High-Volume Sending: Deliverability at 1M+ Emails per Month

Why High Volume Is Different

At low volume, deliverability problems are often invisible. A 0.3% complaint rate on 10K sends is 30 complaints — manageable. At 1M sends, that same rate is 3,000 complaints, and ISPs notice fast.

High-volume sending amplifies every problem: list hygiene issues, authentication misconfigurations, engagement decay. It also demands infrastructure that low-volume senders never need to think about.

IP Infrastructure

Dedicated IPs Are Mandatory

At 1M+ monthly volume, shared IPs are unacceptable. You need full control over your IP reputation. Most ESPs offer dedicated IPs starting at $20-50/month per IP.

IP Pool Sizing

Monthly Volume	Recommended IPs (Marketing)	Recommended IPs (Transactional)
1M-3M	3-5	1-2
3M-10M	5-10	2-3
10M-50M	10-20	3-5
50M+	20+ with rotation	5-10

These assume healthy engagement. Poor engagement means you need more IPs to distribute the reputation damage, which is a band-aid, not a solution.

IP Pool Segmentation

Separate your IP pools by mail type:

• Transactional pool: Receipts, password resets, alerts
• Marketing pool: Campaigns, newsletters
• Re-engagement pool: Win-back campaigns (higher risk)
• Cold outreach pool: If applicable (highest risk)

Never mix transactional and marketing on the same IPs.

Practitioner note: I managed infrastructure for a SaaS company sending 8M emails/month. They started on 3 shared IPs. After moving to 12 dedicated IPs segmented by stream, their Gmail inbox rate went from 72% to 94% in 6 weeks. The IP segmentation alone — without any other changes — made the difference.

Sending Architecture

Provider-Specific Throttling

Each mailbox provider has different rate limits. Configure your MTA or ESP to throttle per-provider:

• Gmail: Start at 500/hour per IP, scale to 5K-10K/hour per IP with good reputation
• Outlook: More conservative — 200/hour per IP initially
• Yahoo: Similar to Gmail but slightly more lenient

If you're using PowerMTA or a custom MTA, configure domain-level throttling rules. If you're on an ESP, check whether they handle this automatically.

Time Distribution

Don't send 1M emails in a 2-hour window. Spread sends across 8-12 hours minimum. This matches natural sending patterns and keeps you under per-hour rate limits.

For marketing campaigns, stagger sends: most engaged recipients first (they generate positive signals that improve delivery for subsequent batches), then less engaged segments later.

Engagement-Based Segmentation

At high volume, engagement-based sending isn't optional — it's survival. Segment recipients by engagement recency:

Segment	Definition	Send Priority
Active	Clicked in last 30 days	First — generates positive signals
Engaged	Opened in last 60 days	Second
Passive	Opened in last 90 days	Third — reduced frequency
Inactive	No engagement 90+ days	Re-engage or sunset

List Hygiene at Scale

Real-Time Bounce Processing

Process bounces in real-time, not batch. At 1M+ volume, a batch process that runs daily could send thousands of additional emails to addresses that bounced yesterday. Hard bounces should be suppressed within minutes.

Continuous Validation

Run your list through a validation service quarterly at minimum. At high volume, addresses go stale faster — people change jobs, abandon mailboxes, providers recycle addresses into spam traps.

Complaint Processing

Register for all available feedback loops:

• Gmail: Feedback Loop (via Postmaster Tools)
• Outlook: JMRP
• Yahoo: Complaint Feedback Loop

Process complaint suppressions within hours, not days.

Practitioner note: At high volume, I always set up automated alerting on complaint rates. If Gmail complaints exceed 0.15% on any campaign, the system should pause sending and alert the team. A 0.1% to 0.3% jump at 1M volume means 1,000-3,000 additional complaints hitting your reputation within hours.

Monitoring Infrastructure

Daily Checks

• Google Postmaster Tools: Domain and IP reputation
• Microsoft SNDS: IP status
• Blacklist monitoring: Automated checks every 4-6 hours
• Bounce rates: Per-campaign and per-provider
• Complaint rates: Per-campaign and per-provider

Automated Alerting

Set threshold alerts for:

• Bounce rate exceeding 2% on any campaign
• Complaint rate exceeding 0.15% on any provider
• IP appearing on any major blacklist
• Postmaster Tools reputation dropping below "High"
• SNDS status changing to yellow or red

Inbox Placement Testing

Run seed tests before every major campaign and on a weekly cadence for automated sends. At high volume, a 5% inbox placement drop means 50K+ emails going to spam.

Practitioner note: The companies doing high-volume email well all have one thing in common: a dedicated email operations person or team. At 1M+/month, email infrastructure requires ongoing attention — it's not a "set and forget" system. If you don't have that resource internally, outsource it.

Scaling Up Safely

When increasing volume (new product launch, acquisition, seasonal spike):

1. Increase no more than 25-50% per week
2. Warm additional IPs before you need them
3. Monitor reputation daily during the ramp
4. Have rollback plans if metrics degrade
5. Send to engaged segments first at new volume levels

If you're scaling to high-volume sending or experiencing deliverability challenges at scale, a deliverability audit can design the right infrastructure and monitoring stack for your volume.

Sources

• Google: Email Sender Guidelines
• Google: Postmaster Tools
• Microsoft: SNDS
• M3AAWG: Best Practices for Large-Scale Senders
• Port25: PowerMTA Documentation

---

v1.0 · March 2026