The three numbers that matter#
Sender reputation is the score receiving servers give you. It's not published anywhere — you reverse-engineer it from three measurable signals:
| Signal |
Healthy |
Warning |
Critical |
| Bounce rate |
<2% |
2-5% |
>5% (Gmail throttles) |
| Complaint rate |
<0.1% |
0.1-0.3% |
>0.3% (most ISPs blocklist) |
| Inbox placement |
>85% |
70-85% |
<70% (re-engagement campaign overdue) |
AcelleMail surfaces all three in the campaign report. The fourth signal — engagement rate (open + click) — is the leading indicator: a 30% open campaign predicts good reputation next month; a 5% open campaign predicts blocks next month.
Reading the campaign report#
Where the bounce signals live#
Open the campaign → Bounce log tab. Every failed delivery is row-listed with the recipient, bounce type (Hard/Soft chip), and the raw reason from the receiving server:
For machine-codes (5.x.x / 4.x.x DSN classifications) the per-row reason text is the receiving server's verbatim response. See Decoding bounce messages for the full code reference.
The complaint signal — feedback log#
Same campaign → Feedback log tab. Every "Mark as Spam" click recipients made gets recorded here:
Complaints are the most expensive signal. A 0.3% complaint rate gets you blocklisted at Gmail / Outlook within 24 hours. Read the log row-by-row when complaints spike — patterns reveal:
- All from one list segment? Bad source — pause sending to that segment, audit signup flow
- All on first send to a freshly imported list? Permission unclear at signup — full audit needed
- Spread evenly across the list? Content issue (subject line over-promised, body misleading)
Sending-server health signal#
Open the sending server detail and check: SPF/DKIM/DMARC chips green, recent throughput, error-rate trend:
If any chip flips red mid-campaign, that's your authentication setup breaking — receiving servers will start rejecting + downgrading reputation immediately.
The diagnostic test for reputation triage#
When you suspect a problem, fire a one-off test to your own personal Gmail / Outlook / Yahoo accounts:
Open each on its receiving side. Three outcomes:
- All inbox → reputation OK; the original issue is content or list-quality, not transmission
- Gmail spam / Outlook inbox → Gmail-specific reputation hit; check Google Postmaster Tools
- All spam → systemic auth break; check DKIM/SPF/DMARC alignment
Acting on each signal#
| If you see... |
First action |
| Bounce rate jumps from 1% → 6% on one campaign |
List hygiene — bad import; run email verification, prune syntactic + role-account addresses |
| Complaint rate > 0.3% |
PAUSE the campaign. Audit consent. Send re-engagement BEFORE next send to this list. |
| Engagement drops 30% → 12% over 3 sends |
Send frequency too high. Cut cadence in half for 2 weeks; engagement recovers, reputation follows. |
| Inbox placement on test send: Gmail spam, Outlook inbox |
Gmail reputation hit specifically. Check Google Postmaster Tools (separate setup). |
| SPF / DKIM chip red mid-campaign |
DNS change broke alignment. Re-verify domain immediately. |
| Daily-quota exceeded errors |
Vendor cap hit. Either lower send rate or add a second sending server. |
When to add a second sending server#
Once you cross any of these thresholds, run on two servers minimum:
- Sending more than 50k/day
- Mixing marketing + transactional traffic
- Operating in regulated industry (finance, healthcare) where outage costs > $1k/hr
- Single vendor cost > $300/mo (free Postmark tier + paid SES is often cheaper)
See Configuring multiple sending servers for the setup walkthrough.
Advanced: postmaster-tool integrations + per-recipient reputation tracking
The campaign-level signals tell you the aggregate. The next layer is per-recipient-domain reputation, which AcelleMail doesn't surface natively but which postmaster tools expose.
Google Postmaster Tools (gmail.com / googlemail.com):
- Add your authenticated domain at https://postmaster.google.com
- Verify via TXT record (Google walks you through it)
- Within 24 hours you get dashboards for: Spam rate (real Gmail user complaints), IP reputation (High/Medium/Low/Bad), Domain reputation, Authentication results (SPF/DKIM/DMARC pass rates), Encryption (TLS adoption), Delivery errors
Daily check: spam rate. If it crosses 0.3% Google starts throttling silently. Fix immediately by pausing sends + auditing list.
See Gmail postmaster tools walkthrough for full setup.
Microsoft SNDS (outlook.com / hotmail.com / live.com):
- Sign up at https://sendersupport.olc.protection.outlook.com/snds/
- Add your sending IPs (one entry per IP)
- Get per-IP reputation (Red/Yellow/Green) + filter results + complaint rate
See Microsoft SNDS walkthrough for full setup.
Yahoo / AOL — historically had postmaster tools, mostly retired. Best signal is bounce + complaint rates from AcelleMail.
Per-recipient-domain rejection mining — for advanced ops, aggregate the bounce log by recipient domain:
# From your DB host:
mysql -e "
SELECT
SUBSTRING_INDEX(email, '@', -1) AS domain,
COUNT(*) AS bounces,
SUM(CASE WHEN bounce_type = 'hard' THEN 1 ELSE 0 END) AS hard
FROM bounce_logs
WHERE created_at >= NOW() - INTERVAL 30 DAY
GROUP BY domain
ORDER BY bounces DESC
LIMIT 25
" your_database
If one domain (e.g. yahoo.com) accounts for 60% of bounces while contributing only 8% of your list, you have a Yahoo-specific reputation hit. Pause sending to that domain, work the reputation recovery (lower volume, higher-engagement-segment-first, monitor SNDS / Y!Mail postmaster signals).
Programmatic alerting — wire AcelleMail's campaign.bounce_rate_spike webhook (configurable threshold) to Slack:
[POST to your webhook URL]
{
"campaign_uid": "...",
"bounce_rate": 0.062,
"threshold": 0.04,
"exceeded_at": "2026-05-20T14:30:00Z"
}
Your bridge service formats + pings the on-call channel. Catches problems 6-12 hours before customers notice missing emails.
Reputation recovery cycle — when you've taken a hit:
Day 0: Pause all bulk sending. Send ONLY to engaged-last-7d segment (highest open rate).
Day 7: If engagement-segment sends stayed clean, expand to engaged-last-30d.
Day 14: Expand to engaged-last-90d. Monitor SNDS / Postmaster daily.
Day 30: Resume normal segment if reputation back to green/healthy.
The slope of recovery depends on volume + complaint severity. Mild reputation dips recover in 1-2 weeks; severe (>1% complaint) takes 4-6 weeks or never (sometimes you need a fresh IP + domain).
Related articles#
