The Canvas: Workflows, Nodes, and Connections

The Editor Interface

When you open n8n and create a new workflow, you land on the canvas — a large, dotted grey grid that stretches infinitely in every direction. This is your workspace. Everything you build lives here.

The left sidebar lists all your workflows. The top bar has the workflow name, execution controls (Run, Save, Activate), and settings. The canvas itself is where you drag, connect, and configure nodes.

What a Workflow Is

A workflow is a collection of connected nodes that automate a process. Think of it as a recipe: it starts with a trigger (something that kicks it off), followed by a sequence of actions (steps that do the work). When the trigger fires, n8n runs the workflow from left to right, passing data through each node in sequence.

Every workflow has one of two states: inactive (it only runs when you manually click "Run") or active (it runs automatically whenever the trigger condition is met — on a schedule, when a webhook fires, when a new email arrives).

Node Types

n8n has four categories of nodes:

• Trigger nodes — Every workflow starts with one. A trigger defines when the workflow runs. Common triggers: Schedule (cron), Webhook (HTTP call from another app), Gmail (new email), Slack (new message). You can spot trigger nodes by their lightning bolt icon.
• Core nodes — Built-in logic and utility nodes: IF, Switch, Code, Edit Fields, Merge, Loop Over Items, HTTP Request, Wait. These are the building blocks of flow control and data manipulation.
• Action nodes — App-specific nodes that interact with external services: send a Gmail, update a Google Sheet, post to Slack, create a HubSpot contact. n8n has 400+ action nodes.
• Cluster nodes — Groups of nodes that work together, primarily for AI features. The AI Agent cluster node, for example, uses a root node (the agent) connected to sub-nodes (LLM, memory, tools).

How Data Flows

Data in n8n flows as JSON items. Every node receives a list of items from the previous node, does something with them, and passes a new list of items to the next node. Each item is a JSON object.

For example: a Google Sheets node reading a spreadsheet returns each row as a separate JSON item. A Gmail node that sends an email receives those items and can use values from them (like an email address in column A) as parameters. The next node receives one item per email sent.

You connect nodes by clicking the output circle on the right side of one node and dragging to the input circle on the left side of the next. A grey arrow appears between them — that is the connection, and it is the path data travels.

Adding and Configuring Nodes

To add a node: click the + button to the right of any existing node (or press Tab on the canvas), search for the node you want, and click it. The node opens a configuration panel on the right where you set its parameters. Most nodes need credentials (covered in the next lesson) and at least one operation selected.