Back to docs

Product Basics

How Cold Mail Server Works

Understand the system model behind Cold Mail Server: what each resource represents, how requests are authenticated, and where to look when delivery quality changes.

Audience: All customersTime: 12-18 minutes

Updated April 2026

Before you begin

  • Account access

Table of contents

  1. 1) What Cold Mail Server actually does
  2. 2) The resource model: what each object is for
  3. 3) How an outbound message is processed
  4. 4) Why this structure improves reliability

1) What Cold Mail Server actually does

Cold Mail Server is a sending infrastructure layer between your campaign tool and recipient inbox providers.

  • Your campaign tool submits SMTP traffic through Cold Mail Server instead of sending directly from your app infrastructure.
  • Cold Mail Server enforces domain policy, mailbox identity, and authentication before accepting sends.
  • Accepted messages are relayed using your configured sending identity and tracked for delivery outcomes.
  • The platform is designed to make sender operations observable: setup state, auth state, and message-level results are all visible in one workspace.

2) The resource model: what each object is for

The platform uses a layered model so each object has a clear responsibility.

  • Domain: trust boundary and policy container. DNS authentication and send limits live here.
  • Mailbox: sender identity under a domain, usually aligned to a persona, team, or campaign lane.
  • Auth Key: credential bound to one mailbox. It authorizes tools without exposing broader account access.
  • Email Log Entry: immutable event record used for troubleshooting and trend analysis.
  • Workspace: collaboration boundary for teammates, visibility, and ownership.

3) How an outbound message is processed

Understanding the message path helps you debug faster and make safer changes.

  • Step 1: your sending tool authenticates with mailbox credentials (username + mailbox auth key).
  • Step 2: Cold Mail Server validates mailbox state, domain verification, and policy controls.
  • Step 3: if accepted, the message is relayed and logged with status events for follow-up.
  • Step 4: you use Email Logs and domain/mailbox signals to decide whether to scale, pause, or adjust.

4) Why this structure improves reliability

The model separates concerns so issues are easier to isolate.

  • Credential issues stay mailbox-scoped instead of affecting an entire workspace.
  • Policy changes (like limits) can be managed at the domain level without rotating every credential.
  • Operational visibility is centralized, so teams can diagnose problems from logs instead of guesswork.
  • A clear boundary between identity, auth, and policy lowers blast radius during incidents.

Related guides