How email authentication works
Email was designed without any way to prove who sent a message, so anyone can put your address in a from field. Authentication closes that gap using records you publish in your domain’s DNS, which receiving servers look up as the message arrives.
Three records do the work together. SPF lists the servers allowed to send mail for your domain, so a receiver can check whether the message came from one of them. DKIM adds a cryptographic signature to each message, letting the receiver confirm the content was not altered and that the signing domain really authorised it. DMARC ties the two to the domain your recipients actually see, states what should happen when a check fails, and asks receivers to send you reports.
None of them says the message is honest or wanted. They only establish that the sender is who they claim to be, which is the foundation everything else is built on.
Why authentication matters
Without it, your mail competes against forgeries carrying your own name. Anyone can send an invoice or a payment request that appears to come from your business, and your customers have no way to tell the difference. With DMARC set to reject, most of those forgeries are stopped before anyone reads them.
It has also stopped being optional. The major mailbox providers now require bulk senders to authenticate, and unauthenticated bulk mail is routinely filtered or refused. This is finite, one-off technical work with a lasting benefit, which puts it among the cheapest improvements available to an email programme.
Common mistakes with authentication
The most frequent is partial setup. A domain with SPF but no DKIM, or with both but no DMARC policy, passes some checks and fails others, and the sender only discovers this when a campaign lands in spam. All three belong together.
Next is forgetting a sender. Marketing platforms, invoicing tools, helpdesks and booking systems all send on your behalf, and each needs to be included. A new tool added months later, authorised by nobody, quietly starts failing.
Third is publishing a DMARC policy of none and calling the job finished. That policy asks receivers to do nothing at all when a check fails. It is a sensible place to start while you read the reports, but leaving it there gives you no protection against forgery.
How to act on it
List every system that sends mail using your domain before touching DNS, because the list is always longer than expected. Then publish a single SPF record covering all of them, enable DKIM signing in each platform that offers it, and add a DMARC record starting with a monitoring policy while you check the reports for legitimate senders you missed.
Once the reports are clean, tighten the policy so failures are quarantined and then rejected. Verify the records with a public checking tool rather than assuming the platform wrote them correctly, and re-check whenever you change hosting, migrate email, or add a new tool. If you send marketing from a separate subdomain, remember it needs its own records — inheriting them from the parent domain is not how DNS works here, and this is the single most common reason a careful setup still fails.
