Email Deliverability Infrastructure: Complete Technical Guide

Email deliverability is infrastructure, not copywriting. Your message reaches the inbox or the spam folder before a single word is evaluated. The sending architecture, authentication configuration, domain reputation, and warm-up protocol determine inbox placement rates independent of message content — and poor infrastructure makes good copy irrelevant.

Why Cold Email Infrastructure Differs from ESP Infrastructure

Consumer email service providers (Mailchimp, Klaviyo) manage deliverability on behalf of their users by maintaining shared IP pools with established sender reputation, managing suppression lists collectively, and enforcing content and engagement standards across all accounts. Cold email platforms (Instantly, Smartlead) give you greater control but place deliverability responsibility on you. The tradeoff is flexibility at the cost of responsibility.

Cold email to purchased or prospected lists is categorically different from email to opted-in subscribers. Opted-in lists have positive engagement signals accumulated over time. Cold lists start with zero signal — inbox providers cannot use historical engagement to infer inbox placement decisions. Authentication, domain reputation, and warm-up signals become the primary inputs to inbox filtering decisions when engagement history is absent.

Domain Architecture: Primary vs Sending Domains

Never send cold email from your primary business domain. This is the foundational rule of cold email infrastructure. If your business operates from company.com, create a dedicated sending domain — company-outreach.com or trycompany.com. The sending domain is expendable in a way your primary domain is not.

For sustained cold email volume, operate three to five sending domains in rotation. Each domain hosts two to four sending mailboxes. Rotating sends across multiple domains means no single domain carries your full sending volume, reducing the risk that one bad day damages your entire outreach capability. Domains should be aged a minimum of 30 days before beginning warm-up; domains aged 60 to 90 days have higher initial trust signals from inbox providers.

SPF, DKIM, and DMARC: Implementation and Verification

SPF (Sender Policy Framework) declares which IP addresses are authorised to send email from your domain. For Google Workspace sending: include:_spf.google.com ~all. For Microsoft 365: include:spf.protection.outlook.com ~all. Verify with mxtoolbox.com/spf.aspx — an invalid SPF record causes mail to be treated as unauthenticated.

DKIM (DomainKeys Identified Mail) adds a cryptographic signature to each outgoing message. Your sending platform generates DKIM keys; you add the public key as a TXT record in DNS. DMARC tells receiving servers what to do when SPF and DKIM checks fail. Start with p=none (reporting only), transition to p=quarantine after 30 days of clean authentication data, and p=reject after 60 days. BIMI requires DMARC at p=quarantine or p=reject and adds your brand logo to inbox displays.

IP Warming: Protocols and Timelines

A new sending IP carries no reputation. Inbox providers see a pattern of zero email followed by sudden volume as a spam indicator. Warming gradually builds sending volume to establish positive reputation before scaling. The warm-up schedule for a new mailbox: Days 1 through 7 — 20 to 30 emails per day via warm-up network. Days 8 through 14 — 40 to 50 emails per day, 20 to 30 real cold emails mixed in. Days 15 through 21 — 60 to 80 emails per day, 40 to 60 real cold emails. Days 22 through 30 — 100 to 150 emails per day. Days 31 through 45 — ramp to target volume of 150 to 300 per mailbox per day.

Most cold email platforms (Instantly, Smartlead) have automated warm-up that manages the warm-up network sends in the background while you begin real campaign sends from day 8. Do not disable warm-up while running campaigns.

Monitoring: The Metrics That Reveal Infrastructure Problems

Six metrics form the deliverability monitoring dashboard. Hard bounce rate: target below 2%; above 3% is a domain reputation warning. Spam complaint rate: target below 0.08%; above 0.1% triggers automated filtering at Gmail and Outlook. Inbox placement rate: measure with GlockApps or MailReach; target 85%+ inbox. Open rate: a proxy for inbox placement and subject line quality. Reply rate: the ultimate signal of campaign health. Unsubscribe rate: above 0.5% indicates relevance problems.

Monitor Google Postmaster Tools and Microsoft Smart Network Data Services (SNDS) for domain and IP reputation signals. Both are free and provide direct feedback from the two dominant inbox providers. A bad or medium reputation rating in Postmaster Tools requires immediate investigation and usually a pause-and-recover protocol.

Platform Selection and Recovery Protocol

Three platform architectures for cold email at different scales. Solo / small team (under 1,000 emails per day): Instantly or Smartlead with Google Workspace or Microsoft 365 mailboxes, 2 to 4 mailboxes per domain, 1 to 2 sending domains. Managed sending team (1,000 to 10,000 emails per day): 8 to 20 mailboxes across 4 to 6 domains, inbox rotation configured, automated warm-up running at all times. Enterprise outbound (above 10,000 emails per day): dedicated IP addresses, custom SMTP infrastructure, advanced monitoring with Postmaster Tools API integration.

Recovery protocol when reputation degrades: pause all sending on the affected domain immediately; continue warm-up network sends at low volume (20 to 30 per day) during the pause; clean the list that generated the bounce or complaint spike; review and revise copy if complaints were related to irrelevance; resume at low volume (50 to 70 per day) and ramp slowly over 3 to 4 weeks. Full recovery takes 6 to 10 weeks. This is why operating multiple sending domains in rotation is a risk management requirement, not a preference.