N8N FULL COURSE 1 HOUR (Build & Automate Anything)

This is your N andN full 1hour course. Learn how to build and automate anything. Today, I'm going to show you the complete N8N course that builds and automates anything you want. This is a compilation of all my recent N8N workflows that have saved businesses thousands of hours every month. You're going to see how to scrape unlimited leads for free, build AI agents that work 24/7, create WhatsApp bots that get customers while you sleep, and automate your entire social media presence. I'm talking about systems that can replace entire teams and cost almost nothing to run. By the end of this course, you'll have everything you need to automate your business completely. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency, Goldie Agency, whilst he's helping clients get more leads and customers. I'm here to help you get the latest AI updates. Julian Goldie reads every comment, so make sure you comment below and let me know which automation you're most excited to build. Look, I've been testing these N8 systems for months now, and what I'm about to show you is going to change how you think about business automation forever. We're talking about replacing manual work with AI that never sleeps, never makes mistakes, and costs basically nothing to run. And what I'm about to show you is absolutely wild. This AI agent doesn't just post random stuff. It actually searches the internet in real time, finds breaking AI news, and creates posts that get engagement. Real engagement, not bot likes. The system I'm showing you today has three main parts. First, it searches the web automatically using something called firecrol. Second, it uses AI to write the perfect tweet. And third, it posts directly to your Twitter account. All of this happens without you doing anything. Now, here's what most people don't understand. Social media growth isn't about posting more. It's about posting the right content at the right time. This automation does exactly that. It finds trending topics and creates content around them instantly. I've been running this on a test account and the results are crazy. The engagement rates are higher than anything I've posted manually. Why? Because the AI knows exactly what content performs well right now. Not what worked last month, what works today. But before I dive into the technical setup, let me show you exactly what this looks like in action. You can see right here on my screen, this automation has already created several posts. Each one is about current AI news. Each one follows proven engagement patterns. And each one was created completely automatically. Now, you might be thinking, "This sounds complicated. " It's not. The whole system runs on something called N. Think of N as the brain that connects everything together. It tells the AI when to search for news. write a post. And it tells Twitter when to publish. Here's the beautiful part. Once you set this up, it runs forever. You could go on vacation for a month and come back to hundreds of new posts, all high quality, all relevant, all posted at the perfect times. Let me walk you through the exact seat up process. And don't worry if you're not technical. I'm going to explain everything in simple terms. Plus, I've already created a JSON file that does most of the work for you. More on that in a minute. The first component is theuler. This tells the system when to run. You can set it to post every hour, every day, or whatever schedule you want. I recommend starting with every 24 hours while you test things out. Here's something most people mess up. They think more posts equals more growth. Wrong. Quality beats quantity every single time. This automation focuses on creating fewer, better posts instead of spamming your audience. The second component is the AI brain. I'm using Chat GPT for this, but you can actually use completely free alternatives. There's something called Open Routter that connects to free AI models. So, this entire system can run without costing you a penny. Now, here's where it gets interesting. The AI doesn't just create random posts. It follows specific instructions that I've tested and optimized. It knows exactly what style of content works on Twitter. It knows which emojis to use. It knows how to create curiosity gaps that make people want to click. These instructions took me weeks to perfect. I tested hundreds of different approaches. I analyze what content performs best in the AI space and I distilled all of that knowledge into a simple prompt that anyone can use. But the real magic happens with the third component, the web scraping tool called Firecrawl. This is what makes the content feel fresh and relevant. Instead of recycling old information, it finds brand new stories that just broke. Firecrawl searches specific websites for AI news. It could be Reddit, Twitter, Google News, or any site you choose. It finds the latest headlines and passes them to the AI. The AI then creates a post about that news. The result, content that feels timely and valuable. The best part about Firecrawl, it's completely free for the first 500 searches. That's enough to run this automation for months without paying anything. Even after that, it costs almost nothing to keep running. Now, let me show you the exact setup process. First, you need to get your Firecrawl API key. You just go to their website, sign up for free, and grab your key from the dashboard. Takes about 2 minutes total. The API key is what lets N8N talk to firecraw. You copy this key

and paste it into your NAM workflow. Once that's connected, your automation can search the internet automatically. Here's something being cool. You can customize exactly what it searches for. Want posts about AI news? Set it to search AI websites. Want posts about crypto? Point it to crypto news sites. Want posts about marketing? You get the idea. The flexibility here is insane. You're not locked into one type of content. You can create multiple automations for different topics, each one running independently, each one growing different aspects of your online presence. But here's where most people get stuck. The Twitter API setup. Twitter recently changed how their API works. It's not as simple as it used to be, but don't worry, I'm going to walk you through the exact process that still works today. You need to create a developer account on Twitter. Then you create an app. Then you get your API keys. The whole process takes about 10 minutes if you know what you're doing and I'm going to show you exactly what buttons to click. The tricky part is the authentication settings. You need to make sure you have read and write permissions. You need to set the call back URL correctly. And you need to configure the oorthth settings properly. Miss any of these steps and your automation won't work. But here's the thing. I've already done all the hard work for you. Remember that JSON file I mentioned? It contains the entire workflow preconfigured. all the settings, all the connections, all the prompts, everything. You just download the file, import it into NAN, add your API keys, and you're ready to go. What would normally take hours of setup takes about 5 minutes. This is the power of having the right systems. Now, let me show you what happens when the automation runs. First, theuler triggers at your chosen time. Let's say every 24 hours at 9:00 a. m. The system wakes up and starts the process. Step one, Firecross searches for the latest AI news. It might find an article about a new AI breakthrough or a story about AI regulation or news about a major AI company. The key is that it's fresh and relevant. Step two, the AI analyzes this news and creates a Twitter post. It doesn't just copy the headline. It adds commentary. It creates intrigue. It uses proven formatting that drives engagement. Step three, the post goes live on your Twitter account automatically. Your followers see valuable, timely content. They engage with it. The algorithm notices the engagement and shows your content to more people. This creates a virtuous cycle. Better content leads to more engagement. More engagement leads to more reach. More reach leads to more followers. More followers leads to more opportunities. I've seen accounts grow by thousands of followers using systems like this. Not fake followers. Real engaged people who are interested in the content. People who become customers, clients, or business partners. But here's what separates this from other automation tools. Most social media bots are obvious. They post generic content. They use poor grammar. They feel robotic. This system is different. The AI I'm using creates content that sounds human. It uses natural language. It makes relevant observations. It starts real conversations. Your audience won't even know it's