There are two distinct failure modes that look the same from the outside: transactional email broke (users can't log in, receipts missing), or marketing/outbound email broke (replies dropped, campaigns silent). You need to know which you're dealing with before you can triage.
Good news: the first 15 minutes of diagnostics are identical for both. Start here.
First 15 minutes: did anything change?
Nine times out of ten, a sudden email failure correlates with a specific change. Ask your team:
- Did anyone touch DNS over the weekend?
- Did we migrate the domain, renew the domain, or change domain registrars?
- Did we switch ESP? Upgrade a plan? Add a new sending platform?
- Did we increase send volume recently?
- Did anyone install a new plugin on the website?
- Did we run any automated "domain health" tool that modified DNS records?
- Did the ESP send any recent deliverability or compliance notices?
Write down everything that changed in the last 14 days. Even seemingly unrelated changes (a CDN migration, a new spam filter on the receiving side, a credit card expiration on your ESP account) can cascade into mail failures.
Go to Inbox Check and send your current message (transactional or marketing) through a seed test. In 2–3 minutes you'll know: (a) is mail arriving at all, (b) where is it going, (c) is authentication passing. Free, no signup.
Minutes 15–30: interpret the seed result
Scenario A: Mail is not arriving anywhere
Your domain is probably blacklisted, suspended by your ESP, or has a broken MX/SPF record. Check:
- Log into your ESP. Is there a warning banner? A suspended sender?
- Check the domain's payment status. Expired credit cards disable sending on most platforms.
- Run a blacklist check (several free tools; or see the seed test output).
Scenario B: Mail arrives at some providers, not others
Provider-specific rejection. Most common:
- Gmail only: reputation dropped or triggered a content filter.
- Outlook/Microsoft only: SmartScreen or ATP flagged you; check Microsoft SNDS.
- Enterprise-only: your mail is being quarantined by corporate tenants. Usually due to authentication alignment.
Scenario C: Mail arrives but lands in Spam or Promotions
The mail is technically reaching the provider. Filter decisions are sending it away from Primary. Fix vectors:
- Authentication (if any FAIL): repair SPF/DKIM/DMARC.
- Content (if auth is clean): audit for spam triggers, link density, image-to-text ratio.
- Reputation (if auth and content are clean): reduce volume, warm the domain, wait.
Minutes 30–60: triage by business priority
If it's transactional mail (receipts, password resets, invitations)
This is bleeding revenue and trust. Prioritise in this order:
- Communicate to users. Post in your status page or on the login screen: "We're aware of email delivery delays. Please check your spam folder."
- Enable SMS or in-app fallback for critical flows (password reset, 2FA) if you have that capability.
- Contact your ESP's support. Premium plans often have a priority queue. Free plans rarely.
- Check if you can fail over to a second provider. Teams running critical transactional mail should have a backup (e.g. Postmark + SES).
If it's marketing or outbound
Less urgent than transactional — buyers will find you through other channels — but don't keep sending while broken. More volume into a bad-reputation state digs the hole deeper.
- Pause all outbound. Stop the active sequences. Stop the newsletter.
- Fix the root cause. Authentication first, reputation second.
- Resume with reduced volume. When you re-enable, start at 20–30% of previous volume. Ramp over 7–14 days.
Minutes 60–90: if you can't fix it yourself
If the diagnostic points at authentication and you don't manage your own DNS, you need to involve the person who does. Usually that's IT, an agency, or your domain registrar's support.
The escalation message should include:
- Screenshot of the seed test result showing FAIL on SPF/DKIM/DMARC.
- The specific record that's failing (the tool shows you).
- What you tried and what didn't work.
- The business urgency: "transactional mail is failing, customers are locked out."
Common root causes, ranked by frequency
- DMARC policy change, not propagated to senders. Team enables DMARC quarantine/reject without aligning all sending sources. Roughly 30% of "mail suddenly broken" cases in SMB.
- Volume ramp on a cold domain. Team started outbound on a new domain without warmup. Gmail/Outlook start filtering by week two.
- ESP suspension or rate limit. Expired card, TOS violation, abuse report.
- Blacklist. Usually after a list import that hit spam traps, or after an IP-neighbourhood incident.
- DNS propagation error after registrar change. MX or authentication records got dropped in the migration.
- Content trigger. Rare as a sole cause, but can be last-straw.
What you should do before Monday ever comes
A working Monday morning has two prerequisites:
- Monitoring: a daily or weekly seed test, with alerts on placement drop. Catches 80% of issues before they hurt.
- Redundancy: a backup ESP for transactional mail. Even a cold-standby sender saves you when the primary goes down.
Neither is expensive. Both are hours of setup to save days of fire-fighting.