n8n Tutorial for Beginners: Build Your First Workflow in 6 Steps

n8n Tutorial for Beginners: Build Your First Workflow in 6 Steps

In short5 min read

  • The n8n Cloud trial needs no credit card and allows 1,000 executions.
  • Every production workflow needs at least one trigger node to run on its own.
  • Only production executions count toward quotas; manual test runs do not.
  • Debug in editor reloads a failed run's data so you can fix and re-run it.

The fastest way to learn n8n is to build one small, real workflow end to end, then break it on purpose and fix it. n8n’s own Build your first workflow quickstart does exactly that: a workflow that fetches the weather for Berlin every morning, turns it into a short report, and saves it to a data table. This tutorial walks through that build manually in six steps, explains the four concepts beginners trip over (nodes, triggers, credentials and executions), and shows how to debug a failed run.

You need an n8n instance first. The quickstart lists two options: the n8n Cloud free trial or a self-hosted install. If you’re unsure which suits a one-person business, read our guide to n8n self-hosted vs cloud before you sign up.

In this article

  1. Before You Build: Picking Where n8n Runs
  2. Four Concepts That Make Everything Else Click
  3. Build Your First Workflow in 6 Steps
  4. Debugging a Failed Execution
  5. Your First Week With n8n

Before You Build: Picking Where n8n Runs

As of November 2026, the n8n pricing page says no credit card is required for the n8n Cloud trial. The trial includes Pro-level features with limits of 1,000 executions, 5 concurrent executions, a 180-second execution timeout and one team project. That’s plenty for learning. The same page notes that a standard, self-hosted Community Edition is available on GitHub.

Paid plan prices on the official page are listed in euros, so check the official pricing page for what applies to you. If you’re still choosing between tools rather than plans, our n8n vs Make vs Zapier comparison covers that decision.

Your first n8n workflow, step by step: Create the workflow, Schedule Trigger, HTTP Request, Edit Fields (Set), Data table, Test and publish

Four Concepts That Make Everything Else Click

Nodes

n8n’s documentation on nodes defines a node as “an entry point for retrieving data, a function to process data, or an exit for sending data.” Every workflow is a chain of nodes, each passing its output to the next.

Triggers vs actions

Node operations come in two types. Triggers start a workflow in response to an event or condition (they show a bolt icon in the node search). Actions do the work: manipulate data, or perform operations in other systems. The docs are explicit that all production workflows need at least one trigger to decide when they run.

Credentials

Credentials are how n8n authenticates with the services your nodes connect to. The weather example uses a public sample API, so you won’t need any yet. When you connect real accounts later, follow n8n’s step-by-step credentials docs for each service.

Executions

An execution is a single run of a workflow. There are manual executions (you click to test) and production executions (the trigger fires on its own). This distinction matters for your quota: according to n8n’s executions docs, only production executions count, and manual test runs don’t.

Concept What it is Where beginners go wrong
Node One step that fetches, processes or sends data Cramming too much logic into one step
Trigger The node that decides when the workflow runs Forgetting that a workflow needs one to run on its own
Credential Stored authentication for a connected service Setting one up before checking what the node needs
Execution One run of the workflow Testing endlessly without ever publishing

Build Your First Workflow in 6 Steps

This follows the optional manual path in n8n’s quickstart. The data comes from Mockingbird, a sample API n8n provides at https://mockingbird.n8n.io/v1/weather/berlin, so nothing you do here touches a real account.

Step 1: Create the workflow

Start a new, empty workflow and give it a clear name, such as “Weather Log”. Naming habits matter more than they seem; in a month you’ll have twenty workflows.

Step 2: Add the Schedule Trigger

Add a Schedule Trigger node and set it to run every day at 8 AM. This is your trigger: the reason the workflow will run without you.

Step 3: Add the HTTP Request node

Add an HTTP Request node after the trigger and point it at the Mockingbird weather URL. Run the node once to see what comes back. Look at the output panel closely: these fields are what later nodes will use.

Step 4: Add the Edit Fields (Set) node

Add an Edit Fields (Set) node and create a field called report. Use expressions such as {{ $json.city }} to pull values from the previous node into a sentence. The quickstart’s expected output reads like: “The temperature in Berlin is 6.8 degrees C and it’s currently cloudy.”

Step 5: Add the Data table node

Add a Data table node using the Insert row action to save each report. Now every run leaves a record you can check.

Step 6: Test, then publish

Click Execute to run the whole workflow manually and confirm a row appears in your data table. Then publish the workflow so the schedule takes over.

Prefer chatting to clicking? The quickstart’s main path uses n8n Assistant: you describe the workflow in a prompt and approve its steps. It’s worth trying after you’ve built one by hand, because you’ll understand what it produces. The docs list Assistant availability for Cloud Starter and Pro and for several self-hosted editions.

Debugging a Failed Execution

Every real workflow eventually fails: an API times out, a field comes back empty, a format changes. Here’s the toolkit n8n documents for that moment.

Load the failed run into the editor. n8n’s guide to debugging executions explains that you can open the Executions tab, select a failed execution and choose “Debug in editor”. n8n copies that execution’s data into your current workflow and pins it in the first node, so you can fix the problem and re-run with the exact data that broke it. For successful runs the option is “Copy to editor”. The docs list this feature for all n8n Cloud plans and for Registered Community, Business and Enterprise self-hosted editions.

Pin data while you build. Data pinning saves a node’s output and reuses it instead of fetching fresh data on every test. It keeps you from hammering an API while you adjust later nodes. Limits to know: it’s for development only, works on nodes with a single main output, and can’t pin binary data.

Set up an error workflow. n8n’s error handling docs describe an error workflow that runs when an execution fails, for example to send an email or Slack alert. It must start with the Error Trigger node, and one error workflow can serve many workflows. You set it per workflow under the workflow’s settings.

Use node settings for flaky steps. Each node has settings such as Retry On Fail and an On Error option (stop the workflow, continue, or continue using the error output). Retry is a sensible first step for an occasionally slow API.

Your First Week With n8n

Rather than repeating the build, here’s how to turn one tutorial into a working skill over the next few days:

  • Day 1: Finish the weather workflow and let it run on schedule overnight.
  • Day 2: Open the Executions tab and read what a successful production run looks like.
  • Day 3: Break it deliberately (change the URL), then fix it using “Debug in editor”.
  • Day 4: Build a shared error workflow that alerts you, and attach it to the weather workflow.
  • Day 5: Replace the sample API with one real task from your business, such as the patterns in our guide to automating lead follow-up with n8n or Make.

For more workflows built for one-person businesses, browse our n8n automation hub for solo founders.

Sources: n8n pricing page, n8n Docs (quickstart, nodes, debug executions, data pinning, error handling), checked November 2026.

Build the Berlin weather workflow before lunch

Open an n8n instance and build the four-node weather log: Schedule Trigger, HTTP Request, Edit Fields and Data table. Run it once by hand, then publish it and check tomorrow's execution.

Get new AI automation playbooks by email

One email when new guides and case studies go up. No spam, unsubscribe anytime.

Leave a Reply