Make.com to n8n Migration Guide

The Problem

Many advanced Make.com users face difficulty when switching to n8n due to differences in how looping, integrations, and workflow structure work. Common issues include crashing Google Sheets with large loops, not knowing how to implement conditional flows or error handling, and confusion around integrating AI agents and third-party APIs like Google or Perplexity. This video highlights such pain points and the real challenges developers face in this transition.

The Solution

We’ve crafted a complete solution using n8n to replicate and improve the logic you’re familiar with in Make.com. Here’s how I tackled each of the core issues step-by-step.

1. Controlled Looping with Split Batches

Instead of pushing all data at once (which leads to Google Sheets crashes), I implemented Split In Batches to handle data in smaller chunks—2 rows per cycle. This prevents overload and ensures high reliability even with 100+ records. Each batch completes and writes to the sheet before the next one begins.

2. Smart AI Agent Integration

I configured an AI Agent using OpenAI’s GPT-4 with a clear system message and input prompt structure. When more context is needed, you can add memory or attach tools like HTTP Request nodes to extend capabilities—like fetching live data from APIs or scraping insights dynamically. This enables the agent to work with real-time data and return contextual responses.

3. Reliable Google API Integration

Using OAuth2 credentials set once inside n8n, I connected Google Sheets for both read and write operations. This avoids re-authentication issues and supports continuous workflows. The workflow fetches emails, processes them, and appends AI-generated responses to a second sheet.

4. Streamlined Conditional Routing & Automation

Using IF nodes, I set up conditional logic to check whether data needs to go to AI, be skipped, or passed through other services. Combined with Set and Merge nodes, we built full dynamic flows similar to Make’s routers—only more flexible and easier to debug.

Workflow Highlights

• Manual trigger for testing and control
• Google Sheets → Email data source
• AI generates human-like replies
• Split in batches keeps Google Sheets stable
• Smart logic routes each dataset based on status

Bonus Features

I’ve added optional error handling using Error Trigger nodes to notify on failure. Tools like HTTP Request, external APIs, and dynamic values are plug-and-play. This means workflows can handle AI, data transformation, and posting to platforms like WordPress—completely autonomously.

Want the full JSON file? Or need a live walkthrough to customize your use case? I’d love to help you transition your Make.com flows into clean, powerful n8n systems.