In this lesson
Overview
In this video, Wesley Hoang walks through the foundational setup for building a robust cold email infrastructure. He covers the technical requirements to prevent spam, the process of warming up domains, and how to leverage modern AI agents and MCP connections to automate outbound management.
Foundations of Infrastructure & Deliverability (1:05 - 4:55)
Infrastructure First: Before running campaigns, you must ensure your email infrastructure is secure. This includes properly configuring DNS records (SPF, DKIM, and DMARC) either via automated services or manual import (1:05 - 2:44).
Warm-up & Reputation: To build sender trust, dedicate 2–3 weeks to a warm-up process. Gradually increase volume, starting at 5 emails per inbox daily, ramping up to 10 to mimic natural human behavior (2:44 - 4:10).
Smart Delivery Checks: Run pre-send deliverability tests to verify that emails land in the primary inbox rather than spam or promotions folders before launching live campaigns (4:10 - 4:55).
Campaign Building & Management (5:25 - 13:14)
Sequence Strategy: Use 'spin tags' to create variations in email copy so that every recipient receives a slightly unique message. Avoid tracking open or click rates, as these can negatively impact deliverability (5:25 - 7:28).
Strategic Settings: To bypass spam filters, avoid round-number intervals. Use off-hour sending windows (e.g., 9:33 AM) and randomize sending gaps (e.g., 18-19 minutes) to mimic human activity (7:28 - 10:08).
Delivery Optimization: Always use plain text mode and maintain a healthy ratio of net new leads to follow-ups (recommended 70/30 split) to ensure high engagement without overwhelming your sending reputation (10:08 - 12:10).
Scaling with AI Agents & MCP (14:41 - 19:15)
Smart Agents: Deploy autonomous workflows that trigger based on campaign activity, such as sending instant Slack notifications for positive replies or syncing lead status to CRMs like HubSpot or Airtable (14:41 - 16:35).
Claude MCP Integration: Connect Smartlead to Claude via the Model Context Protocol (MCP) to turn your AI into a command center. This allows you to perform complex analytics—such as identifying top-performing campaigns or lead segments—through simple natural language queries (16:35 - 19:15).
