How to use n8n to build AI agent workflows that run 24/7. Connect ChatGPT, APIs, and automations without writing code. Complete tutorial with examples.
15 min read · Last updated April 2026
n8n (pronounced "nodemation") is an open-source workflow automation platform that has fundamentally changed how businesses approach automation in 2026. Unlike proprietary tools such as Zapier or Make, n8n gives you complete control over your data and workflows — and it's free when self-hosted.
At its core, n8n is a visual workflow builder that connects AI models, APIs, databases, and over 400 services. You drag and drop nodes, connect them with lines, and build complex automations without writing a single line of code. But what truly sets n8n apart in 2026 is its native AI integration — you can plug in OpenAI, Anthropic Claude, Google Gemini, Ollama, or any local LLM directly into your workflows.
Why does this matter for your business? Because AI-powered automation is no longer a luxury reserved for enterprise companies with deep pockets. With n8n, a small business owner in Raipur can build the same intelligent workflows that Fortune 500 companies pay consultants six figures to create. The platform handles everything from simple email triage to complex multi-step AI agents that research, decide, and act autonomously.
The self-hosted option is particularly significant. When you run n8n on your own server, your business data never leaves your infrastructure. For companies handling sensitive customer information, financial data, or proprietary processes, this level of data sovereignty is invaluable. Compare this to Zapier, where every piece of data flows through their servers and is subject to their privacy policies.
If you're exploring AI tools for your business, check our guide on the best AI tools for small business to see how n8n fits into the broader ecosystem.
Choosing the right automation platform can save your business thousands of dollars annually. Here's how the three major players stack up in 2026, with honest assessments of each platform's strengths and weaknesses.
| Feature | n8n | Zapier | Make |
|---|---|---|---|
| Starting Price | Free (self-host) | $20/mo | $9/mo |
| Free Tier | Unlimited (self-host) | 100 tasks/mo | 1,000 ops/mo |
| AI Integration | Native (all models) | ChatGPT only | Limited |
| Self-host Option | Yes (Docker/npm) | No | No |
| Integrations | 400+ | 6,000+ | 1,800+ |
| Code Nodes | JavaScript/Python | JavaScript only | No |
| Learning Curve | Medium | Easy | Medium |
| Data Privacy | Full control (self-host) | Zapier servers | Make servers |
| Best For | AI workflows, developers | Non-technical users | Visual thinkers |
The verdict: If you're building AI-powered workflows and want to keep costs at zero, n8n is the clear winner. Zapier wins on ease of use and sheer number of integrations, but you'll pay $50-100/month for anything serious. Make offers a middle ground with decent pricing and a visual interface, but its AI capabilities lag behind n8n's native support.
For Indian businesses and startups, the cost difference is significant. Zapier's Professional plan at $49.99/month translates to roughly Rs 4,200/month — that's Rs 50,000/year just for automation. With n8n self-hosted on a Rs 500/month VPS, you get unlimited workflows and executions. Learn more about automating your business with AI in our dedicated guide.
Getting started with n8n takes less than 10 minutes. Here are your three options, from easiest to most powerful.
Step 1: Choose Your Setup
Option A: n8n Cloud (Easiest — 2 minutes)
Sign up at n8n.io with your email or GitHub account. You get 5 active workflows and 5,000 executions per month on the free tier. No credit card required. This is perfect for testing and small projects.
Option B: Self-Host with Docker (Free — 5 minutes)
docker run -it --rm \
--name n8n \
-p 5678:5678 \
-v n8n_data:/home/node/.n8n \
docker.n8n.io/n8nio/n8nOpen localhost:5678 in your browser. Unlimited workflows, unlimited executions, complete data privacy. If you're new to Docker, check our guide to running AI locally for setup tips.
Option C: Install via npm (For developers)
npm install n8n -g
n8n startStep 2: Create Your First Workflow
Once n8n is running, click "New Workflow" in the dashboard. You'll see a blank canvas with a Start node. Drag a trigger node (like Schedule, Webhook, or Email trigger) onto the canvas. Then add your action nodes and connect them by dragging lines between the output and input ports.
Step 3: Add an AI Node
Click the "+" button to add a new node, search for "OpenAI" (or "Anthropic", "Ollama", etc.), and connect your API key. You can now send prompts to AI models and use their responses in subsequent nodes. The AI node supports chat completions, text generation, image analysis, and embeddings.
Step 4: Test and Deploy
Use the "Test Workflow" button to run each node individually and verify the data flow. Once everything works, click "Activate" to make the workflow run automatically. Monitor executions in the Executions tab to catch any errors.
Pro tip: Use the JSON formatter tool to inspect API responses when debugging your workflows. Properly formatted JSON makes it much easier to map data between nodes.
These are production-ready workflows that businesses are using right now. Each one can be built in under 30 minutes with n8n's visual editor.
