The Weekly Deliverability Monitoring SOP (Copy This Into Your Agency Playbook)

Most agencies discover deliverability problems the same way: a client asks why replies stopped. By that point the domain has usually been filtering to spam for one to three weeks, the campaign data is polluted, and the recovery clock hasn't even started. A weekly monitoring SOP costs about 20 minutes per client and catches almost everything early.

This is a complete, copy-pasteable routine. Assign it to one person, run it the same day every week, and log results in a shared sheet so trends are visible.

Monday: The 20-minute health sweep

Per sending domain

Run an inbox placement test — log where the email landed (inbox / promotions / spam) per provider
Run a blacklist check on the domain and its sending IP — log any new listings
Verify SPF, DKIM, and DMARC still resolve — DNS records get deleted more often than you'd think
Check domain expiry — an expired domain mid-campaign is an unforced error
Check the tracking domain — confirm CNAME intact, SSL valid, not Cloudflare-proxied

Per client account (from ESP data)

Bounce rate over trailing 7 days — flag anything over 3%
Reply rate trend vs. prior 3 weeks — a 40%+ drop with stable volume is a placement problem until proven otherwise
Open rate trend (directional only — opens are noisy post-privacy changes)
Any inboxes paused, disconnected, or erroring in the sending platform

The traffic-light system

Score every domain green, yellow, or red each week. This keeps judgment consistent across team members.

Green: inbox placement confirmed, no blacklists, auth intact, bounce rate under 2%. No action.
Yellow: promotions-tab placement, one low-severity blacklist listing, bounce rate 2–4%, or a reply-rate dip. Action: reduce volume 30–50%, tighten list quality, re-test in 3 days.
Red: spam placement at Gmail or Outlook, any Spamhaus/Barracuda/SURBL listing, broken DKIM, or bounce rate over 4%. Action: pause the domain today and start the disaster recovery SOP.

The single most expensive mistake in agency cold email is continuing to send on a red domain because a client campaign "can't pause." Sending on a burned domain deepens the damage and extends recovery from weeks to months.

What to do when a domain goes red

Two decisions, in order:

Repair or replace? Use the repair-or-replace calculator. Short version: minor auth breakage → repair; reputation damage with critical blacklist hits → replace.
How does the campaign keep running? If the client can't tolerate a 4–8 week recovery pause, move the campaign to standby infrastructure. Agencies that keep backup inboxes budgeted switch over in a day. Agencies that don't, scramble for weeks. If you need replacement infrastructure fast, WarmInboxes supplies pre-warmed inboxes on fresh domains that can take over sending within 24–48 hours — which is the difference between a hiccup and a churned client.

Monthly additions (first Monday of the month)

Rotate placement tests across every inbox, not just one per domain — individual inboxes on the same domain can diverge
Review DMARC aggregate reports for unknown senders (spoofing attempts show up here)
Re-run the deliverability risk score per client and compare month-over-month
Audit sending volume per inbox against safe limits with the send limits calculator

Tooling and time budget

Everything above runs on free tools on this site plus your ESP dashboard. Time budget: about 20 minutes per client per week once the routine is muscle memory. For a 10-client agency that's half a day per week — dramatically cheaper than one burned-domain fire drill, which typically consumes 15–30 team hours and risks the client relationship.

The agencies with the best deliverability aren't running secret tactics. They're running boring monitoring, every week, without exceptions — and they keep pre-warmed backup capacity from a provider like WarmInboxes so a red domain never becomes a client emergency.