n8n Tutorial for Beginners 2026 - Full Guide

NADN is an automation tool that can connect practically any software you use and turn hours of manual repetitive work into workflows that run automatically in the background. And by integrating AI models like chat GPT, you can even turn these workflows into intelligent AI agents that can reason, make decisions, and handle complex tasks. In the next 15 minutes, I'll show you how to use NAD to build your first workflow from scratch, even if you're a complete beginner. So by the end you'll have everything you need to start automating your own tasks. Let's start by setting up Naden. Naden cloud costs 20 plus dollars per month, but self-hosting runs under $5 per month with unlimited automations and access to community nodes not even available on the cloud plan. We'll use Hostinger because their one-click template handles all the setup for you. Use the link on screen or in the description for 10% off. The KVM1 plan is more than enough to get started. Pick the closest server to you. Confirm Nad is selected and hit continue. Then follow the prompts to finish setup. Hostinger will set up your server automatically. This takes a few minutes. When it's done, click manage app to open NADN. Bookmark this URL to access NAN directly going forward. Then follow the prompts to set up your NAND account. This is important. In the get paid features for free popup, click send me a free license key. Then check your inbox for the NAN email and activate your license. You're in. Welcome to NADN. Let's build your first workflow. Click start from scratch to begin. This canvas is where you build workflows. Each box is a node that performs one specific task. We'll build a workflow that sends a daily inspirational quote to Slack every morning at 8 a.m. Click the plus icon to open the node library. This is your toolkit of hundreds of available nodes. We'll start with a trigger. Triggers are the events that kick off your workflow. Common triggers include schedule, which is time based, web hook, when another app pings you, and manual, click to run, or app events like new email received. Select schedule trigger and set it for 8 a.m. daily. Hit execute step to run it and generate test data. Quick tip. You see that pin icon? If your trigger is pulling live data from an app, you can pin it to reuse the same test data while building instead of triggering real events every time. Click back to canvas. Click the plus button coming off your trigger to add the next node. Nadm connects to hundreds of apps like Gmail, Slack, Google Sheets. You can see them all under action in an app. But let's start with something simpler. An HTTP request from the core library. HTTP requests let you fetch data from any website or API. Basically asking a URL for information. We'll grab a random quote from Zen quotes, a free quote API. Keep the method as get. That means we're just requesting data, not sending anything. Paste this URL. execute step and check the output panel on the right. And there's our quote. You can see the quote text, author, and some other data the API sent back. Now, let's send this quote somewhere useful. Click the plus button, select action in an app, and search for Slack. You can use Discord, Telegram, Gmail, whatever works for you. The process is similar. Select send a message. First, we need credentials. Nitn needs permission to act on your behalf in Slack. Click create new credential under the dropdown. Most apps require some form of authentication like API keys or OOTH tokens. Each app is a little different, but NAN makes it easy. Click the docs link at the top for step-by-step setup instructions. For Slack, we'll use OOTH 2 since we're sending messages, not just monitoring them. You'll need to create a Slack app. Click the link in the docs to go to Slack's API page. Create an app from scratch. Name it something like quotebot and pick your workspace. Copy the client ID and client secret from Slack and paste them into Nadn. Important. Keep secrets actually secret. These are like passwords between apps. Copy the OOTH redirect URL from NAN back to Slack under OOTH and permissions. Add the redirect URL and save it. In Slack, scroll to scopes. These define what permissions your app has, like post messages or read channels. Add the scopes listed in the NAND docs. Then scroll up and click install to workspace. Authorize it and you're connected. Save the credential in NAND. Now let's configure what Slack actually does. Set send to as user and pick yourself from the drop down or you can choose a channel. For the message content, look at the left panel. You can see outputs from previous nodes. Most importantly, we've got the quote text and the author. Drag the quote into your message field. Notice that it appears as a variable in brackets. This means it'll dynamically change each time the workflow runs. Add a line break. Type a dash and then drag in the author. You can preview the final message below the field. Execute the step to test it. And a green check mark means success. Let's check Slack. And there's your message from the quote bot. Right now, this only runs at 8 a.m. Let's add a manual trigger so you can run it any time. In the upper right, click plus. Add another trigger on app event. manually. Drag it next to your schedule trigger and connect it to the HTTP request node. Now you have two entry points, scheduled daily or on demand whenever you click. Hit execute workflow and everything turns green. Check Slack for your quote. Flip the switch at the top to activate your workflow. And now your automation runs automatically every morning. Don't forget to name it. Click my workflow in the upper left and give it something descriptive like daily quote to Slack. You just built your first real automation. Nice work. You've got the basics down. Now, here's where it