Domain setup, warmup, monitoring, and scaling -- the full infrastructure stack for cold email that gets delivered.
TL;DR
Email deliverability is not a copywriting problem. You can write the best cold email in the world and it will never get read if it lands in spam. Deliverability is an infrastructure problem -- and it requires the same rigor as any other technical system you build.
This guide covers the full infrastructure stack: domain strategy, DNS setup, mailbox provisioning, warmup, platform selection, and ongoing monitoring. These are the exact systems we use across client campaigns generating hundreds of meetings per month.
Your primary domain (company.com) carries your entire brand reputation. One damaged sending domain can tank your transactional email, your marketing automation, and your team's internal email deliverability. Never risk it.
Instead, purchase alternate domains that are close variations of your brand and use those exclusively for cold outreach.
Good alternate domains look like they belong to your brand. They should be recognizable as yours but not identical to your primary domain.
Standard ratio for cold email infrastructure. More than 3 per domain increases risk; fewer reduces daily capacity.
Three DNS records are non-negotiable before you send a single email. Missing any of them will get you filtered or blocked immediately.
SPF tells receiving mail servers which IP addresses are authorized to send email on behalf of your domain. Without it, your emails look like they could be spoofed.
v=spf1 include:_spf.google.com include:sendgrid.net ~all -- authorizes Google Workspace and SendGrid to send on your behalf. The ~all means unauthorized senders get a soft fail (not hard reject). Use -all only when you are certain all senders are listed.
DKIM adds a cryptographic signature to every email. The receiving server checks the signature against a public key in your DNS. It proves the email was not tampered with in transit.
DMARC tells receiving servers what to do when emails fail SPF or DKIM checks. It also sends you reports on emails sent using your domain.
v=DMARC1; p=none; rua=mailto:dmarc-reports@yourdomain.com -- Start with p=none (monitor only). Do not jump to p=quarantine or p=reject until you have confirmed all legitimate senders are passing DKIM and SPF. Moving too fast on DMARC blocks your own emails.
The provisioning layer is where domains get purchased, DNS gets configured, and mailboxes get created. Doing this manually at scale is operationally impossible. These are the tools we use.
| Tool | Role | Why we use it |
|---|---|---|
| Dynadot | Domain registrar | API-based domain purchase and DNS management. Supports bulk registration and automated provisioning via v3 API. |
| ZapMail | DNS + mailbox setup | Configures SPF, DKIM, DMARC automatically. Creates mailboxes and manages sender profiles. Integrates directly with Google Workspace and Outlook. |
| InboxKit | Inbox testing + monitoring | Tests where your emails actually land across major providers (Gmail, Outlook, Yahoo). Identifies deliverability issues before they affect campaigns. |
| Google Workspace | Mailbox provider | Higher inbox placement rates for B2B recipients. Required for targeting Google-hosted domains. |
| Microsoft 365 | Mailbox provider | Better placement when targeting Outlook/Microsoft-hosted domains. Warmup settings differ from Google. |
From domain provisioning to campaign-ready. Rushing this step is the single most common deliverability mistake. A domain that skips warmup typically burns within 2 weeks of sending.
Warmup is the process of establishing a sending reputation for new mailboxes. You start sending small volumes of email between real mailboxes -- with high engagement signals -- and gradually increase over time.
Mail servers see a new mailbox suddenly sending 100 emails per day as suspicious. The same mailbox that has been sending 2 emails per day for three weeks and receiving replies is trusted.
| Setting | Value | Why |
|---|---|---|
| Initial send volume | 2 emails/day | Conservative start avoids immediate flag |
| Max warmup volume | 15 emails/day per mailbox | Standard cap during warmup phase |
| Auto-reply rate | Enabled | Simulated replies increase engagement score |
| Warmup duration | 2-3 weeks | Score needs to reach 90+ before live sends |
| Score threshold for campaigns | 90+ (100 preferred) | Below 90 risks spam placement |
| Google vs Outlook settings | Different | Outlook warmup requires slower ramp -- adjust settings per provider |
The sending platform is where your campaigns live, your sequences run, and your warmup gets managed. Each platform has different strengths.
| Platform | Best for | Key strength |
|---|---|---|
| EmailBison | High-volume multi-workspace operations | Isolated workspaces per client, aggressive pricing at scale ($599/mo for 500K emails + unlimited workspaces), native warmup, EmailGuard integration for inbox testing |
| Instantly | Ease of use, fast campaign setup | Clean UI, strong warmup system, good deliverability analytics, straightforward account management |
| Lemlist | Multichannel (email + LinkedIn) | Best-in-class LinkedIn sequencing alongside email, strong personalization features, image and video support |
| SmartLead | Analytics-heavy teams | Deep analytics and A/B testing, good for data-driven optimization -- note: infrastructure-heavy setup |
For most campaigns, EmailBison or Instantly are the default choices. Lemlist is the right call when LinkedIn sequencing is part of the motion. SmartLead works for teams that run heavy analytics-driven optimization.
Scaling email infrastructure is not just buying more domains. Each addition to the sending pool needs the same provisioning and warmup process. Shortcuts at scale create compounding problems.
At 15 emails/day capacity per mailbox across 10 domains with 3 mailboxes each. This is the practical throughput of a standard infrastructure build.
Deliverability degrades silently. A domain can move from healthy to blacklisted in 48 hours without any obvious trigger. Daily monitoring catches problems before they become campaigns.
| Metric | Check how | Alert threshold |
|---|---|---|
| Warmup score per mailbox | EmailBison / Instantly dashboard | < 90 -- pull from campaign rotation |
| Bounce rate per domain | Sending platform analytics | > 2% -- pause and investigate |
| Spam complaint rate | Google Postmaster Tools / Microsoft SNDS | > 0.1% -- immediate investigation |
| Blacklist status | MXToolbox or automated check | Any listing -- pull domain immediately |
| Inbox placement | InboxKit or EmailGuard | < 80% inbox -- flag as degraded |
| DMARC pass rate | DMARC aggregate reports | < 95% -- DNS configuration issue |
| Warmup enabled status | Sending platform | Any disabled mailbox -- re-enable and investigate |
Domain degradation is not a failure -- it is expected. Domains used for cold email have a lifecycle. The system you build should handle degradation automatically.
Never assume a warmed domain is landing in the inbox. Test it explicitly with seed-based inbox placement tools before attaching it to any campaign.
Run placement tests weekly during warmup. Daily once domains are in active sending rotation. A placement test that shows less than 80% inbox is a problem to fix before it affects real prospects.
We implement these systems end-to-end. First sends within 14 days.