automated. Of course, you still need to engage with your audience personally. This automation handles the content creation and posting, but you should still reply to comments, join conversations, and build relationships. Think of this as your content engine, not a complete replacement for human interaction. Now, let me address the elephant in the room. Is this ethical? Absolutely. You're not stealing content. You're not misleading anyone. You're creating original commentary on public news. It's no different from having a social media manager. Except your manager is AI. The key is transparency and value. As long as you're providing genuine value to your audience, the method doesn't matter. Your followers care about getting useful information. They don't care whether a human or AI wrote it. Here's another benefit most people don't think about. Consistency. How many times have you started posting regularly on social media only to stop after a few weeks? Life gets busy. You forget to post. Your audience disappears. This automation solves that problem completely. It posts consistently whether you're busy or not, whether you're motivated or not, whether you remember or not. Consistency is one of the biggest factors in social media growth. And this gives you consistency automatically. But what if Twitter bans automated posting? The truth is they don't ban automation when it's done correctly. They ban spam. They ban low-quality content. They ban misleading information. This system creates highquality original content. It follows Twitter's terms of service. It provides value to users. There's no reason for Twitter to have a problem with it. In fact, they want good content on their platform. That said, I always recommend starting with a test account. Don't run this on your main business account right away. Create a separate account, test the system, see how it performs. Once you're confident it's working well, then consider using it on your main account. Here's something else to consider. You can run multiple versions of this automation for different niches. One for AI news, one for marketing tips, one for business insights, each one growing a different aspect of your online presence. The scalability is incredible. Once you understand how this works, you can create entire networks of automated content, all providing value, all growing your influence, all running

without constant supervision. Now, I know what some of you are thinking. This sounds too good to be true. I get it. I was skeptical when I first heard about automated social media systems. But here's the difference. Most automation tools create generic content. They use templates. They feel robotic. This system uses advanced AI to create unique, relevant content every single time. The difference in quality is night and day. Plus, the content isn't just random. It's specifically about AI and technology news. Topics that are always trending, topics that people are actively searching for, topics that drive real engagement. The timing aspect is crucial, too. By searching for news in real time, your posts are always about current events, not last week's news, not last month's trends, today's breaking stories that people are actually talking about. This gives your content a massive advantage in the algorithm. Social media platforms prioritize timely, relevant content. They want to show users the latest information. And this system ensures your content always qualifies. Here's another benefit I didn't mention earlier. This system is a learning tool. By watching what content the AI creates and how your audience responds, you'll start to understand what works in your niche. You'll see which topics generate the most engagement, which writing styles resonate with your audience, which posting times work best. This knowledge will make you a better marketer. Overall, the data you collect from this automation can inform your entire content strategy, not just on Twitter, but across all your marketing channels. You're essentially getting free market research every single day. I figured out how to build nan AI agents in seconds using Claude. Not minutes, not hours, seconds. You can literally take a screenshot of any NAN workflow you find online, any automation you see, any template you want to copy, paste that screenshot into Claude and boom, you get perfect JSON code that you can copy and paste directly into N8N. The entire automation gets created instantly. Every node, every connection, even the sticky notes. I'm going to show you exactly how I do this, the custom Claude project I built, the exact prompts that make Claude an N expert instantly, and how you can replicate any workflow with just a screenshot. But first, let me show you this in action. Here's a screenshot I took of an N810 workflow I wanted to replicate, just a simple automation with a few notes. I pasted this into my custom Claude project, asked it to recreate this workflow, and look what happened. Claude gave me perfect JSON code. I copied this code, went to N8N, pasted it in, and the entire workflow appeared exactly as it was in the screenshot. Same nodes, same connections, even the same sticky notes in the exact same positions. This took me 30 seconds. From screenshot to working automation, 30 seconds. Now, I know what you're thinking. And honestly, when I first discovered this, I thought the same thing. But I've tested this with dozens of workflows. simple ones, complex ones, multi-step automations with API calls and data processing, and it works about 80% of the time perfectly. The other 20% just needs minor tweaks. But here's the crazy part. Most people don't know how to set this up. They don't know the right prompts to use. They don't know what knowledge to feed Claude, and they definitely don't know the tricks that make this work consistently. That's what I'm going to teach you today. By the end of this video, you'll have your own custom Claude project that can replicate any N8N workflow you throw at it. You'll never have to build automations from scratch again. You'll just screenshot what you want and let Claude do the heavy lifting. And before we dive in, Julian Goldie reads every comment. So, make sure you comment below and let me know what automations you want to build. Now, let me show you exactly how to set this up step by step. First, you need to create a custom clawed project. This is where the magic happens. Without this setup, Claude won't understand N8's JSON format. It won't know how to structure the nodes, and it definitely won't create working automations. But with the right setup, Claude becomes an N8 expert instantly. Here's how you do it. Go to Claude and click on projects in the left sidebar. Then click new project. Give your project a name. I call mine N8N AI agent builder, but you can call it whatever you want. Now, here's where most people mess up. They just start asking Claude to build workflows without giving it the right knowledge. That's like asking someone to build a car without showing them how engines work. You need to feed Claude the right information first. Click on the project you just created. Then scroll down to project knowledge. This is where you're going to upload all the information Claude needs to become an N8N expert. Now, I've already done all the hard work for you. I've created a complete knowledge base with N8N documentation, JSON examples, and workflow templates. I'll tell you how to get all of this for free in just a minute, but first let me show you what goes into this knowledge base. You need the N8N JSON cheat sheet. This tells Claude exactly how N8N's JSON format works, how nodes connect to each other, how delta flows between nodes, and how to structure everything properly. You also need example workflows. These show Claude what good NAN automations look like. Simple ones, complex ones, different types of integrations. The more examples you give