1. Intelligent Email Automation
Set up a Gmail or Outlook trigger that fires on every new email. Route it through an AI classifier that categorizes emails as "urgent," "needs response," "FYI," or "spam." For emails needing responses, the AI drafts a reply based on your previous email patterns and company knowledge base. You get a Slack notification with the draft — approve with one click, and it sends automatically. This workflow alone saves 2-3 hours daily for busy professionals.
2. Social Media Content Scheduling
Connect an RSS feed or Google Sheet as your content source. The AI reads each article, generates platform-specific posts (Twitter threads, LinkedIn carousels, Instagram captions), adds relevant hashtags, and schedules them through Buffer or directly via platform APIs. Include a password generator step to rotate API keys monthly for security. One n8n workflow can manage content across 5+ platforms simultaneously.
3. Customer Support Routing
Incoming support tickets from Telegram, WhatsApp, or your website form trigger the workflow. AI analyzes the query, checks your FAQ database (you can use Airtable, Google Sheets, or a vector database), and either provides an instant answer or routes to the right human agent with full context. This handles 60-80% of repetitive questions automatically, letting your team focus on complex issues. See how this fits into building AI agents for your business.
4. Lead Scoring and Qualification
When a new lead fills out your contact form, the webhook triggers n8n. AI enriches the lead data (company size, industry, budget signals), scores them 1-100 based on your ideal customer profile, and takes different actions: high-value leads get a personalized email from the CEO and a Slack alert to sales, medium leads enter your nurture sequence, low leads get an automated response. This workflow converts 3x more leads than manual processing.
5. Data Sync Between Tools
Keep your CRM, project management, and invoicing tools in sync without manual data entry. When a deal closes in your CRM, n8n automatically creates the project in your PM tool, generates the invoice, sends a welcome email, and updates your analytics dashboard. Use the JSON formatter to verify data structures when mapping fields between different APIs. This eliminates the "copy-paste between 5 apps" problem that plagues most small businesses.
Bonus tip: Combine multiple workflows using n8n's "Execute Workflow" node. Your lead scoring workflow can trigger your email workflow, which can trigger your CRM sync workflow — creating a fully autonomous business process.
One of n8n's biggest advantages is its pricing model. Here's the complete breakdown so you can choose the right plan for your needs.
Self-Hosted (Free Forever)
The self-hosted version uses a sustainable-use license — free for companies earning under $5M annually. You get unlimited workflows, unlimited executions, all features, and complete data ownership. The only cost is your server (a $5/month VPS from DigitalOcean or Hetzner works fine for most businesses). This is the option most Indian startups and small businesses should choose.
n8n Cloud Starter
The cloud free tier gives you 5 active workflows and 5,000 executions per month. No credit card required. This is enough for testing, personal projects, or a very small business. The Starter plan at $24/month removes these limits and adds priority execution and more storage.
n8n Cloud Pro
At $60/month, the Pro plan adds SSO, advanced permissions, and dedicated support. This is for teams of 5-20 people who need collaboration features but don't want to manage their own infrastructure. For larger enterprises, custom pricing is available with SLA guarantees.
Cost comparison for Indian businesses: A Zapier Professional plan costs roughly Rs 4,200/month (Rs 50,000/year). n8n self-hosted on a Rs 400/month VPS gives you more features for Rs 4,800/year — a 90% saving. That's money better spent on growing your business. For more ways to optimize costs, see our guide on AI tools for small business.
Is n8n free to use?
Yes. The self-hosted version is completely free for companies under $5M revenue — you get unlimited workflows, unlimited executions, and every feature. The cloud version has a generous free tier with 5 active workflows and 5,000 executions per month. Unlike Zapier, which limits you to 100 tasks on the free plan, n8n's free tier is actually usable for real business workflows.
Can n8n replace Zapier?
For most use cases, absolutely. n8n has 400+ native integrations, and for anything missing, the HTTP Request node can connect to any API on the internet. Where n8n actually surpasses Zapier is in AI integration (native support for all major models), code flexibility (JavaScript and Python nodes), and data privacy (self-hosting). The only area where Zapier wins is the sheer number of pre-built integrations (6,000+) and slightly easier setup for non-technical users.
What AI models does n8n support?
n8n has native nodes for OpenAI (GPT-4o, GPT-4 Turbo), Anthropic (Claude 3.5 Sonnet, Claude 3 Opus), Google (Gemini Pro, Gemini 1.5), Ollama (any local model like Llama 3, Mistral, Phi), Hugging Face models, and any OpenAI-compatible API. You can also use the HTTP Request node to connect to any AI API that isn't natively supported. Local model support via Ollama is particularly powerful — run AI workflows with zero API costs and complete data privacy. Learn more about running AI locally in our detailed guide.
Do I need coding skills for n8n?
No, but they help. 90% of n8n workflows can be built with zero code using the visual drag-and-drop editor and pre-built nodes. However, knowing basic JavaScript unlocks powerful custom logic through code nodes — loops, data transformations, conditional branching, and API response parsing. If you can write a simple if/else statement, you can use n8n's code nodes. For complex workflows, the JSON formatter tool is invaluable for debugging data flow between nodes.