Why Does DKIM Pass but Fail DMARC Alignment?
Published 2026-11-14 · Reviewed 2026-11-14 · 6 min read · By the Ingenious AI team
Is someone impersonating your domain right now?
Run the free DomainGuard scan: lookalike registrations, MX-armed variants, certificate-transparency and CAA checks for your domain, in one report.
Run my free DomainGuard scanThe short answer
DKIM passes but DMARC alignment fails when the domain in the DKIM signature — the d= value — is not your organisational domain. The signature is valid; it just vouches for somebody else, usually your sending vendor's domain instead of yours. DMARC needs at least one aligned pass, meaning SPF or DKIM must both succeed and match the domain in the visible From header. The fix is almost always configuration: have each sender sign with your domain (or a subdomain of it), publish the public key they give you, and confirm alignment in your aggregate reports.
Two different questions: valid, and aligned
DKIM answers a cryptographic question: was this message signed by whoever holds the private key for the stated signing domain, and did the signed parts survive transit unchanged? A pass means yes. DMARC then asks a second, separate question: is that signing domain the same as — or under — the domain your customers see in the From header? A newsletter platform signing as its own domain gives you a perfectly valid signature that DMARC counts for nothing.
This distinction explains the most confusing line in authentication reports: dkim=pass sitting next to dmarc=fail. Nothing is broken and nothing was forged. The vendor simply signed as itself. Until you configure custom DKIM for your domain with that vendor, every message it sends depends on aligned SPF alone to pass DMARC — a fragile arrangement that breaks the first time the vendor changes its bounce handling.
Strict versus relaxed alignment
DMARC's default is relaxed alignment: the signing domain and the From domain must share an organisational domain, so mail from news.example.com can align with a signature for example.com. Strict alignment requires an exact match. Most organisations should stay on relaxed — it is what makes legitimate subdomain sending workable — and understand that relaxed is also why a signature from a subdomain you control counts, while one from a lookalike never can.
You set the mode in the DMARC record with adkim=r or adkim=s (and aspf= for SPF). Changing to strict before every sender signs with the exact From domain is a reliable way to break your own mail, so treat strict as a finishing move for mature, tightly inventoried estates, not a starting setting. For most small and mid-sized domains, relaxed alignment plus p=reject is the correct end state.
Finding every misaligned signer
Your DMARC aggregate reports contain the whole story: for each source, they show the DKIM signing domain alongside the result. Sort by signing domain and the pattern jumps out — your own domain's signatures aligned, vendor domains not. Mail-testing services that show full Authentication-Results headers are the second tool: send yourself a message through each platform and read the d= value in the DKIM-Signature and the auth-results line.
Make a simple table as you go: sender, d= value, aligned or not, and the vendor setting that controls it. The usual fixes are labelled differently in every dashboard — custom DKIM, domain authentication, branded signing, DKIM keys — but they all do the same thing: the vendor signs with your domain using a key whose public half you publish in your DNS, often via two CNAME delegations rather than a pasted key.
The traps: forwarding, mailing lists, and key problems
Some alignment failures are not vendor settings. Forwarding breaks SPF almost by definition, because the forwarding server is not in your SPF record, which is exactly why aligned DKIM matters: a valid aligned signature survives forwarding intact. Mailing lists that rewrite bodies or footers break the signature itself. If a source shows dkim=fail rather than a pass with the wrong d=, suspect content modification, a selector mismatch, or a key that was rotated on one side only.
Key hygiene causes its own class of failures. A selector published in DNS but no longer used by the signer, a new key signed with before DNS propagation, or a 1024-bit key where receivers increasingly expect 2048 — all produce intermittent results that look random until you line up the selectors. Keep a register of selectors per vendor, and rotate deliberately: publish the new key, wait for propagation, switch the signer, then retire the old key.
Verify alignment before you enforce DMARC
Alignment is the gate your whole DMARC rollout passes through. Before stepping any policy toward quarantine or reject, confirm from at least a week of aggregate reports that every legitimate high-volume source shows either aligned DKIM, aligned SPF, or preferably both. Dual alignment is the resilient target: if one mechanism breaks in transit — a forward, a list, a vendor change — the other still carries the message through DMARC.
Build a light monthly habit around this: skim the aggregate summary for any source whose aligned pass rate dropped, and any new signing domain you do not recognise. Alignment failures found in reports are a nuisance. The same failures found by your customers, after enforcement, are an incident. The reports are free and they arrive whether you read them or not — read them.
Five actions to take today
- Pull a week of DMARC aggregate reports and list every DKIM signing domain (d= value) sending as your domain.
- For each vendor signing as its own domain, enable custom or branded DKIM so it signs with your domain instead.
- Publish the vendor's public key or CNAME delegations in your DNS and verify with a real test message before moving on.
- Check Authentication-Results headers for dkim=pass with your domain on every platform, including invoicing and support tools.
- Require dual alignment (SPF and DKIM) for your highest-volume senders before raising DMARC to p=reject.
Worked example: Brightpath Tutoring fixes three silent signers
Brightpath Tutoring, a fictional education company with 3 domains monitored, assumed its email was authenticated because every test showed dkim=pass. Its aggregate reports told another story: its CRM, booking platform, and receipt service all signed with vendor domains, leaving DMARC dependent on SPF alone for 61 percent of its mail. Configuring branded DKIM with all three vendors took one afternoon. Aligned DKIM coverage rose from 39 percent to 98 percent of messages, and when the booking platform later changed its bounce domain and broke SPF alignment, not a single booking email failed DMARC — the signatures carried them.
FAQ
Can DMARC pass on DKIM alone, without SPF?
Yes. DMARC needs one aligned pass, from either mechanism. Aligned DKIM alone is enough, which is why it is the mechanism that survives forwarding and why fixing signing domains matters so much.
Where do I find the d= value for a message?
In the DKIM-Signature header of any received message, and summarised in the Authentication-Results header. Your DMARC aggregate reports also list signing domains per sending source.
My vendor only offers DKIM on a higher plan. What then?
Then that sender depends on aligned SPF: make sure its return-path is a subdomain of your domain so SPF aligns, and weigh the upgrade cost against the risk. Note it in your sender register as a known single point of failure.
Does a subdomain signature align with the main domain?
Under relaxed alignment, yes — a signature for mail.example.com aligns with a From domain of example.com. Under strict alignment it does not. DMARC defaults to relaxed for both SPF and DKIM.
Why do some messages show dkim=fail instead of a wrong domain?
A fail means the signature did not verify: the signed content changed in transit, the selector's public key does not match, or the key was rotated inconsistently. That is a different fault from misalignment and is fixed with the vendor's key setup, not your DMARC record.