How to Use n8n: What It Can Do and Building Your First Workflow

Steps and pricing in this article were checked against n8n's official documentation and pricing page on October 2, 2026. Screen names and steps can change. Before use, check the latest information through the sources at the end.

The shortest path into n8n is to run one official tutorial from start to finish. This article follows those steps and fills in the points where beginners tend to get stuck.

What n8n is: a tool that automates work by connecting nodes

n8n is a workflow automation tool. Its official documentation describes it as a fair-code licensed tool that combines AI capabilities with business process automation. You place components called nodes on the canvas and connect them with lines to build a process flow.

It is pronounced “n-eight-n.” According to its GitHub description, the name is short for nodemation. “Node” comes from the node-based interface and Node.js, and “mation” from automation.

The version you run on your own server is published under the Sustainable Use License. License points to know if you sell what you build are explained in n8n vs. Dify vs. GPTs.

What you can do with n8n

In short, n8n connects services and runs set processes on set triggers. Its GitHub description says it offers over 1500 integrations and more than 9,000 workflow templates.

• Run on a schedule: collect data every Monday and send it
• Run when called from outside: process a form submission or webhook
• Split data: use an If node to branch the flow by condition
• Add AI: use the AI Agent node to let an AI model use tools
• Add code: write code for processing the visual nodes cannot handle

According to the pricing page, n8n's paid cloud plans are priced by the number of workflow executions per month. However many steps a workflow has, one full run from start to finish counts as one execution.

Getting started: Cloud or self-hosted

There are two ways to start with n8n: use n8n Cloud, run by n8n, or install it on your own computer or server. The official documentation recommends n8n Cloud for new users.

Starting with n8n Cloud

n8n Cloud has a trial that lets you try Pro plan features for 14 days. The pricing page says no credit card is needed to start the trial. If you continue, monthly pricing billed annually is €20 for Starter and €50 for Pro. Monthly executions are 2,500 for Starter and 10,000 for Pro.

Starting with self-hosting

To run it in your own environment, install Docker, then run curl -fsSL https://get.n8n.io | sh in the terminal. The official documentation calls this the one-line setup. Once it starts, open http://localhost:5678 in your browser.

The same documentation explains that from n8n 3.0, n8n will be distributed only through Docker and npm installation will no longer be available. n8n 3.0 is scheduled for release in October 2026. If you are starting now, choose the Docker method. You prepare and update the server yourself.

Your first workflow: from trigger to action

In the official tutorial, you build a workflow that pulls solar flare data from NASA's public API, splits it into two groups by size and sends the results. It runs automatically once a week.

1. Create a new workflow. On the first screen, choose Start from Scratch or Create Workflow
2. From Add first step, add a Schedule Trigger. Set the interval to Weeks, every 1 week, Monday at 9:00
3. Add a NASA node and choose Get a DONKI solar flare. Get a key for NASA's public API and save it as a credential
4. Put an expression in the start date to limit the data to the last 7 days. Use Execute step to run just that node and see the result
5. Add an If node and set a condition on whether classType contains X
6. Connect a PostBin node to both true and false, and include classType in the message sent
7. Run the whole thing with Execute Workflow and check the results on the PostBin page
8. To run it automatically every week, click Publish

Two terms in these steps need explaining. Credentials are information such as API keys issued by the connected service. n8n uses them to connect to the service on your behalf. The official documentation warns not to share or expose credentials outside n8n. Expressions are calculations written inside double curly braces. In the tutorial, an expression builds the date 7 days before today and puts it in the start date.

This tutorial covers n8n's core ideas: triggers, credentials, data processing, conditional branching and expressions. When you automate other work, the shape is the same: start from a trigger, chain the processing and send the result at the end.

Common pitfalls

Where beginners most often get stuck is the difference between running manually and running automatically after publishing.

• Forgetting to publish: building a workflow does not make it run automatically. Clicking Publish activates the schedule or the production webhook URL
• No trigger: even when you run only some nodes, they must be connected to a trigger. Connect a manual trigger to test
• No data arriving: you cannot select fields in the next node until you run the previous node once
• External service limits: the PostBin bin in the tutorial disappears 30 minutes after creation. Create a new one when the time is up
• Handling credentials: save API keys and similar in n8n credentials and do not expose them

Pinning saves a node's output so later tests use that data. You can test without calling external services again and again. For a failed production execution, you can load its data into the editor from the Executions screen and fix it.

How saving and publishing relate

n8n saves workflows automatically while you edit, about 1 to 5 seconds after a change. However, saved changes remain a draft. Production executions run the version from when you published. If you edit after publishing, the change does not reach production until you click Publish again.

How executions are counted

Only production executions count toward paid plan limits. How they are counted depends on the trigger type.

• Schedule Trigger: one per run, whether it succeeds or fails
• Polling triggers: one only when new data is found
• Webhook: one per request that fires the trigger
• Not counted: manual executions, sub-workflows called from other workflows, and error workflows

If you want to be notified when a production run fails, create an error workflow. Build a workflow that starts with an Error Trigger node and select it in the original workflow's settings. You can then get an email or chat message when a run fails.

What to learn next: AI nodes and integrations

Once your first workflow runs, try adding AI. Connect a chat model and tools to the AI Agent node, and the AI chooses tools to get the work done. The AI Agent node needs at least one tool connected. Steps are in Building an AI agent with n8n.

According to the pricing page, all cloud plans include agents, MCP and AI nodes. How to connect other services is covered in n8n integrations.

Getting others to use your workflow

Once you have a workflow that helps with business tasks, you can also offer it to other companies. Employee Store is a marketplace where companies can adopt AI agents built by developers, with a one-time purchase or a monthly plan. n8n workflows can be listed too. There is no listing fee and no upfront cost. The commission is 20% of the transaction amount and applies only when a sale closes. See the seller guide for details.