Qlaudia

Dark mode

n8n

Technology & Computing

Connect apps and APIs through visual workflows using nodes that exchange data and trigger actions automatically. Build sequences without coding, then extend them with JavaScript or Python when needed—all while keeping your data self-hosted and under your control.

What you'll learn

NodeThe fundamental building block of an n8n workflow that represents one discrete step, receives input data, performs an action, and produces output data.
Data ItemA JSON object containing structured data and optional binary content that represents a single discrete record traveling between workflow nodes.
Workflow ExecutionA workflow execution is a single, complete run of an n8n workflow from when a trigger fires through every connected node in sequence until the final node finishes or an error halts the run.
CredentialsEncrypted records that store sensitive authentication information such as API keys, passwords, and tokens needed to communicate with external services.
n8n WorkflowA visual automation pipeline made up of connected nodes that executes a self-contained task automatically when triggered by an event.
Execution LogA detailed per-run record of every node that executed in a workflow, capturing input data, output data, status, timing, and error details for inspection at each step.
Action NodeAny node in a workflow that performs an operation by receiving data from the previous step, executing a configured operation, and outputting the result.
Workflow ActivationWorkflow activation is switching a workflow from a passive, editable state into an active, production-ready state where its trigger continuously listens for real events and executes the workflow automatically.
ExpressionsDynamic code snippets enclosed in double curly braces that evaluate at runtime to pull data from previous nodes, built-in variables, or JavaScript logic into node configuration fields.
Error HandlingTechniques for intercepting workflow failures, logging them, and executing recovery actions to prevent silent data loss and unexpected execution stops.
Node ConnectionA directional link between two nodes that defines both the execution order and the channel through which output data from one node flows as input to the next.
AI Agent NodeA mechanism for building autonomous goal-directed automations powered by a large language model that iteratively reasons about which tools to use, invokes them, and inspects results until a goal is satisfied.
LLM Chain NodeA node that sends a prompt template to a connected language model, renders expressions for each item, and returns the model's text response as a standard workflow data item.
HTTP Request NodeA versatile built-in node that sends HTTP requests to any URL with configurable methods, headers, body, and authentication to integrate with REST APIs and web services.
Trigger NodeThe starting point of a workflow that listens for a specific condition and initiates workflow execution when that condition is met, requiring activation to register with n8n's runtime.
n8n Cloudn8n Cloud is the fully managed, Software-as-a-Service version of n8n where the n8n team hosts and operates the platform and handles all infrastructure concerns on your behalf.
Self-Hosting n8nDeploying and operating the n8n application on your own infrastructure rather than using the managed cloud service, giving you full control over data, configuration, and maintenance.
Memory Node (AI)A specialized node that stores and retrieves conversation history so AI agents can maintain coherent context across multiple interactions instead of treating each exchange as stateless.
Chat Model CredentialA configuration object that securely stores the secret information required to authenticate with a language model provider's API.
Embeddings NodeA node that converts text into fixed-length numerical vectors that encode semantic meaning for use in semantic search and vector storage.
Sub-workflow (Execute Workflow Node)A pattern where one workflow triggers another workflow using the Execute Workflow node, enabling modularity, reuse, and separation of concerns across automations.
Webhook TriggerA node that starts a workflow in response to an incoming HTTP request from an external system, receiving the request payload and passing it downstream as the workflow's initial data.
Merge NodeA node that combines data from two or more separate workflow branches back into a single unified flow for continued processing.
Environment VariablesConfiguration values set at the server or process level that govern how the entire n8n application behaves across all workflows.
IF NodeA logic node that evaluates a condition and routes each incoming item to one of two branches: one for items where the condition is true, and another where it is false.
Workflow TemplatesA workflow template is a pre-built, shareable workflow configuration that another n8n user can import with a single click and adapt immediately.
Tool Node (AI)A sub-node wired to an AI Agent that exposes a named capability with a description and input schema, allowing the agent to invoke it during reasoning to interact with external systems and data.
Set NodeA utility node that creates, overwrites, renames, or removes fields on data items flowing through a workflow using a point-and-click or JSON interface.
Code NodeA node that lets you write and execute custom JavaScript or Python code directly inside a workflow for transformations and logic that built-in nodes cannot provide.
Vector Store NodeA node that connects to a vector database to store and query numerical vector representations (embeddings) of text, enabling semantic similarity search and supporting Retrieval-Augmented Generation patterns.
Loop / Split in Batches NodeA mechanism for iterating over large lists in fixed-size chunks, processing each batch separately before moving to the next to avoid API rate limits and manage memory efficiently.
Pinned DataA debugging feature that freezes a node's output to manually defined sample data, bypassing the node's actual execution during test runs so downstream nodes can be tested without re-triggering the workflow.
App Integration NodeA purpose-built node designed for one specific third-party service that handles authentication, API calls, response parsing, and error handling without requiring manual HTTP requests.
Schedule TriggerA trigger node that automatically starts a workflow at predetermined intervals or times without requiring an external event, functioning like a cron job within n8n.
Switch NodeA node that routes each incoming item to one of multiple output branches based on matching rules, providing more than two paths for logic with three or more distinct outcomes.

