How To Use N8N AI Workflow Automations For Beginners
Other Videos📅 2025-10-02
Transcript
[00:00] I'm going to show you as fast as possible as if you've never used an automation software in your entire life. In this video, how to leverage N for your business or personal life. Sound good? Let's jump in. To get started here, we're going to create a very simple workflow that integrates AI, specifically OpenAI's API. And if you're wondering, Corin, what the heck is an API? Don't worry about it. Right now, all you need to understand is how to create a workflow from start to finish. So, we're going to do a form submission, create a automated artificial intelligence draft through AI, and then send that automatically through Gmail.
[00:30] So, let's do it. I'm going to come up here to create workflow. Let's begin by creating our trigger event. This is what's going to initiate the entire automation. Now, right now, Nan provides a bunch of different options such as on form submission, web hook call, on schedule, on app event, on schedule, run it every 100 p.m. PST on every Monday. Each one of these has their own specific use case. Today, we're going to do an onform submission. Maybe tomorrow or in a couple days, I'm going to show you how to do more complex workflows with a web hook. So, obviously, make sure to subscribe here. Let's do on form submission. First, we'll add a form title here. We're going to do lawnmowing
[01:02] inquiry. Form description. I'm going to say I need your data. And what we're going to do is we're going to act like a lawnmowing company here that needs relevant information about your property for me to effectively lawn mow your lawn. Field name. I'm going to say name. What is your name? And for now, we can add placeholders here. So, we'll say first name. Make this a required field. And what I want to do is very simple conditional logic where we're going to do two different types of automations depending on what the user inputs. Therefore, we're going to do element type number. I'm going to say how big is your lawn parenthesis square feet. With
[01:34] that down, we do require field. We can add one last field here. Obviously, we have a lot to choose from. Look at that. So many options. It's like when you go to Cold Stone and you're like, "What ice cream do I like?" And then you typically go with the one you always go with. # mintship but cookie dough can hit as well. Come back over here, though. We're going to choose email so we can send up a follow-up email. Oopsy daisy. And that should be good here. So, we're going to require this field as well. Next step for any automation is we need test data. Execute step. So, here we go. We got James, how big is your lawn? 1,400 ft. And then the email use is contact at webcafi.com. This is a real email and I'm using a real email so I can show you
[02:05] this actually works. Wait, a tutorial where the tutorial person, is that what you call me? Actually shows a real automation that works. Yes. Yes. You found gold. All right. Come in over here. Once we execute the step, the date is above me. hit this little pin data. We're going to use this data throughout the flow. Receive data. What do we do next, Corbin? Well, first off, how do we even move around this little playground to move around like this? You're going to use the middle of your little mouse. What's the middle? This thing right here to zoom in, zoom out. We can use these buttons right here. I'm zooming out. I'm zooming in to zoom to fit where it's
[02:36] like, you know, you a little crazy. Like, where am I going? I'm in no man's land. Click it. We're back. And then we'll use the clean up a little later in this tutorial. So, first off, we need to use our open AI. We have an AI build autonomous agent summarize. Okay, nice. They got a bunch of different options here is a lot has changed. You can choose whichever model you really love. I'm going to use open AI for now. And we're going to do the action of message and model. Now, one thing I want you to note here, especially when it comes to AI and why you clicked on this video, is that AI isn't just text in, text out. Eg. I text you, you give me text back. There's a bunch of different stuff we can do such as analyze images, analyze
[03:07] audio, extract audio. But for now, message a model. No, no, no. Not not an IG model, y'all. message an open AAI model. Go ahead and connect your credentials with your OpenAI account. If you're like, Corbin, how the heck do I do that? Check me out here. I show you in a ton of different tutorials. Resource text, text in, text out. I like it. We're going to do message and model. From the model list, all I want you to do is always just go with the most up-to-date model as of now because you're learning still. So, you don't have to worry about all this other craziness. Next is prompt engineering. For now, we'll do a simple prompt here. I'm going to open it up like this cuz this box is too small. I want it bigger.
