SPF, DKIM, and DMARC work together to show receiving mail servers that messages from your domain are authorized. This guide explains how to inventory senders, publish each DNS record safely, roll out DMARC without unnecessarily blocking legitimate email, and establish a recurring review process for authentication failures and new sending tools.
Overview
Email authentication is a domain-management task, not a one-time checkbox. Your website host, mailbox provider, marketing platform, help desk, invoicing system, and other services may all send email using your business domain. If one of those senders is missing from your authentication configuration, messages can be delayed, placed in spam, or rejected depending on the receiving system’s checks.
Each standard has a different job:
| Record | Purpose | Where it is published | Primary maintenance task |
|---|---|---|---|
| SPF | Lists servers and services permitted to send for a domain or envelope sender | DNS TXT record at the domain root or a specified subdomain | Keep authorized services current and avoid multiple SPF records |
| DKIM | Allows a receiving server to verify a cryptographic signature added to a message | DNS TXT record at a provider-generated selector | Keep the provider’s selector and public key published |
| DMARC | Defines how receivers should handle messages that fail authentication and provides reporting options | DNS TXT record at _dmarc | Review reports, confirm alignment, and tighten policy gradually |
Authentication does not replace careful mail configuration. It should be managed alongside domain registration, managed DNS, mailbox administration, and change control for business systems. For background on related records, see DNS Records Explained.
What to track
1. Build an email-sender inventory
Before editing DNS, list every system that sends messages with your domain in the visible From address or in the underlying envelope. Ask each internal team to identify its tools and record the following details:
- Service name and business owner
- Purpose, such as employee mail, receipts, support, alerts, or newsletters
- Sending domain or subdomain
- Whether the provider supplies SPF instructions, DKIM keys, or both
- Return-path or envelope-from domain, if documented
- Expected sending volume and whether the service is still active
- Date last verified and the date it should be reviewed again
A simple inventory prevents a common failure: removing an unfamiliar but legitimate sender because nobody knows which team owns it. Treat temporary campaign tools, recruitment platforms, website forms, and transactional systems as senders until their status is confirmed.
2. Publish SPF carefully
An SPF record is a TXT record at the relevant domain, commonly represented as @ in a DNS control panel. A provider may give you a value resembling:
v=spf1 include:mail.example-provider.test include:marketing.example-provider.test ~all
Use the exact mechanisms supplied by your mail providers; the domains above are placeholders. Do not copy an example into production without replacing it. A domain should have one effective SPF record. If several teams publish separate SPF TXT records, receiving systems may treat the configuration as invalid rather than combining them. Consolidate authorized mechanisms into one record and remove services that no longer send mail.
SPF is evaluated against the envelope sender, which may differ from the visible From address. This distinction matters when a third-party service sends on your behalf. Ask the provider which domain its SPF instructions apply to, and document whether it uses your primary domain or a dedicated sending subdomain.
3. Enable DKIM for every sender that supports it
DKIM normally requires two steps: generate or activate signing in the sending provider, then publish the provider’s public key at a selector such as selector1._domainkey.example.com. The provider keeps the private key and adds a signature to outgoing messages; DNS publishes the matching public key.
Selectors are provider-specific. If two systems use different selectors, both records can exist without conflict. Copy the hostname and TXT value exactly, including long key strings and any required quotation handling in the DNS interface. After publication, use the provider’s verification control or send a test message to a mailbox where you can inspect authentication results.
4. Add DMARC in monitoring mode first
DMARC is published at _dmarc.example.com. A cautious starting point is a monitoring policy such as:
v=DMARC1; p=none; rua=mailto:dmarc-reports@example.com
Use a reporting mailbox that is monitored and appropriate for aggregate reports. Reporting addresses may require additional authorization when they use a different domain, so follow the instructions of your DNS or reporting provider. DMARC reports can contain information about sending infrastructure and message patterns; limit access to them and choose a reporting arrangement that fits your privacy and retention practices.
The p=none setting requests reporting without asking receivers to quarantine or reject failing messages. Once legitimate senders pass and alignment is understood, consider a staged move to p=quarantine, followed by p=reject when your organization is prepared. The correct timing depends on the completeness of your sender inventory and the quality of your report review.
Cadence and checkpoints
Use a repeatable schedule rather than waiting for a delivery complaint. A practical operating rhythm is:
- During initial setup: verify each SPF, DKIM, and DMARC record in authoritative DNS, then test mail from every known sending system.
- Weekly during rollout: review available aggregate reports, investigate new source IPs or services, and record decisions in the sender inventory.
- Monthly: compare active senders with contracts, application lists, and DNS records. Remove retired includes and unused DKIM selectors only after confirming they are no longer needed.
- Before launching a new tool: request its authentication requirements, add the records in a planned change, and send controlled tests before enabling a large campaign or workflow.
- Quarterly: review DMARC policy, reporting access, subdomain use, forwarding behavior, and ownership of the domain and DNS account.
Keep a change log with the date, DNS record changed, person approving it, provider affected, and test result. If DNS is managed through a third-party DNS hosting provider, include the account and zone where the record was changed. This makes incident response faster when a record is edited in the wrong zone or a domain transfer changes administrative access.
How to interpret changes
A DMARC failure does not automatically mean that a message is malicious. First separate authentication from alignment. SPF or DKIM can pass while DMARC fails if the authenticated domain does not align with the domain shown in the visible From address. A provider may need to configure a custom return path, custom DKIM domain, or both.
Use this troubleshooting sequence:
- Identify the affected From domain, envelope-from domain, DKIM signing domain, selector, source IP, and sending service.
- Check whether SPF passed and whether its authenticated domain aligns with the From domain.
- Check whether DKIM passed, whether the selector exists, and whether the signing domain aligns.
- Confirm that the sender is authorized in your inventory. If it is unknown, investigate before adding it to SPF or changing policy.
- Check forwarding, mailing lists, and message gateways. These can alter message paths or content and may cause SPF or DKIM results to change.
- Retest after the provider or DNS administrator makes a change, allowing for DNS caching and provider processing time.
Common fixes include publishing a missing DKIM key, correcting a selector hostname, consolidating duplicate SPF records, enabling alignment in a third-party sender, or separating marketing and transactional mail onto clearly managed subdomains. Avoid adding broad authorization for an unknown service simply to make a failure disappear.
For related DNS administration, review Managed DNS Compared. If a domain move is involved, use the domain transfer and renewal checklist and verify email records before changing nameservers.
When to revisit
Revisit this configuration monthly while DMARC is in monitoring mode, then at least quarterly after the sender estate is stable. Return immediately after a new mailbox provider, marketing platform, support desk, website form, acquisition, domain transfer, or DNS migration is introduced. Also review it after unexplained delivery changes, a rise in authentication failures, or the retirement of an application.
At each review, answer five practical questions:
- Does every active sender have a named owner?
- Are there any report sources or sending domains that the inventory does not explain?
- Are SPF records consolidated and limited to current services?
- Do DKIM keys and DMARC alignment pass for the important sending flows?
- Is the current DMARC policy appropriate for the organization’s confidence level?
Record the answers and schedule the next checkpoint. If results remain clean across several review cycles, tighten DMARC in a controlled stage rather than making a sudden change. If failures appear, pause policy changes, identify the affected sender, and restore a known-good configuration before investigating further. This recurring process keeps domain email authentication aligned with the way the business actually sends mail.