Claude, the better it gets at creating new workflows. And you need the custom instructions. This is the prompt that tells Claude exactly what you want it to do, how to behave, what format to use, and how to structure its responses. Now, I spent hours figuring out the perfect prompt, testing different approaches, refining the language, making sure Claude understands exactly what I want. And I'm going to give you that exact prompt for free. Here's what it looks like. You are an N expert and workflow generator. Your job is to create fully functional nan workflow JSON that can be directly pasted into N to create working automations. The JSON should be complete, valid, and ready to use without any modifications. But that's just the beginning. The full prompt is much longer and much more detailed. It includes specific instructions for node positioning, connection handling, and error prevention. I'll put the complete prompt in the AI success lab so you can grab it for free. Once you have your clawed project set up with the right knowledge and instructions, you're ready to start building. Here's where it gets fun. Take a screenshot of any N8N workflow you want to replicate. It could be from a tutorial, a template site, or even your own workflow that you want to modify. Go to your Claude project, paste in the screenshot, and say something like, "Recreate this workflow in N8NJS and format. " Claude will analyze the image, understand the workflow structure, and generate perfect JSON code that you can copy and paste directly into N. Let me show you another example. Here's a more complex workflow with multiple nodes and integrations. I'll paste this into Claude and ask it to recreate it. Look at this response. Claude not only recreated the workflow structure, but it also added proper node configurations, connection mappings, and even positioning coordinates. So everything appears in the right place. I can copy this JSON, go to NADN, create a new workflow, and paste it in. And boom, the entire automation appears exactly as it was in the screenshot. But here's where it gets even better. You can also ask Claude to modify workflows on the fly. For example, I can say, add a Google Sheets node to this workflow so I can output the results to a spreadsheet. Claude will take the existing JSON and add the new node with all the proper connections. No manual work required. or I can say change this web hook trigger to a schedule trigger that runs every hour. Claude understands the workflow structure well enough to make these modifications automatically. Now, I want to be honest with you, this doesn't work 100% of the time. Sometimes Claude makes small mistakes. Sometimes node configurations need tweaking. Sometimes connections aren't quite right. But even when it's not perfect, it gives you about 80% of the work done automatically. And that 80% can save you hours of manual building. The key is knowing how to prompt Claude correctly, and that comes from practice and the right setup. Speaking of setup, let me show you an even more advanced approach. Instead of just uploading my pre-made knowledge base, you can actually connect Claude directly to N8N's official documentation on GitHub. Here's how you do it. In your Claude project, go to project knowledge, click the plus sign, then click link. Now you can paste in the URL to N810's official GitHub documentation. Claude will automatically pull in all the latest information about NA10 nodes, features, and best practices. This means your Claude project stays up to date with the latest NAN features automatically. You can even be selective about which parts of the documentation to include. Maybe you only want information about specific integrations or certain types of workflows. Claude lets you choose exactly what knowledge to include. Now, here's something really cool that most people don't know about. Claude now has extended thinking and web search features. This means when you're asking it to build workflows, it can actually search the web for the latest information about APIs, integrations, and best practices. So, if you're building a workflow that connects to a new service, Claude can search for the latest API documentation and include that in its response. This makes the workflow generation even more accurate and upto-date. Let me show you how to use this. When you're prompting Claude, you can say something like, "Create an NAN workflow that connects to the latest OpenAI API. Use web search to find the current API endpoints and authentication methods. Claude will search the web, find the latest information, and build a workflow that uses the most current API specifications. This is incredibly powerful because APIs change all the time. Authentication methods get updated, new endpoints get added, and with Claude's web search, you always get the latest information. And here's another tip that most people miss. Don't just describe what you want to Claude. Show it. Instead of saying, "Build me a workflow that processes emails. " take a screenshot of a similar workflow and say, "Build me something like this, but for email processing. " Visual examples work much better than text descriptions. Claude can see the structure, the node types, and the connections. It gives much more accurate results. Now, I know some of you are wondering about the technical details. How does Claude actually understand NAT's JSON format? How does it know which nodes to use? How does it handle connections and data mapping? The answer is in the training and knowledge you provide. When you

upload NAN documentation and examples, Claude learns the patterns. It understands that certain nodes work together. It knows how data flows from one node to another, and it can replicate these patterns in new workflows. It's like teaching someone to cook by showing them recipes. The more recipes they see, the better they get at creating new dishes. The same principle applies here. The more NA10 examples you show Claude, the better it gets at building new workflows. And here's something really cool. You can even train Claude on your specific use cases. If you're in e-commerce, upload e-commerce workflow examples. If you're in marketing, upload marketing automation templates. Claude will learn your specific patterns and build workflows that match your needs. This customization is what makes the difference between generic workflows and workflows that actually solve your problems. Now, let me address some common questions I get about this system. First, is this legal? Yes, absolutely. You're not stealing code or violating any terms of service. You're using publicly available workflow structures and adapting them for your own use. This is no different than learning from online tutorials or templates. Second, will this replace N810 builders? No way. This is a tool that makes builders more efficient. You still need to understand your business logic. configure integrations. You still need to test and optimize your workflows. This just eliminates the tedious part of dragging and dropping nodes and connecting them manually. Third, what if the generated code doesn't work? Remember, this works about 80% of the time perfectly. The other 20% might need small tweaks, but even a workflow that's 80% complete saves you hours of work. And as you get better at prompting Claude, your success rate will improve. Now, here's something I discovered that makes this even more powerful. You can combine this with other AI tools to create a complete automation pipeline. For example, you can use claw to generate the n workflow, then use another AI tool to create the API endpoints, then use a third tool to generate the documentation. The entire automation ecosystem can be built with AI assistance. I've started experimenting with this approach, and the results are incredible. What used to take weeks of development now takes days, sometimes hours. This is the future of automation building, and you're getting access to it right now. Look, here's what most people don't get about WhatsApp. Everyone's using it wrong. They think it's just for chatting with friends, but smart business owners know the truth. WhatsApp is the most powerful sales tool on the planet. And when you connect it to N with AI, magic happens. I'm about to show you the exact 20step process that turns any WhatsApp into an automated customer getting machine. These are the real steps, not theory. Follow along and you'll have your own AI sales assistant working 24/7. Step one, how to get client secret and client ID for N8 end. To get WhatsApp credentials, go to https business facebook. com or login with Facebook. Create a business page with email name etc. Select business portfolio. This is where everything starts and most people get lost right here. You need Facebook business to connect WhatsApp to N. Don't skip this step or nothing else will work. The business page is required for WhatsApp business API access. Use your real business information here because Facebook checks everything. If you use fake info, they'll shut you down fast. Step two, go to manage apps. This is where you'll create the connection between your Facebook business and N8N. The app section is your control center for all integrations. Step three, click add. Create a new app ID. Every integration needs its own app ID. Think of this as your unique key that lets n talk to WhatsApp through Facebook system. Step four, use cases. Other app type business. Create app. Submit. Business app type gives you access to WhatsApp business API. Consumer apps don't get this access, so make sure you pick business here. Once done, click WhatsApp. This adds WhatsApp functionality to your app. Without this step, your app can't send or receive WhatsApp messages. Step five, click setup and you'll see a screen like this. The setup screen is where you configure your WhatsApp integration. Take your time here because getting this wrong means starting over. Step six, click app settings. Basic. Create a new NAN workflow. Add WhatsApp on messages trigger. This is where you'll find the credentials you need for NAN. These credentials are like passwords that let NAN access your WhatsApp. The trigger listens for incoming WhatsApp messages. When someone messages you, this trigger wakes up your automation and starts the process. Step seven, copy app ID. Create new credential. Paste into client ID. The app ID tells N which Facebook app to connect to. Copy this exactly as it appears or the connection will fail. Step eight, copy app secret. Paste into client secret section. Hit save. Connection tested successfully. Save changes on WhatsApp. The app secret is like your password. Never share this with anyone. This proves to Facebook that you own the app you're trying to

