To monitor email deliverability, track bounce rate, spam-complaint rate, and reply trends per campaign, watch your domain's health in Google Postmaster Tools, and run periodic seed-list placement tests to see where mail actually lands. Deliverability degrades quietly — open rates drift down weeks before anything visibly fails — so a lightweight weekly loop beats a quarterly audit.
Monitoring Email Deliverability: The Short Answer
- Deliverability fails silently first. Spam placement doesn't bounce; the only early symptoms are slow drifts in opens and replies.
- Three numbers carry most of the signal: hard-bounce rate (under ~2%), spam-complaint rate (under ~0.3%, ideally 0.1%), and the trend of replies per hundred sends.
- Provider tools show each provider's own view. Postmaster Tools for Gmail domain reputation, SNDS for IP-level Microsoft telemetry — supplemental readings on the scores described in what sender reputation is.
- Placement tests reveal what dashboards can't: whether delivered mail landed in the inbox or the spam folder.
This is the monitoring guide in our deliverability cluster. For the causes behind the numbers — content, infrastructure, and list quality — start with the pillar guide to cold email deliverability.
Common Misconceptions About Deliverability Monitoring
- "Delivered means it reached the inbox." Delivery rate counts mail a server accepted — the spam folder counts as delivered. Genuine placement only shows up in seed tests and engagement trends.
- "Open rate is a reliable metric." Privacy features that prefetch images inflate opens unevenly across providers. Opens are still useful as a trend within one provider, but replies are the honest engagement metric for outbound.
- "No bounces means no problem." Bounces catch dead addresses; they say nothing about placement. A campaign can bounce nothing and land entirely in spam — which is exactly the failure mode monitoring exists to catch. (When bounces do spike, that's a list problem — see how to reduce bounce rates.)
- "Monitoring is a quarterly audit." Reputation damage compounds weekly; by quarter's end, you're doing archaeology. Fifteen minutes a week catches the same problems while they're still cheap.
The Metrics That Matter, and Their Thresholds
| Metric | Healthy | Investigate | Where to read it |
|---|---|---|---|
| Hard-bounce rate | Under ~2% | Above ~2% | Sequencer / sending platform |
| Spam-complaint rate | ~0.1% or less | Approaching 0.3% | Postmaster Tools, platform reports |
| Replies per 100 sends | Stable trend | Multi-week decline | Sequencer, CRM |
| Domain reputation | High / Medium | Low or falling | Google Postmaster Tools |
| Inbox placement | Consistent | Provider-specific drops | Seed-list placement tests |
| Blocklist status | Not listed | Any listing | Blocklist lookup tools |
Thresholds are tripwires, not grades: the trend matters more than any single reading, and a sharp move on one provider while others hold steady is itself diagnostic information.
Deliverability problems don't announce themselves — they compound quietly until a healthy-looking dashboard sits on top of an empty pipeline.
A Weekly Monitoring Loop That Takes Fifteen Minutes
- Pull the three campaign numbers — bounces, complaints, replies per hundred — and compare against the last four weeks, per sending domain and per list source.
- Glance at provider dashboards. Postmaster Tools domain reputation and spam-rate graph; SNDS (IP-level data) if you send meaningful Microsoft volume. Both can be sparse at low volume — you're looking for direction, not absolute values.
- Send one header check. A test message to a mailbox you control, confirming SPF, DKIM, and DMARC still pass with alignment — authentication breaks silently during vendor changes, as covered in the authentication setup guide.
- Run a placement test when anything moved. Seed tests are the confirmation step, not the routine — use them when trends dip, after infrastructure changes, or before scaling volume.
- Log it. A one-line weekly record per domain turns next quarter's "when did this start?" into a lookup instead of a guess.
What to Check Before You React to a Bad Reading
Diagnose before treating: a single metric out of range has several possible causes, and the fixes don't overlap. Bounce spike → read the status codes; address-not-found means the list, policy-rejection means reputation or authentication. Reply-rate slide with steady bounces → placement problem; run a seed test and check whether one provider is filtering while others deliver. Complaint uptick → targeting problem; tighten the segment before touching infrastructure. And after any domain or tool migration, check authentication first — it's the most common silent casualty.
Frequently Asked Questions
What metrics should I track for email deliverability?
Track hard-bounce rate, spam-complaint rate, and replies per hundred sends for every campaign, plus domain reputation in Google Postmaster Tools and periodic inbox-placement tests. Bounces and complaints are the tripwires providers themselves score you on; reply trend is the honest engagement signal; and placement tests reveal whether delivered mail actually reached the inbox rather than the spam folder.
What is a good spam complaint rate for cold email?
Stay at or below roughly 0.1 percent, and treat 0.3 percent as a hard ceiling — that is the spam-rate threshold in Google's bulk-sender guidelines, and sustained rates near it invite filtering regardless of your volume tier. Because a complaint is a recipient actively marking your mail as unwanted, even small upticks are meaningful: they almost always point at targeting that got looser or an audience that never fit.
How do I check if my emails are going to spam?
Combine three views. Run a seed-list placement test to see where mail lands across providers, watch Google Postmaster Tools for your domain reputation and spam-rate trend at Gmail, and monitor reply rates per provider — a sharp decline isolated to one provider strongly suggests filtering there. A test send to your own accounts helps, but a handful of personal mailboxes is a sample, not a measurement.
How often should I run deliverability checks?
Run the lightweight loop weekly: campaign metrics, a glance at provider dashboards, and an authentication header check take about fifteen minutes. Save full placement tests for when something moves — a trend dips, infrastructure changes, or volume is about to scale. Quarterly-only auditing is too slow because reputational damage compounds for weeks before it becomes obvious.
What is a seed list test?
A seed test sends your actual campaign message to a panel of mailboxes across major providers and reports where each copy landed — inbox, spam, or missing. It is the most direct measurement of placement available and the fastest way to confirm whether a suspected filtering problem is real, which provider it affects, and whether a fix actually worked.
Why are my open rates dropping but nothing is bouncing?
That combination is the classic signature of a placement problem: the addresses are real, so nothing bounces, but the mail is landing in spam, so nobody sees it. Confirm with a seed test and provider dashboards, then look for what changed — a volume jump, a new list source, a content shift, or silently broken authentication after a tool migration are the usual culprits.
References
- Google, Email sender guidelines: https://support.google.com/a/answer/81126
- Google, Postmaster Tools: https://support.google.com/a/answer/9981691
- Microsoft, Smart Network Data Services (SNDS): https://sendersupport.olc.protection.outlook.com/snds/
- IETF, RFC 3463 — Enhanced Mail System Status Codes: https://datatracker.ietf.org/doc/html/rfc3463
- Spamhaus, Blocklist removal center: https://check.spamhaus.org/
Next Steps
Monitoring tells you when something upstream broke — and upstream of almost every alert is the data you loaded. Learn what Lead Seeker is and who builds it, and why verified, signal-led data is the quietest deliverability dashboard you can buy.
