Skip to content
EncryptTechBD
Home Industries Our Work Blog Pricing About Contact Discuss Your Project
বাংলা (BN)
Website & Infrastructure 6 min read

SPF, DKIM and DMARC: a change-control checklist

A preventive checklist for adding senders, changing DNS and rolling out DMARC without breaking legitimate email.

E
EncryptTechBD · 0 comments
Network cables with RJ45 connectors on a white desk
Photo: Piotr Lohunko / StockSnap

Email authentication breaks in the gaps between teams. A marketing manager adds a new sending tool, a developer edits a DNS record during a website migration, a finance manager forwards invoices through a personal rule, and nobody records the change. Weeks later, quotations land in spam and the argument starts about whether SPF, DKIM or DMARC is “broken”.

This checklist treats authentication as change control. It is written for the person who owns the domain, not for a mail-server specialist. Use it when you add a sender, change a DNS provider, migrate a website or tighten a DMARC policy.

1. Build the domain and sender register

You cannot protect a domain you have not listed. The register is the most useful artefact in the programme.

  • Domain inventory. Record the root domain, every subdomain that sends mail, the registrar, the DNS host and the renewal date. Owner: IT lead. Evidence: register with dates and account references. Frequency: reviewed quarterly. Sign-off: operations lead.
  • Sender inventory. For each service that sends as the domain, record the vendor, purpose, envelope domain, DKIM selector and the person who can change it. Include staff mail, invoices, support tickets, website transactions, payroll, marketing and any application that sends notifications. Owner: IT lead. Evidence: one row per sender. Frequency: updated before go-live. Sign-off: operations lead.
  • Ownership record. Name who can edit DNS, who approves a policy change and who receives DMARC reports. Shared logins are not ownership. Owner: operations lead. Evidence: access list and approval note. Frequency: every staff change. Sign-off: business owner.

2. Use a sender intake form

Every new tool is a potential authentication failure. Do not let a trial account send mail before the form is complete.

  • Purpose and volume. What is sent, to whom and how often. A password reset and a monthly newsletter have different risk profiles.
  • Envelope domain and From domain. Ask the vendor which domain appears in each. If they differ, alignment must be checked.
  • DKIM details. Selector, signing domain and whether a custom domain is supported. Shared signing domains may not satisfy DMARC alignment.
  • SPF impact. The exact include or IP range, and the resulting DNS lookup count. Record it before the change.
  • Test evidence. A real message to a Gmail and an Outlook mailbox, with the authentication headers saved.
  • Exit plan. What happens to the DNS record and the data when the trial ends or the contract is cancelled?

Give the form an owner and a sign-off; the business owns the domain and the decision.

3. Control DNS changes like production changes

Most authentication incidents follow a DNS edit. Treat the zone as production configuration.

  • Snapshot first. Export or photograph the current TXT records, DMARC record, DKIM selectors and TTLs. Owner: IT lead. Evidence: dated snapshot in the change ticket. Frequency: before every change. Sign-off: change approver.
  • One change at a time. Do not combine a registrar move, a nameserver change and an SPF edit in one window. Owner: IT lead. Evidence: change ticket with a rollback step. Frequency: every DNS change. Sign-off: operations lead.
  • Check from two resolvers. A single cached resolver can show an old record. Verify the published value from at least two public resolvers after the TTL has passed. Owner: IT lead. Evidence: query output with timestamps. Frequency: after every change. Sign-off: change approver.
  • Send a live test. Authentication headers matter more than a green tick in a checker. Send from each affected flow and save the Authentication-Results header. Owner: IT lead. Evidence: test messages and headers. Frequency: after every change. Sign-off: operations lead.

4. Roll out DMARC in gates, not leaps

A DMARC policy is a decision about what receivers should do with mail that fails authentication. Moving to rejection before every legitimate sender is known will block real business email.

  • Gate 1: visibility. Publish a monitoring policy with aggregate reporting and collect data long enough to see every sender. Owner: IT lead. Evidence: report inventory by sending IP and domain. Frequency: until the sender list is stable. Sign-off: operations lead.
  • Gate 2: alignment fixes. Correct SPF, DKIM and alignment for each legitimate sender. Do not weaken the policy to accommodate an unmanaged tool. Owner: IT lead. Evidence: before-and-after authentication results. Frequency: per sender. Sign-off: operations lead.
  • Gate 3: quarantine. Apply a partial or full quarantine policy and watch reports and user complaints. Keep a rollback path. Owner: IT lead. Evidence: policy value, report review and complaint log. Frequency: weekly during rollout. Sign-off: business owner.
  • Gate 4: reject. Move to rejection only when the reports show no unexplained legitimate failures. Owner: IT lead. Evidence: signed rollout note and final policy. Frequency: once, then reviewed quarterly. Sign-off: business owner.

5. Cover the human changes too

Staff moves create mail flows that no vendor form captures. A leaver’s mailbox may forward to a personal address; a new assistant may add a rule that redirects invoices; a shared inbox may send under the domain without DKIM.

  • Joiner and leaver checklist. Confirm mailbox creation, forwarding rules, shared mailbox membership and signature tools. Owner: HR and IT. Evidence: completed checklist per person. Frequency: every joiner and leaver. Sign-off: line manager.
  • Forwarding review. List every automatic forward and external recipient. Remove anything without a business reason. Owner: IT lead. Evidence: forwarding report. Frequency: quarterly. Sign-off: operations lead.

6. Keep monthly proof

Authentication is not a one-time project. A short monthly check catches drift before customers do.

  • Export the current SPF, DKIM and DMARC records and compare them with the approved baseline.
  • Review DMARC aggregate reports for new senders, failing sources and unexpected volumes.
  • Test one message from each critical flow: invoices, password resets, support replies and marketing.
  • Check the DMARC policy, reporting addresses and DNS TTLs.
  • Record the reviewer, the date and any action taken.

7. Red flags in a handover

  • The previous agency owns the DNS account or will not transfer it.
  • Two SPF records exist and nobody knows which is current.
  • DKIM selectors are undocumented or point to a provider you no longer use.
  • DMARC reports go to an address nobody reads.
  • Marketing sends from the root domain with no separate subdomain.

First week actions

  1. Create the domain and sender register.
  2. Snapshot every authentication record and store it with the change ticket.
  3. Name the person who can edit DNS and the person who approves policy.
  4. Send a test through each critical flow and save the headers.
  5. Put the monthly check on the calendar with a sign-off.

If you need help inventorying senders and running a controlled rollout, start with Managed Technology & Cybersecurity. The pricing page explains how the work is scoped. When mail is already failing, use the diagnostic sequence in SPF, DKIM and DMARC problems and fixes.

Sources

Services that deliver this

Want this implemented properly?

Pricing →

Every engagement starts with a written scope. If a requirement is outside it, we quote it separately before starting.

Responses (0)

No comments yet. Add the first one below.

Leave a response

Your email address will not be published. Required fields are marked *.

Keep reading

More field notes