connect to. Don't forget to save. Your changes won't take effect until you save them. This activates your WhatsApp integration. Step nine. Create new credential. Go back to Facebook WhatsApp drop-down. You need separate credentials for sending messages. The first credential was for receiving. This one is for sending messages back to customers. Step 10. Click API setup. Go with the test number. But add your own number in the to section. Generate an access token. Continue. Select all. Save. Got it. Copy paste. Access token. The test number lets you try everything before going live. Add your real number so you can test the messages yourself. This access token gives n permission to send messages. It expires, so you'll need to refresh it occasionally to keep your automation working. Step 11. Business account ID in a WhatsApp business account ID. This ID tells the system which WhatsApp business account to use. Every business has a unique ID that connects everything together. Step 12, save. Then it should say connection tested successfully. If you don't see this message, something went wrong. Go back and check all your credentials because one of them is wrong. Step 13. Settings here. Step 14. You can test it like this. Tasteing is crucial before going live. Send yourself a message to make sure everything works. Better to find problems now than when a real customer messages you. Step 15. Then send a message to the test number you have once done. Click the test workflow on nan. Use WhatsApp web or your phone to send a test message. The workflow should catch this message and process it. If nothing happens, check your trigger settings. Step 16. Set up your AI agent using this. The AI agent is what makes your bot smart. It understands customer messages and generates helpful responses automatically. This is where the magic happens. Step 17. You will get an error on the memory module, which is okay. Edit the memory module. Don't panic when you see this error. The memory module needs special configuration to work with WhatsApp. This error is normal and expected. Step 18 key click expression. Type this into the memory section. T WhatsApp trigger item. json contacts way g key icing. This expression tells the AI to remember conversations by WhatsApp ID. Each customer gets their own memory so conversations stay personal and relevant. You can change the memory as well. So you could say, "Remember my last 10 conversations," for example. Memory settings control how much your AI remembers. More memory means better conversations but costs more. Find the balance that works for your business. Once you've done the above, then test the workflow again. Send a message to your test number. Example below. You can use the WhatsApp web app. Testing after memory setup is crucial. The AI should now remember who you are and continue conversations naturally instead of treating each message as new. Step 19. Now, if you want the WhatsApp business cloud option to reply to you with the AI, use these settings. So, you drag the output JSON into the text body. This connects your AI responses back to WhatsApp. The AI generates the response and this step sends it back to your customer automatically. Step 20. Then, when you test it again, you'll see something like this. Your AI should now respond intelligently to test messages. If it's not working, check that all connections are properly linked between nodes. Once done, you can activate the workflow like this. Activation makes your bot live and ready to handle real customer messages. Make sure everything works perfectly before activating because customers will start messaging immediately. Now, here's what happens when you follow these exact 20 steps. Your WhatsApp becomes a customer getting machine. People message you and get instant intelligent responses. Your competitors are still manually typing replies while you're sleeping. But here's the real secret that nobody talks about. This isn't just about WhatsApp automation. This is about building a business that works without you. Every message that comes in gets handled by your AI. Every lead gets the perfect response. Every customer feels heard and valued while you're focused on growing your business and the cost. Practically nothing compared to hiring staff. We're talking about maybe $50 per month total for a system that could generate thousands in extra revenue. The ROI is absolutely insane. But here's what I need you to understand. Most people will watch this video and do nothing. They'll say, "That's cool. " And then go back to manually responding to messages. Don't be like most people. The businesses winning right now are the ones embracing automation. They're not working harder. They're working smarter. And this WhatsApp automation system is just the beginning. I'm going to be straight with you. Most people are getting ripped off when it comes to AI automation. They're paying 20, 60, even hundreds of dollars per month for tools like N. And here's the crazy part. You can get the exact same thing running on your computer for free. Not a trial, not a limited

version, the full thing. I just discovered this method and it's honestly mind-blowing. You can run unlimited workflows. You can use any AI model you want through Olma and everything stays private on your machine. No data going to third parties, no monthly bills, just pure automation power. Now, before I show you exactly how to do this, let me paint you a picture of what this really means. The starter plan for N costs €24 per month. That's only five active workflows. The pro plan, €60 per month for 15 workflows. If you do the math, that's €720 per year just for the pro plan. But what I'm about to show you gives you unlimited everything for $0. I've been testing this for weeks now, and it's rock solid. I've got multiple AI agents running. I've got workflows automating my entire business and my computer isn't even breaking a sweat. This is the real deal. So, here's what we're going to cover today. First, I'll show you the exact setup process step by step. Then, I'll walk you through creating your first local workflow. After that, I'll show you how to import existing templates. And finally, I'll reveal some advanced tricks that will blow your mind. But here's where it gets really interesting. Most people think you need to be a tech genius to pull this off. That's not true. The setup I'm about to show you uses something called Docker. And even if you've never heard of Docker before, don't worry. It's basically like installing an app. You copy and paste a few commands and boom, you're done. Now, let me show you exactly what this looks like in action. I've got n running locally on my machine right here. Look at this URL. It's localhost. That means it's running on my computer, not some server in the cloud. And when I open up the chat interface, I can talk to my AI agent that's powered by Olma running locally, too. Watch this. I'm going to ask it a question. Hey, are you working? And look at that response. The AI is responding instantly. No API calls to OpenAI. No credits being used up. This is all happening on my local machine for free. This is what I call the holy grail of AI automation. You get all the power of enterprise level tools without any of the cost. And here's the kicker. It's actually faster than the cloud versions because everything is running locally. Now, you might be wondering about the technical requirements. Do you need some monster computer to run this? Not at all. I'm running this on a Mac M3 Pro, and it handles everything perfectly, but this works on Windows, too. I'll show you both setups. The secret source here is something called the self-hosted AI starter kit from NAN. This is an official project from the NAN team that lets you run everything locally. It's