Pull burning accounts off the line.
Eight codified rules with severity tiers from warn to pause to replace. When bounce velocity climbs or a DNSBL hits, Mailnurse stops the sends before the campaign goes wider.
A bad domain keeps sending until someone notices.
When a sending account starts misfiring — bounce rate climbs, a DNSBL listing lands, complaint rates spike — most agencies discover it after the campaign has already run for hours. By then, the new spam signal has been broadcast wider than necessary, and the recovery curve is longer.
-
Triage rules live in Notion docs, Slack threads, and the founder's memory — not in the system actually monitoring accounts.
-
Two operators on the same fleet apply "when do we pause?" differently. The fleet drifts in ways the dashboard cannot explain.
-
By the time the morning standup catches a burning account, it has been sending for twelve hours.
Rules engine
8 active · Last evaluation 09:24- pauseBounce velocity7-day climb vs baseline
- pausePlacement drop< 85% inbox placement
- warnBlacklist hits2+ DNSBL listings
- pauseAuthentication regressionSPF/DKIM/DMARC fail rate > 5%
- warnWarmup driftengagement slope < expected
- replaceEngagement collapseopen rate -25% over 7d
- warnAge mismatchsend volume > age-banded threshold
- warnFleet anomalycross-account signal
Eight rules. Three severity tiers. One audit log.
- 01
Codify.
Mailnurse ships with eight default rules covering the most common deliverability failure modes. Each has a tunable threshold per workspace.
- 02
Evaluate.
Rules run nightly against every account in your fleet — and continuously for fast-moving signals like bounce velocity. Each evaluation produces an automation-log entry.
- 03
Act.
Rules emit one of three severity actions: warn (notification only), pause (sending halted, account preserved), or replace (rotation triggered). The action is logged with the rule that fired.
The minimum playbook to run cold email without daily intervention.
Eight default rules.
Bounce velocity, placement drop, blacklist hits, authentication regression, warmup drift, engagement collapse, age mismatch, and fleet anomaly. The minimum playbook to run cold email without a human in the loop every day.
Three severity tiers.
Warn, pause, replace. Each rule resolves to one action; you decide which threshold maps to which tier per workspace.
Automation log.
Every rule evaluation is recorded — when it fired, against which account, with what observation, leading to what action. Audit-trail-grade history; queryable, exportable.
Per-workspace tuning.
Rule thresholds adjust per workspace. The aggressive thresholds appropriate to a fast-burn agency client are not the thresholds for a slow-burn enterprise client. Same rules, different dials.
What the rules do, and what you can tune.
What are the eight default rules?
Can I add my own rules?
What does replace actually do?
How are rule evaluations logged?
Two disciplines that compound this one.
Health monitoring
Continuous placement testing, 11-DNSBL probes, warmup vitals every four hours. A composite risk score per account so decay is visible before deliverability collapses.
Read more 02Auto-rotate domains
When a domain burns, replace it end-to-end: Spaceship → Cloudflare → Workspace → Instantly → warmup. The ten-hour playbook, run while you sleep.
Read moreCare, expressed as precision.
Cold-email infrastructure that watches itself — so you can focus on the campaign, not the chassis.
14-day free trial · No credit card · Instant setup