About this course

n8n Overview

What Is n8n?

n8n (pronounced "n-eight-n", short for "nodemation") is an open-source workflow automation platform. It lets you connect apps, APIs, and services together so that they can exchange data and trigger actions automatically — without you having to write a full application from scratch. Think of it as a visual programming environment where the "programs" you write are sequences of connected steps, and each step does one specific thing: fetch data from a URL, send an email, write a row to a spreadsheet, or transform a piece of text.

If you have used tools like Zapier or Make (formerly Integromat), n8n occupies the same conceptual space. The key differentiators are:

Self-hosting: You can run n8n on your own server or machine, keeping your data and credentials entirely under your control.
Code-friendly: When built-in nodes aren't enough, you can drop down to JavaScript or Python inside the workflow itself.
Open source: The source code is publicly available, meaning the community can inspect it, extend it, and contribute new integrations.

The Core Mental Model: Workflows and Nodes

Everything in n8n is built around two concepts: workflows and nodes.

A workflow is a directed graph — a set of steps connected by arrows that define the order in which data flows. You build workflows visually in the n8n editor by placing nodes on a canvas and drawing connections between them.

A node is a single unit of work inside a workflow. Every node has:

Inputs: data arriving from the previous step (structured as JSON objects called items)
Outputs: data it produces after doing its job, passed along to the next step
Parameters: configuration options that tell the node how to do its job (e.g., which URL to call, which field to read)

A minimal n8n workflow: a trigger node starts the process, action nodes do the work, and data flows left to right as JSON items.

Types of Nodes

n8n ships with hundreds of built-in nodes. They fall into a few broad families:

Family	Purpose	Examples
Trigger nodes	Start a workflow in response to an event	Schedule, Webhook, Email Trigger
App / service nodes	Interact with a specific external platform	Gmail, Slack, Airtable, GitHub
Core / utility nodes	General-purpose data manipulation	IF, Switch, Merge, Set, Code
Flow nodes	Control execution branching and looping	IF, Switch, Loop Over Items

When no dedicated node exists for a service, the HTTP Request node can call any REST API directly.

How Data Flows: Items and JSON

Data in n8n is always represented as a list of items. Each item is a JSON object with a json key containing your actual data. For example, after calling an API that returns a list of customers, n8n represents this as:

\text{items} = \bigl[\,\{\text{"json"}: \{\text{"name"}: \text{"Alice"},\; \text{"id"}: 1\}\},\; \{\text{"json"}: \{\text{"name"}: \text{"Bob"},\; \text{"id"}: 2\}\},\; \ldots\,\bigr]

Each node receives the full list, processes it (often item-by-item), and outputs a new list. This design makes it natural to work with bulk data — one workflow execution can process thousands of records in a single run.

You reference data from previous nodes using expressions, written with double curly braces:

{{ $json.name }}
{{ $node["HTTP Request"].json.status }}

Triggering Workflows

A workflow must have a starting point. n8n supports several trigger mechanisms:

Manual trigger: You click "Execute Workflow" in the editor — useful for testing.
Schedule trigger: The workflow runs on a cron schedule, e.g., every day at 9 AM.
Webhook trigger: n8n exposes a unique URL; any external service that sends an HTTP request to it will start the workflow immediately.
App event triggers: Many app nodes (like Gmail or Typeform) have dedicated trigger variants that poll or listen for new events.

Credentials and Security

Connecting to external services requires authentication — API keys, OAuth tokens, username/password pairs, and so on. n8n stores these as credentials, which are encrypted and reusable across workflows. You configure a credential once and then select it inside any node that needs it, rather than pasting secrets directly into node parameters.

The n8n Editor: Key Areas

When you open the n8n web interface, you interact with three main areas:

Canvas: The visual workspace where you build workflows by dragging, dropping, and connecting nodes.
Node panel: A searchable sidebar listing all available nodes, filterable by category or keyword.
Execution log: A history of past workflow runs, showing what data each node received and produced — essential for debugging.

Expressions and the Code Node

n8n's expression language lets you reference and transform data inline using JavaScript syntax wrapped in {{ }}. For more complex logic — conditionals, loops, custom calculations — the Code node lets you write full JavaScript (or Python) that runs inside the workflow, receiving items as input and returning transformed items as output.

Self-Hosting vs. n8n Cloud

n8n can be used in two ways:

n8n Cloud: A managed, hosted version run by the n8n team. No server management required.
Self-hosted: You run n8n on your own infrastructure (Docker, a VPS, a local machine). This gives you full control over data, scaling, and customisation — at the cost of managing the service yourself.

For most learners, starting with n8n Cloud or a local Docker installation is the fastest path to building your first workflows.

How the Pieces Fit Together

The diagram below summarises the relationships between the major concepts:

Key concepts in n8n and how they relate: workflows contain nodes; nodes consume and produce items; credentials, expressions, and the execution log support the process.

Why Learn n8n?

Automation removes repetitive manual work. Instead of copying data from one app to another, checking a form response and sending a follow-up email, or polling an API every hour and logging results, you describe the process once in n8n and it runs automatically — reliably, on schedule, at any scale your server can handle. n8n is a practical tool that rewards incremental learning: you can automate something useful on day one with a two-node workflow, and keep adding sophistication as you grow more comfortable.