Cold Email Disaster Recovery Plan for Agencies with Active Client Campaigns

Something has gone catastrophically wrong. A domain is burned, inboxes are sending to spam, a client's campaign is producing zero results, and you need to fix it now. You do not have a plan. You are making it up as you go. Here is what that plan should look like — written before the next crisis hits.

Phase 1: Detection (0–2 hours)

You discover the problem. Maybe a client reports no replies. Maybe your weekly placement test shows a crash. Maybe you get a bounce alert. Immediate actions:

• Stop all cold outreach from affected inboxes. Do not send "just one more campaign" to test. Stop.
• Run placement tests from all client inboxes using the placement test to determine the scope. Is it one inbox, one domain, or all domains?
• Check Postmaster Tools, SNDS, and the blacklist checker for red flags.
• Check authentication headers on a test email using the DKIM checker, SPF checker, and DMARC lookup.

Phase 2: Triage (2–6 hours)

Categorize the damage:

• Contained: One inbox or one domain affected. Other infrastructure is healthy.
• Moderate: Multiple inboxes or domains affected. Some healthy infrastructure remains.
• Total: All or nearly all sending infrastructure compromised.

Identify the likely cause: authentication failure (DNS change, key rotation issue), reputation damage (bad campaign, complaint spike), blocklist listing, provider infrastructure change, or volume spike. Check the burn score calculator on affected domains for a comprehensive health snapshot.

Phase 3: Stabilize (6–24 hours)

For contained damage

Remove affected inboxes from campaigns. Redistribute volume to healthy inboxes. Begin warmup-only on affected inboxes.

For moderate damage

Remove all affected inboxes from campaigns. Deploy prewarmed backup inboxes from reserves. Redistribute campaigns to healthy inboxes plus backups.

For total damage

Deploy all available backup inboxes. If backup reserves are insufficient, source additional prewarmed inboxes from WarmInboxes immediately — they can deploy same-day. Reduce campaign volume across the board until new infrastructure is tested and stable.

Phase 4: Communicate (within 24 hours)

Inform affected clients. Be honest about the situation but lead with the solution. Template:

"We identified a deliverability issue affecting your campaign. We've already deployed backup infrastructure and campaigns will resume at full volume within [timeframe]. I'll send a technical summary today. No action needed from you."

Clients respect transparency and competence. What they do not respect is finding out weeks later that their campaign was sending to spam.

Phase 5: Recover (1–6 weeks)

Put all damaged infrastructure on recovery protocol: warmup only, no cold outreach. Monitor Postmaster Tools and SNDS weekly. Run placement tests with the placement test every 2 weeks. Set clear recovery benchmarks — 80%+ placement for 3 consecutive tests before resuming.

Phase 6: Post-mortem (within 1 week of stabilization)

Document what happened, why, and what was done. Identify what could have prevented the disaster. Update the plan based on lessons learned. Replenish backup reserves immediately. Most agencies experience the same failure modes multiple times because they don't run post-mortems.

What to include in your written plan

How to make this plan actually work

The biggest gap in most disaster recovery plans is backup infrastructure. Agencies plan for detection and communication but don't have replacement inboxes ready to deploy. With prewarmed inboxes from WarmInboxes available on demand, the stabilization phase shrinks from weeks (warming new inboxes) to hours (deploying prewarmed ones). The time to set this up is before a crisis — not during one.