n8n

Freemium

n8n is a workflow automation platform for technical teams. You build flows by dragging nodes on a canvas and filling in settings.

It sits in workflow automation, but with a twist: you can drop JavaScript or Python code into any step. You can also self-host it for free.

The catch is the learning curve. Everything assumes you understand APIs, JSON, and some server basics. It is not a tool you pick up in an afternoon.

Open it if you need deep control and can write code. Skip it if you want something that just works without thinking.

These are the people who will get value from the trade-offs.

• Technical teams comfortable with JavaScript or Python: You get custom logic in a visual builder. No other no-code tool matches this.
• Teams needing advanced workflow logic beyond simpler tools: Conditional branching and multi-path flows work without hitting limits.
• Budget-conscious operators avoiding per-action billing: Execution-based pricing means a 50-step workflow costs one execution, not fifty.
• Developers who want self-hosted infrastructure control: The free Community Edition runs unlimited workflows on your own server.
• Teams with compliance requirements needing data residency: Self-hosting keeps data on your hardware, not theirs.
• Agencies managing multiple client workflows cost-effectively: One self-hosted instance handles many clients without per-seat costs.
• AI developers building custom agentic workflows: AI nodes with memory, RAG, and human-in-the-loop controls integrate into larger processes.
• Organizations with high automation volumes: Per-execution pricing saves money when task-based competitors would charge per step.

These are the parts that keep you using it after the setup pain.

• Self-hosted Community Edition is free with unlimited executions: You pay server costs only. No vendor lock-in, no credit limits.
• Code nodes for JavaScript and Python anywhere in the flow: When the visual builder cannot express something, you write it directly.
• Execution-based billing beats task-based competitors: Complex workflows do not multiply your costs. This matters for data-heavy processes.
• Live execution watching and step-level debugging: You see exactly what failed and rerun just that step. No starting from scratch at 2am.
• 1,500+ integrations plus HTTP Request node for any REST API: Built-in nodes cover most apps. When they do not, you wire custom calls.
• AI agents integrated into workflows, not triggered externally: Agents with memory, RAG, and human approval live inside your automation, not bolted on.

You can self-host or use cloud. Self-hosted is free but needs technical setup.

1. Install n8n with Docker or npm, or sign up for cloud.
2. Drag nodes onto the canvas and connect them to show data flow.
3. Fill each node's settings, using expressions like {{ $json.customer.email }} to pull data through.
4. Test with real data and watch executions live to catch errors.
5. Deploy and let it run, checking execution history when things break.

These are the real limits that push people away.

• Steep learning curve for non-technical users: The interface looks visual, but you regularly need JSON and JavaScript expressions. Error messages assume technical context.
• Cloud support is slow and unresponsive: Lower plans get forum-first support. Multiple users report 24+ hour email delays and no account recovery even with proof of ownership.
• Self-hosting transfers risk to you: Security patches, backups, scaling, and uptime become your job. Recent vulnerabilities hit self-hosted instances.
• Token consumption is unpredictable and costly: Debugging AI workflows burns through credits fast. The billing descriptor "paddlecom" confuses people checking statements.
• Interface gets cluttered on complex workflows: Large flows become hard to follow. The three-column UI feels awkward for data mapping on smaller screens.