What Is n8n? A Plain-English Guide

n8n is a workflow automation tool. You connect the apps your business already uses, describe what should happen and when, and n8n does the work automatically: a form submission becomes a CRM contact, a new Stripe payment becomes an invoice and a Slack message, a weekly report assembles itself from three data sources and lands in your inbox.

The name is short for nodemation, node plus automation, with eight letters between the two n’s. You build workflows on a visual canvas out of nodes: each node is one step, like “watch for new rows in Google Sheets,” “call this API,” or “send this email.” Connect nodes with lines and you have a workflow.

Two things separate n8n from the automation tools you have probably already met:

1. You can run it yourself. n8n is source-available and self-hostable. Your data, credentials, and workflows can live on your own server instead of a vendor’s cloud.
2. It does not stop at simple triggers. Branching, loops, error handling, custom JavaScript, and AI agent nodes are all first-class. It scales from “forward attachments to Drive” up to systems that run real business operations.

What can n8n do?

n8n connects more than 400 services out of the box: CRMs, spreadsheets, email, Slack, databases, payment processors, ad platforms, and anything else with an API via the generic HTTP Request node. The pattern is always the same: a trigger starts the workflow (a schedule, a webhook, a new record somewhere), and steps transform and route the data.

Typical things teams automate:

• Lead handling. Capture a form, enrich the contact, score it, route it to the right person. See the lead capture use case for a concrete walkthrough.
• Reporting. Pull numbers from ads, analytics, and your CRM into one scheduled digest. This is the classic automated reporting use case.
• Content operations. Draft, review, publish, and distribute content across channels on a schedule.
• Back office. Invoices, order processing, onboarding checklists, notification sequences. Browse the use case library to see the shapes these take.

If you would rather start from something working than from a blank canvas, we maintain a set of importable n8n templates with the workflow JSON included.

Is n8n free?

There are two ways to run n8n, and the honest answer differs by path:

• Self-hosted Community Edition: free. Unlimited workflows, unlimited executions, the full integration catalog. Your only cost is the server it runs on, which for most small setups is a modest VPS. This is the path we use and recommend for teams that want ownership.
• n8n Cloud: paid. The hosted service starts at $20 per month billed annually ($24 month to month) for 2,500 executions, with higher tiers as volume grows. As of April 2026 every plan includes unlimited active workflows; you pay by execution.

Pricing is execution-based: one workflow run counts as one execution no matter how many steps it contains. That single fact is why complex workflows that would cost hundreds of dollars a month in per-task tools often cost almost nothing on n8n.

On licensing: n8n is fair-code under its Sustainable Use License, not OSI open source. In practice, using n8n to automate your own business, or building workflows for clients on instances they own, is allowed and free. Reselling n8n itself as a hosted service or embedding it in a paid product is not. For the details, see the license text.

n8n vs Zapier vs Make

The short version:

Zapier is the easiest to start with and the most expensive to scale. Make sits in the middle. n8n asks a little more of you up front and gives back control, capability, and a dramatically better cost curve at volume.

We keep detailed, honest comparisons if you are choosing right now: n8n vs Zapier and n8n vs Make.

Can n8n build AI agents?

Yes, and this is where n8n has pulled ahead of the older automation tools. n8n ships an AI Agent node with tool calling, memory, and support for OpenAI, Anthropic Claude, Google Gemini, and self-hosted models. An agent in n8n can read a request, decide which of your systems to query, act, and report back, all inside a workflow you can inspect and control.

The practical wins are things like support triage, research assistants wired to your own data, and content pipelines with an approval step before anything ships. Start with n8n AI agents explained, then the hands-on LangChain agent tutorial when you want to build one.

How do I get started with n8n?

The fastest path from zero to a working workflow:

1. Pick a home for it. n8n Cloud for zero setup, or a small VPS if you want the free self-hosted route. The self-hosting tutorial covers both AWS and DigitalOcean.
2. Build one small, real workflow. Not a toy: pick a task you actually do weekly. Our first workflow in 10 minutes tutorial walks a webhook-to-Slack build end to end.
3. Steal structure from templates. Import a template close to your use case and modify it rather than starting clean.
4. Add error handling before you rely on it. The difference between a demo and a system you trust is what happens when a step fails. When you get there, read our production error handling guide.

If you want the longer roadmap from first workflow to a real automation system, the n8n automation guide is the deeper companion to this page.

When is n8n the wrong tool?

Credibility requires saying this part out loud. n8n is a poor fit when nobody on the team is willing to touch anything technical, when you need one trivial two-step zap and nothing more, or when a dedicated product already solves the exact problem better than a general-purpose workflow ever will. We wrote up the specific cases in why you shouldn’t use n8n.

For everyone else, the combination is hard to beat: real automation capability, AI agents included, at a cost that stays flat while your volume grows, on infrastructure you can own.