[03:39] There we go. So, here's our prompt. We're going to add in variables here, don't worry. For now, we got context. We are writing an email as a lawnmower company. Identify who you are. We received a form submission form data. This is how you format data coming in. Eg this stuff up here. If the lawn size is bigger than a thousand square feet, then offer them an extra window cleaning service. Important. Never put in variables that need to be replaced. Eg. When a AI does an output and it's like best regards, your name, and then puts like phone number in brackets, but doesn't actually put a phone number. Let's mitigate that. important is where you identify very important rules that
[04:10] you want the AI model to follow on outputs. Now, what I want you to understand is let's add these little variables to make this dynamic. We're going to do expression. We're then going to click this open. We then get a way better UI. So, from this UI, I can add the relevant information that I care about. So, we're going to name there how big is your lawn? There's that's how big your lawn is. 1,400. And what you'll notice is how it's going to be read by the AI model here. I like it. For now, everything else we're going to say the same. I'm also going to show you how to maybe not necessarily rely on the AI model when it comes to dynamically
[04:41] offering the window cleaning service. Maybe you want a more set way of doing that with a different block. Let's go ahead and just execute step. For right now, we should see a window cleaning service be offered on top with the lawnmowing service. Here we go. We can also add our window cleaning service as well. You might be like, Corbin, what is this slashn? That is a way for the AI to indent the new line. When I say new line, essentially it's right now it's a paragraph. we'd want it a little bit more structured for obvious reasons. I jump into that in other tutorials. For now though, we got our nice little AI output. Coming back over here, if you didn't want to use conditional logic
[05:11] within the AI prompt, we can alternatively attach a different type of block here in nan add low if value. What is the value? We're going to do how big is your lawn? We're going to say is equal to or alternatively since it's a number, we're going to say is greater than. If you've code, you know what I'm talking about. What's the if else? If else. And then we're going to say a th00and. And then these simply the logic that would incur here is out execute step. And then if the underlying lawn is bigger than a th00and it's going to go down true. And then we would set up our
[05:43] relative AI prompt and flow for true. And then if the lawn is smaller than a th00and then we would go for false. And what we would do in that context maybe the lawns that are not 1,000 square feet bigger there. They're not worth their time. So therefore we just send them an email like you know what you're not worth their time. For now though we're going to go and delete this step. We don't need to do this cuz the AI is already doing it for us. attach this and let's do the last block here because you clicked on a very fast tutorial. I want it fast. I'm giving you fast Gmail. Coming down to Gmail here, we're going to say create a draft. Any type of account you need to connect with, you're going to simply add the credentials. And
[06:13] a lot of times you just sign into the relevant account. As a side note, if you're like Corbin, what even connects to NAN? Like what can I do? I'm going to leave this in the description down below. This gives you all the apps and different integrations N currently integrates with. On top of that, don't worry about it. In our more complex tutorial, we're going to go over something called web hooks, which is going to allow us to essentially integrate any type of software found on the internet as long as they provide an API. An API is just a key that tells the software like you own this, like you have access to this. We'll get deeper into that later. From here though, we're
[06:44] going to go ahead and just add a nice little subject. We are then going to choose between text or HTML. For now, I'll just do text. The message, we're going to add the previous output. So, I need to real quickly get that output from this. So, we're going to execute step again. What you have to remember and what I forgot right there is we always have to hit pen in order for that data to be used in the next block for testing. Coffee. It's here again. Pen. Perfect. Great draft. Go to message. I'm going use the content here above my head. Drag it here. And we've successfully added the relevant AI output that would get rid of the message. We can add fixed text here.
[07:15] What I mean by fixed text is that will show up in every single output. So I can say, "Hey, what's up dot?" And then boom, we shall execute this step. And I should automatic wait hold up. I forgot one step. The last step here is going to be the additional options that we would want to add functionality to this block. This is going to be same for any type of software you use where there are going to be task specific things we can do for that application. Eg add an attachment BCC from alias name or ID. For me, I'm going to do a tune email. I want to set it up so there's automatically a draft set to the email provided in the
[07:45] previous output from the form. So come over here to form submission. Click this. Drag the email here. So now you just learned something very fundamental. We grabbed data from two different blocks previous in the workflow. Nice. Execute step. Execute order 66. And what do you know? It showed up in my Gmail as a draft. Lawn more inquiry. Hey, what's up? Subject AI output. It looks perfect. Let me go ahead and fire off my last couple tips about nan before you click off and start doom scrolling. All right, don't doom scroll just yet. I'm still here. And the first tip's got
[08:15] to be, you know, check me out on my live. Okay, it's in the description down below. Watch me live. Just click to Twitch, follow. I'm streaming on YouTube, Twitch, NX, and maybe even Tik Tok pretty soon here every 4 to 7 PM. Just check me out. PST. Here's the next tip. If you're going crazy, because these workflows can go a little crazy and you're going like this, you're checking this around. You're like, "Hold up, hold clean up." And we're back to normal. Next thing, we're going to always want to hit save. In addition, if you want to actually activate this so it works in a line production environment, activate. And with all that done, you've successfully set up your first n
[08:46] workflow. So, make sure to leave a like if you felt like you learned something in today's video. So, it's completely free. I'm going to do a very big tutorial on this platform pretty soon here that's going to be way more complex. This is very much like how do I make an automation? The next one's going to be Whoa, that's a lot. Don't worry about that. Two random videos. That is my face. I'll see you in the next automation.