all packaged up in Docker containers. So, the setup is incredibly simple. Here's what happens when you follow my method. Docker pulls down all the necessary components. It sets up N8N. It sets up Olmer for the AI models. It configures everything to work together. And in less than 5 minutes, you have a full automation platform running on your machine. But wait, there's more. And this is where most people get excited. You can use any AI model you want with Olma. Want to run the latest Llama model? Done. Want to try Deep Seek? No problem. What about Quen? Absolutely. The choice is yours and it costs nothing extra. Let me break down the exact process for you. First, you need to have Docker installed on your machine. Docker is free and it's like having a virtual computer inside your computer. Once you have Docker, the rest is just copy and paste. For Windows users, here's what you do. You open up your terminal or command prompt. Then you paste in these commands. First, get clone https github. com nio selfhosted AI starter kit. Git. This downloads the starter kit. then cd self-hosted AI starter kit to enter the folder. Finally, docker compose profile GPU Nvidia up to start everything up. For Mac users, it's almost identical. Same first two commands, but the last one is docker compose profile CPU up instead. That's because Mac handles the processing differently. Now, here's what's happening behind the scenes that most people don't understand. A Docker is creating isolated containers for each service. One container runs nan, another runs. They can talk to each other, but they're separate from your main system. This means if something goes wrong, it won't mess up your computer. The first time you run this, it might take a few minutes to download everything. Docker is pulling down all the necessary files. But once it's done, starting up is instant. And here's the cool part. You can see everything running in Docker Desktop. You can restart services. You can view logs. You have complete control when everything is up and running. You open your web browser and go to localhost fee 5678. That's where N810 is running. The first time you visit, it'll ask you to create an account. This is just a local account on your machine. No registration with NAN required. Now, let me show you something that will blow your mind. Remember those expensive NAN plans I mentioned? Here's what you actually get with the starter plan at €24 per month. you get five active workflows and 10,000 executions. With the pro plan at €60 per

month, you get 15 active workflows and 100,000 executions. But with this local setup, unlimited workflows, unlimited executions, the only limit is your computer's resources. And unless you're running hundreds of workflows simultaneously, you'll never hit that limit. I've been running this setup for weeks, and I currently have over 20 active workflows. That would cost me way more than €60 per month on the cloud version. But here I am paying nothing. Now, let's talk about the AI part because this is where it gets really exciting. Olma is like having chat GPT running on your computer, but instead of being limited to one model, you can run dozens of different models. And here's the crazy part, they're all free. Want to use the latest Llama 3. 3 model? Just download it through Olma. Want to try the new Deep Seek model that everyone's talking about? Download it. Want to experiment with different models for different tasks? Go for it. There are no API costs, no usage limits, no restrictions. To set up Alama, you go to alarm. com and download the app. It's available for Windows, Mac, and Linux. Once it's installed, you can download models with simple commands. For example, alarm llama 3. 3 downloads the Llama 3. 3 model. Alarm Deep Seek Coder gets you the Deep See coding model. The beautiful thing is that once you have these models downloaded, they work offline. Your internet could go down and your AI automations would keep running. Try doing that with Chat GPT or Claude. Now, here's where most people make a mistake. They think bigger models are always better. That's not true. For many automation tasks, smaller models work just as well and run much faster. A 7B parameter model might be perfect for processing emails or generating reports. You don't always need the biggest model. I've tested this with multiple workflows and the results are impressive. I have one workflow that processes customer support emails, another that generates social media posts, a third that analyzes website data, all running locally, all using different AI models, all completely free. But here's what really sets this apart from cloud solutions. Privacy. When you use chat GPT or clawed through APIs, your data goes to their servers. With this setup, everything stays on your machine. Your business data never leaves your computer. For companies handling sensitive information, this is huge. Let me show you how to import existing workflows because this is where the real magic happens. I have templates saved from my AI profit boardroom that I can import directly into this local setup. I just go to the import section, select the file, and boom, instant automation. These aren't basic workflows either. I'm talking about complex multi-step processes that would normally require expensive tools. customer onboarding sequences, lead qualification systems, content generation pipelines, all running locally for free. Here's something most people don't realize. The N8N ecosystem has thousands of pre-built workflows. You can find workflows for almost any business process you can imagine. And with this local setup, you can run all of them without worrying about costs or limits. I've imported workflows for social media automation, email marketing, data analysis, and even complex AI agent conversations. Each one works exactly the same as it would on the paid platform. The only difference is I'm not paying for it. I just discovered this insane method using N and Apify that lets you scrape thousands of highquality leads completely free. We're talking Google Maps, Instagram, Facebook, LinkedIn, you name it. I'm talking about pulling 50, 100, even 500 leads at a time with full contact details, phone numbers, email addresses, business names, addresses, everything you need to start making money. Then I'm going to show you how to plug this into an AI agent that writes personalized outreach emails that don't sound like spam. Julian Goldie reads every comment, so make sure you comment below with your questions. Check this out. I've got this web scraper running right now and with just a few clicks, I pulled 50 dentist leads from Google Maps. Look at this data. We've got business names, full addresses, phone numbers, email addresses, everything you need to start reaching out. All of this data automatically gets sent to a Google sheet, no manual work, no copy and paste. I ran this system this morning and it only cost me 41 cents to scrape 50 leads. That's less than a penny per lead. You get $5 of free credits every month that resets. So for most people, this is completely free. We're using N connected to Ampify. Appify has thousands of pre-built actors. That's what they call their scrapers for pretty much every platform you can think of. Google Maps, Instagram, Facebook, LinkedIn, Twitter, Tik Tok, Amazon, eBay. If you can browse it, they probably have a scraper for it. The workflow is simple. Send request to Appify to start scraping. Wait for Scraper to finish. Get the results. Send results to Google Sheets. That's it. Four steps that take 30 seconds to set up. Let me show you exactly how to build this from scratch. First, sign up for N and Appify. Both have free plans. In N. Create a new workflow and add an HTTP request node. Set it to post method. Go

