# Email Going to Spam: Troubleshooting for Your Own Domain

*2026-11-11*

> Almost all email from your own domain that lands in spam comes down to SPF, DKIM, DMARC or reverse DNS. Here is how to read the header and find the fault in the right order.

A message from your own domain that lands in spam almost always means the recipient cannot verify that it really came from you, or that the sender has a poor reputation. It is a configuration problem: SPF, DKIM, DMARC and reverse DNS decide whether your message gets read.

## Start by reading the header

Send a test message to an address you control and show the full header.

- `Authentication-Results`: what the receiving server concluded, such as `spf=pass`, `dkim=pass`, `dmarc=pass` or a fail.
- `Return-Path`: the address the message was handed over with, which should belong to your sending domain.
- `DKIM-Signature`: holds `d=`, the signing domain, and `s=`, the key used.

If `dmarc=fail` appears while SPF and DKIM pass, the domains do not match. More on that below.

## Check SPF: is your sending server listed?

SPF is a TXT record stating which servers may send email using your domain ([Wikipedia: Sender Policy Framework](https://sv.wikipedia.org/wiki/Sender_Policy_Framework)). It starts with `v=spf1` and usually ends with `~all` or `-all`.

- `-all`, fail, means messages from other servers should be rejected.
- `~all`, softfail, means they are accepted but flagged as suspicious — useful while testing.
- `?all`, neutral, effectively says nothing.

The common mistake: you send through a provider or external service whose servers are not in the record, so the message gets `spf=fail` as soon as it leaves them. Add them with `include:` or `ip4:` instead of publishing a second SPF record next to the first. The record must not get too heavy either: one check may perform at most ten DNS lookups, and if the policy needs more, the check aborts and the domain behaves as if it had no SPF at all.

## DKIM: the signature has to travel all the way

DKIM signs the outgoing message with a private key, and the recipient verifies it with the public key in DNS ([Wikipedia: DomainKeys Identified Mail](https://sv.wikipedia.org/wiki/DomainKeys_Identified_Mail)). The sender signs, not you as domain owner, so it is your provider's key that must sit under `selector._domainkey.your-domain`. Three things go wrong:

1. The key sits under the wrong name; the `s=` in the signature must point to the record that exists.
2. A forward or mailing list rewrites subject or body and kills the signature; DKIM survives forwarding better than SPF, but not rewriting.
3. You changed provider and left the old key, so messages are signed with a key nobody can verify.

Compare `d=` with your domain. If it says something else, you send through a service signing with its own domain, and DMARC alignment must be solved another way.

## DMARC: this is where delivery is decided

DMARC builds on SPF and DKIM and is published as a TXT record on `_dmarc.your-domain` ([Wikipedia: DMARC](https://en.wikipedia.org/wiki/DMARC)). It holds a policy for messages that cannot be verified:

- `p=none` leaves the message alone but sends reports to the address in `rua=`. A good starting point.
- `p=quarantine` puts suspicious messages in spam.
- `p=reject` rejects them.

The decisive part is alignment: the domain in `From:` must match either the SPF domain or the DKIM domain, and a message can pass both SPF and DKIM and still fail DMARC if another domain was checked. Start with `p=none` and read the reports for a few weeks before tightening: tighten too early and you lose messages from services you forgot to list, and a rejected message never arrives, not even in spam.

## Reverse DNS: owned by whoever owns the IP

The PTR record translates an IP address into a hostname. It lives in `in-addr.arpa` and is recommended by RFC 1912 for every IP address reachable on the internet ([Wikipedia: Reverse DNS lookup](https://en.wikipedia.org/wiki/Reverse_DNS_lookup)). Since February 2024, Google's sender rules require valid forward and reverse DNS records for delivery to Gmail.

The PTR is set by whoever owns the IP, meaning your hosting provider, not you as domain owner. If you send through a shared IP without reverse DNS, nothing in your own zone helps. Ask the provider to point the PTR at the hostname your server presents itself with, then check with `dig -x` and the IP address.

## Reputation and blocklists

Receiving servers weigh what they already know about the IP and the domain. A DNS-based blocklist, a DNSBL, lets a recipient ask whether a sending IP is on a list of known spammers ([Wikipedia: DNS blocklist](https://en.wikipedia.org/wiki/Domain_Name_System_blocklist)). Getting listed usually means an account of yours sent bulk mail, or the IP is shared with someone who did. Gmail also wants a spam rate below 0.3 percent in Postmaster Tools, and that number falls when you stop mailing addresses that never open anything.

## Content and habits that send email to spam

- A short message with one link, from a sender the recipient has never heard of, looks like advertising.
- Text baked into images cannot be read by the filters.
- The same message to many recipients in blind copy from a private mailbox looks like bulk mail; use a sending service.
- A `From:` address that does not match the sending domain, for example Gmail in From and your domain in Return-Path.
- A subject line in capitals with exclamation marks and words like "free" or "last chance".

## Troubleshooting when email goes to spam

1. Read `Authentication-Results`: pass or fail on SPF, DKIM and DMARC?
2. Do `Return-Path` and `d=` match your sending domain?
3. Is the sending server in the SPF record, under ten DNS lookups?
4. Does the DKIM signature verify, under the right `_domainkey` name?
5. Is there a DMARC record on `_dmarc`, and what policy does it state?
6. Does the IP have a PTR record pointing back to the server's hostname?
7. Is the IP on a blocklist, and does another account send from it?
8. Resend after each change, or you will not know which one helped.

## When the settings are right and the message still gets stuck

Then it is usually the recipient's own filters or simply first contact: a new domain has no reputation to lean on. Ask the recipient to mark the message as "not spam" and reply to it; a reply weighs more than ten technical adjustments. Check at the same time that the domain is not about to expire, since mail from a domain without working DNS looks like forgery however well written it is.

The records themselves are covered in [Konfigurera SPF, DKIM och DMARC för egen domän](https://bushmail.cc/artiklar/konfigurera-spf-dkim-dmarc-egen-doman), and if receiving is what fails, start with [MX-post för egen domän](https://bushmail.cc/artiklar/mx-post-egen-doman). More on mailbox security is in [Är min e-post säker?](https://bushmail.cc/artiklar/ar-min-epost-sakker-testa-din-epost-sakkerhet).
