Vendor email compromise (VEC)

Vendor email compromise (VEC) is a fraud in which an attacker takes over a real supplier's mailbox, or imitates it closely, and waits for a payment to come due. The attacker then sends the supplier's customers a real-looking request to pay a different bank account.

Updated

Vendor email compromise is a form of business email compromise that is hard to see, because the mail can come from the supplier's true address.

How vendor email compromise works

Vendor email compromise works by entering a real business relationship and waiting.

  1. Get into the supplier's mailbox. The attacker steals a password, often by credential phishing, and signs in. See account takeover.
  2. Read quietly. The attacker studies invoices, customers, payment dates and the way the supplier writes.
  3. Hide. Mailbox rules move replies out of sight, so the real owner does not notice.
  4. Strike at the due date. A reply in a real thread tells the customer that the bank details have changed. See thread hijacking.
  5. Or imitate. Without access, the attacker copies a stolen thread into mail from a lookalike domain.
  6. Repeat. One supplier mailbox exposes every customer that supplier bills.

What vendor email compromise looks like

A vendor email compromise message looks like the next reply in a thread the target already trusts. In this fictional example, Northwind is the supplier, and an attacker is writing from Megan Park's real mailbox.

Example only, not a real message
From: Megan Park <megan@northwind-supply.example>
To: Jordan Ellis <jordan@trotta.email>
Subject: Re: Invoice 8841

Hi Jordan, one change before you pay invoice 8841.
We have moved banks, so please use the new account below.
[new account details]

The address is real, the thread is real and the invoice is real. The bank details are the attacker's.

Signs of vendor email compromise

SignWhat it means
New bank details inside a known threadThe one fact that changed is the one that moves the money.
A reply address or a copied address that is one character offLater replies are being steered to a mailbox the attacker owns.
A shift in tone, sign-off or wordingSomeone else is writing as the supplier.
A request not to call, or a new phone numberThe second channel is being closed or redirected.
A known thread that now comes from a near-match domainThe attacker copied the thread and is writing from a lookalike.
An invoice sent again with one line changedThe original was copied and edited.

Why vendor email compromise gets through ordinary filtering

Vendor email compromise gets through ordinary filtering because mail from a taken-over mailbox is genuine in every technical sense. The mail leaves the supplier's own mail service and passes SPF, DKIM and DMARC.

The sender has a long history with the target, so reputation checks pass as well. There is no malicious link or file. Filters in Gmail and Outlook, built to stop spam and malware, see a trusted contact continuing a real conversation.

A customer cannot fix its supplier's security. A customer can only check the request.

When an AI agent reads the mail

An AI agent that handles accounts payable or answers suppliers is exposed to vendor email compromise. The agent checks that the sender matches the supplier record, and the sender does.

An agent given a whole thread tends to treat the latest message as the current truth. Unless the agent is told that bank details never change by email, it may update the record and pay.

How to stop vendor email compromise

Vendor email compromise is stopped by verifying the change itself, because the sender cannot be trusted even when the address is right.

  • Call back on a number on file. Confirm new bank details with a known contact, using a number from the contract or the supplier record.
  • Agree the process up front. Tell suppliers in writing that bank changes are never accepted by email alone.
  • Separate change from payment. One human updates supplier details and another approves the payment.
  • Check the first payment. Any payment to a changed account gets a second approval.
  • Protect your own mailbox. Sign-in methods that resist phishing keep you from becoming the compromised vendor.

At the door, before delivery

A pre-delivery inbox judges what a message contains, not only who sent it. Trotta Mail screens the contents for payment fraud. A copy sent from a lookalike domain also meets the sender checks: SPF, DKIM and DMARC, lookalike domains and first contact. In the supplier payments use case, a notice of changed bank details from a lookalike address is held before the agent reads it.

A held message stays outside the inbox until a human releases it. An API key gets the verdict and nothing the sender wrote, so an agent cannot copy bank details it never received.

No filter catches everything. Mail from a supplier's real mailbox passes authentication, so the sender checks have nothing to find, and a well-written request with no technical tell can still be delivered. That is why uncertain mail is held or flagged for a human, and why the callback stays.

Questions

What is the difference between BEC and VEC?

Business email compromise is the wider group of scams that pose as a trusted sender. Vendor email compromise is the form in which that sender is a real supplier, often writing from its true mailbox.

How does an attacker get into a supplier's mailbox?

Often with a stolen password, taken through a fake sign-in page. A reused password or a missing second factor makes it easier.

Does DMARC stop vendor email compromise?

Not when the supplier's real mailbox is used, because that mail is authentic. DMARC set to reject does stop mail that fakes the supplier's exact domain.

How do I know if a supplier's request is real?

Call a known contact at the supplier on a number already on file and confirm the change. Do not reply to the message or call a number it gives.

Sources

Every check is on, on every plan. Free for two inboxes. No card.