to Appify and choose your scraper. Let's use Google Maps scraper. Set your search term. Let's say dentist. Set your location. New York. Set number of results 50. Copy the API endpoint from Appify and paste it into your N HTTP request URL. In the body section, select send body and choose JSON format. Copy the JSON parameters from Appify and paste them into N. Test this step. It should work perfectly. Now add another HTTP request node, but set this one to get method. Go back to Appy API endpoints and find get last data set items. Copy that URL and paste it into your second HTTP request. Add a weight node between these two requests. Set it to 30 seconds. This gives the scraper time to finish. Finally, add a Google Sheets node. Choose append to row and select your spreadsheet. You can map the columns automatically or manually. I recommend automatic to save time. Test the whole workflow and boom, you've got 50 lists of dentists with all their details flowing into your Google sheet. Want to scrape SEO agencies instead? Just change the search term to SEO agencies and update the GS on parameters. Want to scrape Instagram posts? Use the Instagram scraper actor and set your hashtags or usernames. Want to scrape LinkedIn profiles? Use the LinkedIn scraper and set your search criteria. The process is exactly the same, just different parameters. Let me show you what's available in Apey's actor store because this is where things get really powerful. They have Instagram scrapers that can pull posts by hashtag, location, or user account. Perfect for finding influencers or trending content in your niche. They have Facebook scrapers that can pull group members, page followers, or post comments. Great for finding engaged communities. They have LinkedIn scrapers that can pull company employees, job postings, or profile information. Perfect for B2B outreach. They have Twitter scrapers that can pull tweets by keyword, user, or hashtag. Great for finding people talking about your industry. They even have Amazon scrapers that can pull product reviews, seller information, or pricing data. Perfect for competitive research. Each scraper comes with detailed documentation showing you exactly what data you can extract and how to configure it. Let me give you some real examples of how you can use this system. If you're a web designer, you can scrape local businesses that don't have websites or have outdated websites. Then reach out offering your services. If you're a social media manager, you can scrape businesses with low social media engagement. Then offer to help them grow their following. If you're a copywriter, you can scrape businesses with poor website copy or outdated content. Then offer to rewrite their messaging. If you're in lead generation, you can scrape any type of business and sell the leads to agencies or service providers. The possibilities are endless once you start thinking about what data businesses need. Now, here's where it gets powerful. Once you have leads, you can add an AI agent to write personalized emails. Add a chat GPT node to your workflow after the Google Sheets step. In the system message, write you are an outreach assistant. Write personalized emails based on the business information provided. Then in the user message, include the scraped data like business name, address, and website. The AI will create custom emails that mention specific details about each business. Instead of generic, "Hi, I can help your business emails," you get, "Hi, business name. I noticed your Chinese restaurant in location, and I love what you're doing with specific detail. " Here's a more advanced prompt you can use. Write a short, friendly, personalized email to this business owner. Mention their business name, location, and one specific detail about their business. The email should offer website design services and include a soft call to action. Keep it under 100 words and make it sound conversational, not salesy. You can also create different prompts for different types of businesses. A prompt for restaurants might focus on online ordering and delivery. A prompt for law firms might focus on client acquisition and reputation management. The key is to make each email feel like you personally research that business. Once you have the basic system working, you can scale it in several ways. First, you can run multiple scrapers simultaneously. Set up workflows for different business types, different locations, or different platforms. Second, you can automate the scheduling. Set your workflows to run daily or weekly, constantly feeding new leads into your system. Third, you can add more sophistication to your AI outreach. Create follow-up sequences, AB test, different email templates, or add conditional logic based on business type. Fourth, you can integrate with other tools. Connect to your CRM, email marketing platform, or scheduling software to create a complete sales pipeline. Don't just set it and forget it. Monitor your workflows regularly to make sure they're working properly. Check your Appify usage to make sure you're not going over your free credits. Monitor your Google Sheets to make sure data is flowing incorrectly. Track your email open rates and response rates to optimize your outreach. Keep an eye on changes to the platforms you're scraping. APIs and websites change, so your workflows might need updates.

Always respect the terms of service of the platforms you're scraping. Most platforms allow reasonable scraping for legitimate business purposes, but don't abuse it. Don't scrape personal information that isn't publicly available. Stick to business contact information that's meant to be found. Always give people a way to opt out of your outreach. Include an unsubscribe link in your emails. Use the data responsibly. Don't sell it to third parties or use it for anything sketchy. Sometimes workflows fail. Here are the most common issues and how to fix them. If your scraper times out, increase the wait time between requests. If you're getting empty results, check that your search parameters are correct and the target platform hasn't changed. If Google Sheets isn't updating, make sure your authentication is working and you have the right permissions. If your AI responses are generic, improve your prompts with more specific instructions and examples. You can combine multiple scrapers for more complete data. Scrape Google Maps for basic info, then Instagram for social accounts, then LinkedIn for decision makers. You can use the data for competitive research. Scrape competitor reviews to find what customers complain about. Then position your services as a solution. You can create industry reports by scraping data from multiple sources and presenting insights to potential clients. You can build automated outreach sequences that send different messages based on how recipients respond. Don't scrape too much at once. Start with 50, 100 leads and make sure everything works before scaling up. Always respect rate limits and terms of service. Getting banned from a platform kills your workflow. Don't send spammy emails. Quality over quantity always wins in outreach. Always verify important information before using it. Scrape to data isn't always perfect. Don't ignore the human element. Even with AI, personal touches make a huge difference. This is revolutionary stuff. property management is about to get a massive upgrade and I'm going to walk you through every single step so you can build this exact system today. We actually built this for one of our clients and the results have been incredible. If you want us to build you a custom automation like this, make sure you book an AI automation session below. Julian Goldie reads every comment, so make sure you comment below and let me know what you think about this automation. Let me start by explaining what this system actually does. Because when you see how smart this thing is, you're going to want to build one for every business you know. The system has two main flows. The first flow handles messages from tenants. Whether they send a WhatsApp message or an email, this system catches it and processes it automatically. The second flow handles form submissions from contractors when they accept or reject jobs. But here's where it gets really smart. This isn't just about forwarding messages. This system actually thinks about what's happening. When a tenant sends a message, the system first figures out who they are. It looks them up in the database using their email or phone number. Then it finds their property. Then it checks if there's already an open job for that property. This is genius cuz if someone emails about a broken faucet today and then emails again tomorrow about the same forcet. You don't want to contact the contractor twice. The system is smart enough to add the new message to the existing job instead of creating a duplicate. But what if the tenant mentions multiple issues like first they talk about plumbing, then later they mention electrical problems. The system handles that too. It can add multiple contractors to the same job. So your plumber handles the pipe issue and your electrician handles the wiring issue. All tracked in one job. Now here's where the AI magic happens. The system uses Open AI to classify what type of job this is. Is it plumbing, electrical, HVAC? The AI reads the tenants message and figures out exactly what kind of contractor is needed. And this classification can handle multiple job types from a single message. So if someone says, "My sink is leaking and the lights are flickering," the AI knows to contact both a plumber and an electrician. But what happens if the AI can't figure out what the issue is? Maybe the tenant just says, "Hi," or sends something really unclear. The system has a built-in safety net for this. If the system can't classify the message, it waits 1 hour. Then it checks if the job is still unclassified. If it is, it sends an email to the property manager to review manually. This prevents the system from getting stuck on unclear messages. This is the kind of smart automation that actually works in the real world. It's not just blindly forwarding messages. It's making intelligent decisions about what to do next. Once the AI classifies the job type, the system looks up the right contractor. It finds the plumber's contact information from the database. Then it adds that contractor to the job in air table. Here's another smart feature. The system checks if the contractor has a phone number on file. If they do, it sends a WhatsApp message by default. If not, it sends an email. This ensures contractors get contacted through their preferred method. But the message isn't just, "Hey, there's a job. " The system includes a link to a form where the contractor can confirm or reject the job. and they can include their availability, timing, and pricing right in the form. This is where the