gets powerful. So far, we've done simple A to B automation. Trigger fires, fetch data, send message. But real automation needs decision making and logic. That's where flow nodes come in. They let you add conditions, loops, branches, and error handling. Think of them as the if this then that brain of your workflows. Let's make this workflow smarter. We only want quotes on weekdays, not weekends. Hover over the line between schedule trigger and HTTP request. Click the plus that appears. Select flow and if. This creates two paths. True meaning conditions met and false conditions not met. We need to set rules for the true branch to execute. From the left input panel, drag day of the week from the schedule trigger into the value one field. Click the center dropdown and change it to is not equal to. Under string conditions, set value two to Saturday. Click add condition. For a second rule, make sure the drop- down between conditions says and. We need both conditions true, not Saturday and not Sunday, not just one of them. Set up the second condition. Day of the week is not equal to Sunday. Execute the step. Since I'm testing on a Wednesday, the output goes to the true branch. Back on the canvas, it got a bit messy. Click the tidy up button to autoarrange everything. That's the brush icon at the bottom left. See the two branches? True goes to HTTP request. False goes nowhere. So if it's a weekday, the workflow continues. If it's a weekend, it stops. You could add different actions to the false branch or leave it empty to just terminate the workflow. Test the whole workflow with execute workflow. Check slack and there's your quote. Flow nodes add logic to your automation. But wait till you see what AI can do. Let's make your workflows even smarter. A quote without the context isn't very inspiring. So, let's have an AI agent research the author and provide a short paragraph with more info. I've pinned some earlier data so we can test with the same quote. Hover between HTTP request and Slack. Click the plus AI AI agent. Back on canvas, notice the three connections. Chat model is the brain of the agent. Memory tracks conversation and tools are how the agent interacts with the world. We'll wire these up one by one. Starting with the chat model. Click plus under chat model and pick any LLM. I'm using OpenAI. For credentials, we need an API key from platform.openai.com. Like most nodes, the docs walk you through it. The quick version is create an account, generate an API key, paste the API key in NAND, and add at least $5 in credits. Save and select the chat model. I'm using GPT5. Next, let's give our agent memory. Click plus under memory and select simple memory. Since we're not chatting with this agent, we need to set a session ID. Set session ID to define below. And then in the field below that, enter anything. I'll just put session ID. Next, click the plus under tool. Then search for and select Wikipedia. That's it. There's no credentials needed for that one. Next, let's configure the AI prompt. Open the AI agent node and change the prompt source to define below. Here's what we'll tell it to do. You are an agent that looks at a daily quote and its author. Looks up additional information about the author and the context of the quote using Wikipedia. Then provides a short supplementary paragraph to be included with the daily quote to help inspire the reader to have a great day. Add quote, then drag in the quote from HTTP request on the left side. Add author. Then drag in the author. Finally, add only return the paragraph in text format. Nothing else. Test it by clicking the play button on the agent node from canvas. Then watch it work using Wikipedia, the memory, and the LLM together. Check the output when it's done. And there's your enhanced paragraph. Next, we need to update our Slack message. Open the Slack node and drag the AI text output into your message. The old variables are grayed out because we added a node before this one. Just delete them and redrag quote and author from HTTP request. Test it. Check Slack. And there's your AI enhanced message. Now let's run a final full workflow test. Unpin everything. Execute the full workflow with fresh data. Check Slack. And perfect, there's a new quote with context. Finally, let's reactivate the workflow. That's AI agents. Powerful stuff. Now, building your own workflow from scratch is a crucial skill, but why [snorts] reinvent the wheel? Let's steal from the best. Next, we'll explore templates to accelerate your automations. Nadn has a massive template library to jumpstart any automation project. Click the templates button, the open box icon in the bottom left. Search by keyword, app name, or scroll to find what you need. Click use for free to import it directly to your workspace. Set up any required credentials. Many templates include helpful notes, videos, and instructions. Now, let's talk about community nodes. Community nodes are custom integrations built by NAND users. They extend to NADN beyond its 400 plus native nodes, giving you access to virtually any service or tool. To install one, click your logo icon, settings, community nodes, install a node, and browse. Here, find a node you like. Copy its name and paste it into the npm package name field. Check the risk acknowledgement box and click install. To use it in a workflow, just search for it in the node library. Look for the box icon next to the name. You just learned how to self-host NAN, build workflows from scratch, add logic with control nodes, integrate AI agents, and leverage templates and community nodes. You've got everything you need to start automating on your own. Use my Hostinger link below to save 10% on your VPS. you'll get a discount and support the channel so I can keep making tutorials like this one. Thanks for watching. Now go build something awesome.