s own mailbox has been hacked, so the email comes from the right address, in the middle of a real conversation, with a real invoice attached. Nothing looks wrong, because almost nothing is.

CEO fraud: the urgent request from “the boss”

Here the fraudster poses as your director, or a lawyer acting for them. The request is urgent and often secret: an acquisition, a tax matter, a deal “nobody must know about yet”. The pressure is the point: it pushes the person who pays to skip the usual checks.

The same tricks appear as an “employee” asking HR to change their salary account, or an adviser’s new account for fees, often from a lookalike domain such as supp1ier.com.

Warning signs of CEO fraud and invoice fraud

Each of these is a reason to stop and check before paying:

  • A change of bank details, especially by email alone, or an account in a different country from the supplier.
  • Urgency or secrecy: “today”, “before the bank closes”, “don’t discuss this with anyone”.
  • A request that bypasses the normal process: no purchase order, no approval, a new payee.
  • A sender address or reply-to address that is slightly different from usual.
  • A change in tone: a director who never writes to you suddenly does.
  • An invoice that does not match an order you placed.

The most dangerous sign is no sign at all: from a compromised mailbox, the email is perfect. That is why the controls below rely on process, not on spotting mistakes.

Invoice fraud prevention: the controls that work

Call back on a number you already know

Every bank details change and every unusual payment request is confirmed by phone, on a number you already had: your supplier file, a contract, the official website. Never the number in the email. The call should confirm the full new account details, not just “did you send us an email?”.

Dual approval above a threshold

Above a set amount, two people approve each payment: one prepares it, another releases it. Urgent requests from directors included. A real director will accept a two-minute check; a fraudster will push back.

A written procedure for bank details changes

The bank details change scam succeeds when anyone can update a supplier’s account in a few clicks. Write a short procedure:

  1. Changes are accepted only on a signed form or letterhead, then confirmed by the call-back above.
  2. The person who records the change is not the person who approves payments to it.
  3. Your system logs who changed what, and when.
  4. The first payment to new details is flagged for an extra check, or kept small until the supplier confirms receipt.

Many banks now check that the payee name matches the account holder (Confirmation of Payee in the UK, Verification of Payee for euro transfers in the European Union). Heed their warnings, but treat them as an extra layer, not a replacement.

Protect your own domain: SPF, DKIM and DMARC

Three email settings make it much harder for anyone to send emails that appear to come from your domain:

  • SPF (Sender Policy Framework) lists the servers allowed to send email for your domain.
  • DKIM (DomainKeys Identified Mail) adds a digital signature that proves an email was not altered.
  • DMARC (Domain-based Message Authentication, Reporting and Conformance) tells receiving servers what to do with emails that fail those checks, and sends you reports.

Start DMARC in monitoring mode, fix the services that send on your behalf (invoicing tool, newsletter, CRM), then move to a policy that quarantines or rejects. The limit: this stops spoofing of your domain, not a lookalike domain or a hacked supplier mailbox. Multi-factor authentication on every mailbox, covered in our guide to cybersecurity for SMEs, protects you from being that compromised supplier for someone else.

Let your systems do the checking

Controls that live only in people’s heads fade on a busy Friday. A well-configured ERP (enterprise resource planning) system can enforce them: invoices matched to an order before payment, supplier bank details locked behind an approval step, alerts on any change. The principle is the one behind a project led by our founder, Anass Chabbab, in his previous roles: at Innova Home in Morocco, moving from paper and Excel to one order-to-invoice tool, where each invoice follows from a recorded order, saved more than 50% of the time per order.

If you are already automating invoice processing with AI, add one rule: the agent compares the bank details on each invoice with your supplier file and sends any difference to a person. The extraction can be automatic; the decision to pay new details should never be.

Payment-control checklist

Use this as a review of your current process. Any “no” is a gap worth closing.

Control Yes / No
Bank details changes are confirmed by phone, on a number already on file
The person who changes supplier details cannot approve payments to them
Payments above a set threshold need two approvers
Urgent requests from directors follow the same approval rules
Every invoice is matched to an order or contract before payment
Changes to supplier records are logged and reviewed monthly
SPF, DKIM and DMARC are set up, and DMARC is beyond monitoring mode
Multi-factor authentication is on for every mailbox
Everyone knows who to call, and in what order, if a fraud is suspected

What to do if money has already left

Speed matters more than anything else. In order:

  1. Call your bank at once and ask it to stop or recall the transfer, then confirm in writing.
  2. Report it to the police and to your national cybercrime service. As of October 2026, for example: ReportCyber at cyber.gov.au in Australia, the FBI’s Internet Crime Complaint Center (ic3.gov) in the United States. Check the current channel in your country.
  3. Keep the evidence: save the emails with their full headers, the invoices and bank confirmations.
  4. Secure your mailboxes: change passwords, check forwarding rules and recent logins, and switch on multi-factor authentication where it is missing.
  5. Warn the supplier or person impersonated, using a number you already know: their mailbox may be the one compromised.
  6. Inform your insurer if you have cyber or fraud cover, and ask your lawyer about your position before discussing who bears the loss.

Frequently asked questions

How can I tell if a supplier’s change of bank details is genuine?

Not from the email itself, so don’t try. Call the supplier on a number already on file, not one in the message, and ask them to read out the new account details. If they confirm, record the change following your procedure and note who you spoke to. If they know nothing about it, you have just stopped a fraud: warn them, since their mailbox may be compromised.

Does DMARC stop CEO fraud?

Partly. A DMARC policy that rejects or quarantines failing emails stops fraudsters sending messages that appear to come from your exact domain, which blocks the crudest CEO fraud. It does nothing against a lookalike domain, a personal email address (“I’m writing from my phone”) or a hacked mailbox. That is why call-backs and dual approval remain essential.

Is the bank responsible if we pay a fraudster?

It depends on the country, the payment type, your contract with the bank and the facts, so there is no general answer, and reimbursement rules have been changing in several countries. Call your bank immediately, report the fraud, keep every document, and ask a lawyer to assess your position.

In short

Business email compromise succeeds because the email looks right and the payment process trusts it. Call-backs on known numbers, dual approval, a written procedure for bank details changes, and SPF, DKIM and DMARC close most of the gaps. Your systems can enforce those rules so they do not depend on memory. And if money has left, call your bank first, then report and secure.

Want your systems to check bank details before anyone presses “pay”? Book a free 30-minute discovery call: we look together at where your invoices and approvals could be locked down.

Related servicesERPAgentic AI

Have a project in mind?

Let’s talk about it on a 30-minute discovery call, free and with no obligation.

Book a discovery call

Further reading

All articles