Someone is sending email pretending to be you. Here's how to stop it.
If a customer or vendor told you they got a "from-you" email asking to wire money or change banking info, your domain is being spoofed. I'm a USAF veteran with CompTIA Security+. I'll fix it the right way: SPF, DKIM, and DMARC properly deployed, with monitoring so you see the spoofing attempts going forward.
What a spoofed email looks like
Here's an example. The bookkeeper at one of your customers gets an email from [email protected]. It says you've changed banks and asks her to update the wire instructions for the next invoice payment. The email looks normal: same signature, same tone, same logo. She replies, the criminal answers as you, and a few days later $40,000 lands in a stranger's account.
You never sent that email and you never see the reply. The first you hear of it is when your customer calls asking why you haven't credited their payment.
This is called email spoofing. The technique behind it drives Business Email Compromise (BEC), which the FBI's IC3 unit tracks as one of the largest-dollar-loss cyber crimes in the US year after year. The fix is mostly three DNS records and a process. It isn't expensive, but it has to be done right.
Email was designed in the 80s. Anyone can forge the from line.
By default, anyone on the internet, anywhere, can send an email and put whatever they want in the From field. The original SMTP protocol from 1982 has no built-in verification of sender identity. It just trusts whatever the sending server claims.
The fix is three layers added on top, all configured in your domain's DNS. That means the changes happen at your domain registrar (GoDaddy, Cloudflare, Namecheap, etc.), not in your inbox. Receiving mail servers (Gmail, Outlook, Yahoo, every business) already check these records. They just need to find them when they check, and they need them to be correct.
Those three records are SPF, DKIM, and DMARC. Here's what each does.
SPF, DKIM, DMARC in plain English
Each does one specific job. All three are required for spoofing protection to work.
SPF: who's allowed to send
Sender Policy Framework. A TXT record in your DNS that lists every server / service authorized to send email "from" your domain: Microsoft 365, Google Workspace, Mailchimp, your CRM, your e-sign tool, anything legitimately sending as you. When a receiving server gets a message claiming to be from you, it checks SPF: was the sending server on the list? If not, that's a fail.
DKIM: cryptographic signature
DomainKeys Identified Mail. Your mail server signs each outgoing message with a private key, and a matching public key is published in your DNS. Any receiver can verify the signature came from your domain and the message wasn't altered in transit. DKIM is the proof that the message came from your infrastructure and isn't a forged claim.
DMARC: what receivers should do
Domain-based Message Authentication, Reporting & Conformance. A TXT record that tells receiving mail servers: if a message claiming to be from us fails SPF AND DKIM alignment, what should you do? Let it through (none), drop it in spam (quarantine), or reject it outright (reject)? DMARC also asks receivers to send you reports of every authentication attempt. That's how you see spoofing happen in real time.
How to roll this out without breaking email
Doing this wrong can dump legitimate mail in spam folders or block it entirely. The right sequence matters.
Inventory every sender
List every system that legitimately sends email "from" your domain: your mail platform (M365, Google Workspace), marketing tools, transactional email (Mailgun, SendGrid), payroll, e-sign, helpdesk, CRM. Miss one and SPF will block its mail. I use DMARC reporting to find senders nobody remembered.
Deploy SPF and DKIM correctly
Build the SPF record to include only your authorized senders, with no extras or overlaps, and watch the 10-DNS-lookup limit. Enable DKIM signing on your mail platform (in M365, this is two CNAME records and a Defender setting). Verify both with mail-tester.com and dmarcian's free tools.
Start DMARC at p=none with reporting
Initial DMARC: v=DMARC1; p=none; rua=mailto:[email protected]. The "none" policy doesn't block anything yet. It just asks receivers to start sending you aggregate reports of every authentication attempt. Watch the reports for 2-4 weeks to find legitimate senders that aren't authenticating. Authorize them.
Move to quarantine, then reject
Once your reports show 99%+ authentication on legitimate mail, move DMARC to p=quarantine for 2-4 weeks of observation. If nothing legitimate breaks, advance to p=reject. Now anyone trying to spoof your domain has their mail rejected at the receiving server, so your customers never see the forgery.
What Idaho Proper IT does
Spoofing fix done badly is worse than no fix at all. You'll either still be spoofed OR your real mail will start landing in customers' spam folders. Done right, it's quiet, defensive, and visible only when an attacker tries something.
I'm a USAF veteran, hold CompTIA Security+, and have worked in IT support and administration roles since 2022, including hands-on work with email authentication (SPF, DKIM, DMARC).
Spoofing audit & fix
Pricing
$500
one-time audit + setup · single primary domain
Multiple domains, parked domains, and complex hybrid setups quoted separately.