second flow comes in. When a contractor submits that form, the system processes their response automatically. If the contractor rejects the job, the system immediately emails the property manager. This way, they know to find another contractor right away instead of waiting around wondering what happened. But if the contractor accepts the job, that's when the real automation kicks in. The system updates the job in Air Table with the contractor's date, time, cost, and confirmation status. Then it adds the cost to a monthly tracking sheet so everything stays organized for billing. Then the system figures out how the tenant originally contacted them. Did they use WhatsApp or email? Because it makes sense to reply using the same platform the tenant used. Finally, the system emails the property manager to confirm that the booking is complete, so they know everything's handled without having to check manually. This level of automation is insane. Think about how much time this saves. A property manager normally has to read every tenant message, figure out what's wrong, look up contractors, send messages back and forth, track responses, update records, and follow up on everything. This system does all of that automatically, and it does it faster and more consistently than any human could. Let me walk you through how to build this exact system in N8N. We're going to start with the tenant message flow. First step is setting up triggers for both WhatsApp and Gmail. These will catch incoming messages from tenants automatically. Next, we need to normalize the data whether the message comes from WhatsApp or Gmail. We want the rest of the flow to see the same information format. So, we extract the contact info and message content in a standard way. Then, we look up the tenant in our database. We search by email or phone number to find their record. Once we have the tenant, we can find their property. Here's where we implement that smart duplicate detection. We check if there's already an open job for this property. If there is, we add the new message to that existing job. If not, we create a new job. This is crucial for preventing duplicate work. And it's the kind of logic that separates professional automation from amateur scripts. Now, we get to the AI classification part. We send the message content to open AI with a prompt that asks it to classify the job type. The AI can return one or multiple classifications or none if the message is unclear. If there's no classification, we start that 1 hour timer I mentioned. After an hour, we check if the job is still unclassified. If it is, we alert the property manager. But if we do have classifications, we loop through each one. For each job type, we look up the appropriate contractor from our database. Then we add the contractor to the job and send them a message with the form link. The message includes all the job details and ask them to confirm their availability and pricing. When the contractor submits the form, that triggers the seeant flow. This flow processes their response and updates everything automatically. If they reject the job, we alert the property manager immediately. If they accept, we update the job record, add the cost to our tracking sheet, and notify the tenant through their preferred contact method. Finally, we send a confirmation email to the property manager so they know the job is booked. This entire system runs 24/7 without any human intervention. It's like having a property management assistant that never sleeps, never makes mistakes, and never takes a day off. The time savings are massive. Instead of spending hours every day managing contractor communications, property managers can focus on growing their business. And the tenant experience is incredible. They send one message and get a professional response with contractor details and scheduling. No back and forth, no delays, no confusion. For contractors, it's just as good. They get clear job details and can respond on their own schedule. No phone tag, no miscommunication, no wasted time. This is the future of property management and you can build it today with N8N. The best part is that this system is completely scalable. Whether you manage 10 properties or 10,000, the automation handles the same way. You're not limited by human capacity anymore. And because everything is tracked in Air Table, you have complete visibility into your operations. You can see response times, contractor performance, job costs, and tenant satisfaction all in one place. This kind of data is gold for optimizing your business. You can identify your best contractors, track seasonal patterns, and make datadriven decisions about pricing and staffing. The ROI on this automation is incredible, even if you only manage a few properties. The time savings pay for themselves within weeks. And for larger operations, we're talking about thousands of dollars in savings every month. But here's what I find most exciting. This system demonstrates how AI can handle complex multi-step processes that require real decision-making. It's not just copying and pasting. is actually thinking about what to do next based on the situation. This is the direction all business automation is heading. Simple task automation is just the beginning. The real value comes from systems that can handle entire workflows intelligently. And the technology is getting better every day. The AI classification will become more accurate. The integrations will become more seamless. The possibilities are endless. If you're in property management, you need to build this system. If you're not, you should still build it to understand how these advanced automations work because this

approach applies to every industry. Customer service can be automated this way. Sales follow-up Project management can be automated this way. The principles are the same. The key is thinking about the entire workflow, not just individual tasks. What decisions need to be made? What information needs to be tracked? How should the system handle edge cases? When you design automation at this level, you're not just saving time. You're creating a competitive advantage that's hard to replicate. Your competitors are still handling everything manually. While they're busy with operational tasks, you're focused on growth and strategy. That's how you win in business. This real estate AI agent is just one example of what's possible. Every business has workflows that can be automated this way. The question is whether you're going to build them or let your competitors get there first. The tools are available right now. N8N makes it easy to build complex automations without coding. AI services like Open AI provide the intelligence. Database tools like Air Table handle the data management. Everything you need is ready to go. The only thing missing is someone to put it all together. That someone should be you. Start with one workflow. Build it, test it, and optimize it. Then move on to the next one. Before you know it, you'll have automated the majority of your business operations. This is how modern businesses operate. This is how you scale without burning out. stay competitive in an AIdriven world. The future belongs to businesses that embrace automation. Not the simple stuff, but the complex intelligent workflows that actually move the needle. Property management is just the beginning. Every industry is about to get disrupted by this level of automation. The question is whether you'll be leading that disruption or trying to catch up. But I hope this tutorial showed you what's possible, but more importantly, I hope it inspired you to start building. Because the best way to understand AI automation is to actually build it yourself. Free N Scraper Agent. Today, I'm going to show you how to scrape anything for free with N and get unlimited leads without coding. You're going to see exactly how I scraped 50 dentists in seconds and sent them straight to Google Sheets. I'll give you the full template for free, so you don't need to set anything up yourself. Plus, I'll show you how to turn these leads into personalized AI emails that actually get responses. