Can someone send email pretending to be you?
· 4 min read
Email was designed in a more trusting age. When a message arrives saying it’s from the accounts address at your domain, the system that receives it has no built-in way of knowing whether that’s true. Anyone with a little know-how can put your address in the “from” line.
That’s how a customer ends up with an invoice that looks like yours, from an address that looks like yours, with somebody else’s bank details on it.
Three small additions to your domain close that gap. They have unfriendly names, but the idea behind each one is simple.
The three records, without the jargon
SPF is a guest list. It’s a public note on your domain that says “these are the only services allowed to send email for us”. If a message arrives from somewhere that isn’t on the list, the receiving system knows something is off.
DKIM is a signature. Every email you send gets an invisible stamp that only your genuine email service can produce. The receiving system checks the stamp. If it’s missing or wrong, the message didn’t come from you, or it was altered on the way.
DMARC is the instruction. SPF and DKIM only raise a flag. DMARC tells the receiving system what to do about it: let the message through anyway, put it in junk, or reject it outright. It also asks other mail systems to send you reports, so you can see who is sending email in your name.
You need all three. The first two do the checking. The third makes the checking count.
What this stops, and what it doesn’t
It stops someone sending email from your exact domain. With DMARC fully switched on, a fake message using your real address is turned away by any mail system that checks, which today is nearly all of them.
It doesn’t stop two other things, and it’s worth being clear about them.
Lookalike domains. Somebody can still register a domain that’s one letter different from yours and send from that. Your records only protect your real domain.
A real mailbox that’s been taken over. If an attacker has the password to a genuine account in your business, the email they send is genuine as far as any check can tell. The defence against that is multi-factor authentication on every account, which we cover in is Microsoft 365 secure by default?
So these records aren’t the whole answer to email fraud. They are the part that protects your name, and the part your customers can’t do for you.
Why it has to be done in stages
Here’s the catch. Your email service isn’t the only thing that sends email as you. Think about:
- The contact form on your website
- Your invoicing or accounting software
- Your newsletter tool
- The booking system, the CRM, the scanner in the corner
If you tell the world to reject anything that isn’t on the guest list, and you’ve left one of those off, its emails stop arriving. Invoices vanish. Nobody tells you, because the people who didn’t get them don’t know they were sent.
So the safe way is in three steps:
- Watch. Set DMARC to monitor only. Nothing is blocked. You just collect the reports for a few weeks and see everything that sends in your name.
- Fix. Add each genuine service to SPF, and switch on DKIM for it. Anything you don’t recognise is either forgotten or fraudulent.
- Enforce. Once the reports are clean, move to “junk”, then to “reject”.
Skipping straight to step three is how businesses break their own email and decide the whole thing was a bad idea.
Two things that catch Microsoft 365 users out
DKIM isn’t on for your own domain until you turn it on. Microsoft 365 signs email from day one, but with Microsoft’s default domain, not yours. Signing with your own domain means adding two records and flipping a switch.
“We have SPF” isn’t the same as “we’re protected”. Plenty of domains have an SPF record that was set up years ago, and no DMARC record at all. Without the instruction, the fake may well still be delivered.
Don’t forget the domains you don’t use
If you own other domains, perhaps an old trading name or the .com you bought to stop anyone else having it, those can be faked too. Because they never send email, the fix is quick: publish records that say so. The government has a short guide on protecting domains that don’t send email.
Is it worth the effort?
The government’s latest Cyber Security Breaches Survey found phishing was the most common type of attack on UK businesses, reaching 38% of them in a year. Email is where most small businesses actually lose money, and a faked invoice in your name damages a customer relationship even when you did nothing wrong.
There’s a practical reason too. The big mailbox providers increasingly expect these records from anyone sending in volume. Without them, your genuine email is more likely to end up in junk.
One habit that helps either way
Whatever your records say, agree one rule with your customers and suppliers: if anyone’s bank details change, the change is confirmed by phone, on a number you already had. It costs nothing, and it stops this kind of fraud even when the technology hasn’t.
If you’d like it done for you
Setting this up properly, in stages, without losing a single invoice, is part of our email security work. We find every service that sends as you, get each one set up correctly, and only then switch on enforcement.
Sources: the NCSC’s guidance on email security and anti-spoofing and the Cyber Security Breaches Survey 2025/26.