My First Real n8n Workflow: Why It Took 12+ Hours (and What I Learned)

Context

Today’s blog is about my beginner journey with n8n.

I jumped into n8n even though I already have a background in full-stack development, machine learning, deep learning, and data science. The reason was simple: during an AI internship assessment, I was introduced to low-code tools, and n8n was one of them.

At first glance, n8n looks easy. But building my first end-to-end workflow took me 12+ hours. If it looks simple from the outside, it definitely wasn’t simple the first time.

This post is not theory-heavy. I wasn’t just reading docs — I was building, breaking things, and figuring out how automation actually works.

What I Needed to Understand First

Before touching the platform, I realized there are some basic terms you must be comfortable with:

• Node – a single step in the workflow
• Trigger – what starts the workflow
• API keys – required for most integrations
• LLMs & AI applications – if you’re building AI-related flows
• Basic prompt engineering – especially when working with AI nodes

The most important one is the trigger.

A trigger is basically the button of your workflow. When it fires, the whole workflow starts. That trigger could be:

• A schedule (Cron)
• A Telegram bot message
• WhatsApp
• Or anything else supported

Local n8n vs Cloud (A Mistake I Almost Made)

Before using the platform, I tried thinking about running n8n locally with Docker.

You can do that, but I don’t recommend it for beginners.

Here’s why:

• Local n8n runs on HTTP
• Services like Telegram and WhatsApp require HTTPS
• That instantly blocks many real-world integrations

Because of this, I used:

• n8n’s hosted platform (free tier)
• Or third-party hosted providers

That decision saved me a lot of unnecessary debugging time.

The Actual Workflow I Built

Enough theory. I built a very simple but complete workflow:

"Every day at 9 AM, send a motivational quote to my Gmail."

High-Level Flow

Cron → Get Quote → Format Message → Send Gmail

Step 1: Create a New Workflow

• Open n8n
• Click New Workflow
• Name it: Daily Motivation Email

Step 2: Cron Node (Trigger)

This is what starts everything.

Node: Cron
Settings:

• Mode: Every Day
• Hour: 9
• Minute: 0
• Timezone: Your local timezone

This guarantees the workflow runs daily at exactly 9:00 AM.

Step 3: HTTP Request Node (Get Quote)

Node: HTTP Request
Method: GET
URL: https://zenquotes.io/api/random
Response Format: JSON

The API returns something like:

[
  {
    "q": "The best way to get started is to quit talking and begin doing.",
    "a": "Walt Disney"
  }
]

Step 4: Format the Email (Code Node)

This part took some trial and error.

• Add a Code Node
• Language: JavaScript
• Name it: Format Email
• Connect it after the HTTP Request node

Code used:

const quote = $json[0].q;
const author = $json[0].a;

return [
  {
    subject: "🌅 Your 9 AM Motivation",
    body: `Good morning ☀️

“${quote}”
— ${author}

Have a great day 💪`
  }
];

This outputs exactly what the Gmail node needs:

• subject
• body

Step 5: Activate the Workflow

• Click Save
• Toggle Active

🎉 Done. Every day at 9 AM, a motivational email lands in my inbox.

What Actually Frustrated Me

There is a lot of text-and-try error in n8n.

From an engineering perspective:

• I constantly wanted to add more logic
• I wanted more control and flexibility
• Many times, I felt limited by the platform

Most of my time wasn’t spent building the workflow — it was spent fighting constraints.

If you’re used to building rigid and robust systems from a backend or API point of view, n8n can feel restrictive.

What I Want Future-Me to Remember

• n8n is great for non-tech users and quick automation
• It’s fine for simple AI or integration workflows
• Trial and error is the real way to learn it
• If you already think like a systems engineer, you will feel its limits fast

It impressed me in how much it hides complexity — but it didn’t impress me from a deep engineering standpoint.

Still, it helped me finish an assessment and understand automation at a higher level.

We’ll go deeper later into agentic AI, framework-based approaches, and more advanced systems.

For now — this was my honest beginner experience.

Happy learning, and Namaste 🙏
— Omkar Chebale