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency, whilst he's helping clients get more leads and customers. I'm here to help you get the latest AI updates. So, what I'm about to show you is going to blow your mind. This is a complete game changer for anyone who needs leads. And I mean anyone. Whether you're running an agency, doing affiliate marketing, or just trying to build your business, you can scrape leads from Google Maps, Instagram, Facebook, Twitter, LinkedIn, pretty much anywhere on the internet. And here's the crazy part. You can do it all for free. No coding required. And then you can use AI to write personalized emails that actually convert. Let me show you exactly what I mean. Look at this example. We scraped 50 dentists and got all their details. Phone numbers, email addresses, locations, everything. And this took literally seconds to do. Not hours, not days, seconds. Then we blasted all that data straight into a Google sheet. Completely automated. You don't even need to map the fields yourself. N does it all for you. But here's where it gets really interesting. And once you have all this data, you can plug it into AI and create personalized outreach emails for each lead. So instead of sending the same boring template to everyone, you're sending custom messages that mention their business name, their location, their specific industry. And trust me, this works. When someone gets an email that's clearly written just for them, they actually read it. They actually respond. Now, you might be thinking, "Okay, Julian, this sounds expensive. How much is this going to cost me? " Here's the beautiful part. You get $5 of usage every month with Appify. Let me show you exactly how this works step by step. And stick around because I'm going to give you the complete template at the end so you can set this up in minutes, not hours. So, the magic happens with something called Ampify. Inside their app store, they have these things called Actors. Think of actors as pre-built agents for web scraping. They've already done all the hard work for you. You can scrape Google Maps for local businesses. You can scrape Instagram for viral posts. You can scrape Facebook groups for comments, LinkedIn profiles for prospects, Twitter for trending content. The list goes on and on. There's literally thousands of these scrapers already built and ready to use. And the best part is you just plug them into NAN and boom, you have a complete automation. Let me walk you through this from scratch so you can see exactly how easy it is. First, we start with an HTTP request in N8N. We set this to post. Then we go to Ampify and pick which scraper we want to use. For this example, let's use the Google Map scraper. Inside the Google Maps scraper, you tell it what to search for. Let's say we want to search Chinese restaurants in New York. You can choose how many results you want. You can choose whether you want to extract 50 different responses or 30 or whatever you need. Obviously, the more responses you use, the more API credits you're going to use in your free usage section. But the point here is that you can scrape almost anything you want using these free agents. Then you copy the API endpoint from Appify and paste it into your NAM workflow. You also copy the JSON configuration that tells the scraper exactly what to look for. Hit test and boom, your scraper is now running. It's going through Google Maps, finding every Chinese restaurant in New York, and grabbing all their details. But we're not done yet. We need to get

the results. So, we add another HTTP request, but this time we set it to get. We use a different endpoint called get last data set items that grabs the scraped data. And just like that, we have all the restaurant data flowing through our N workflow. names, addresses, phone numbers, websites, reviews, everything. Now, here's where it gets really powerful. We can take all this data and send it straight to a Google sheet. N has a built-in Google Sheets integration. You just tell it which sheet to use, and you get an option to map each column manually or map automatically. I always choose map automatically just to save time. Basically, you get all the headings inside the spreadsheet mapped to the Google map scraping search. So, now you have a Google sheet that's automatically filled with fresh leads every time you run your scraper. No copy pasting, no manual work. It's all automate. Now, if you want to learn more advanced AI automation techniques like this, you should check out my AI success lab. I've got over 14,000 members in there learning how to use AI to grow their businesses. Inside the lab, you get access to over a 100 different AI tutorials, templates, and use cases. I show you exactly how to set up automations like this one, plus way more advanced stuff. And the best part is it's completely free. Link is in the comments and description. Let me show you another example. Let's say you want to scrape for SEO agencies. Same process. We go back to the Google map scraper. We change the search to SEO agencies. We want to scrape in New York. 50 places to extract. We grab the endpoint. We get the run actor section. Copy that. Go back to n. Insert that into the first HTTP request. Post URL. Send JSON using JSON. And obviously, if you change any of the filters, then the JSON changes as well. Hit test step. Make sure it works. And then you can see all your previous runs. You can see what you're doing, how it worked, how much it cost. What I would typically do if you wanted to run this from scratch is add a wait section. Wait about 22 seconds before you go ahead. That will just allow the website scraper to scrape and complete scraping before you move on to the next step. Then we test that step and boom. Look at that. We've got all the internet marketing services and SEO agencies in the local area of New York using that process. Really simple and easy. Now, if you want to clean this up and keep it all neat and tidy, you can use the tidying up button. Then if you want to add a new module, let's say you want to send this to Google Sheets, we can add a Google Sheets section. Click on append row and sheet. Inside this section, we update that sheet. And you get an option to map each column manually or map automatically. I'm going to map automatically just to save time. Basically, we've got all the headings inside the spreadsheet mapped to the Google map scraping search. If we click on test step now, that's going to send all this information to the Google sheet. You can see it's all been appended to the spreadsheet. So, we've got all these SEO agencies inside New York from Google Maps listings, and we're living the dreams. Now, let's say you wanted to take this information and do some outreach. There's a couple of ways you could do that. You could take the information from the spreadsheet. You could go over to Jack GBT. You could say something like, "Create a short, funny, personalized email to this person from my SEO agency to sell white label link building to them. " And you can see it creates an interesting outreach email template based on the personalized details of that lead. For example, it might say something about Millionify. And Millionify is the name of the agency. Even more Millionify with easy link building. So, it says something like, "How many tabs you have open right now? " If the answer is too many, you don't need another thing on your plate. Here's how we can help you. Millionify plus our links equals your client's ranking, your margins laughing, and you're finally having time for lunch. So, that's one way to do it. The other way you can do it is you could take the information and get an AI agent. Then, we can define this information. We can take all the information from the Google sheet, the title, the address, the website. Then, we can add a chat GPT agent, open AI chat model. We'll select GPT. We can take the information from Google Sheets and send that to the AI agent. We can tell the AI agent with a system message, you are an outreach assistant helping me write emails for my leads based on the information in the prompt and we've put in the address, the website, the title. We could put the neighborhood as well. And this is basically taking the information from the Google sheet about the lead and then using that to automate the outreach. For example, we can say lead details here based on the information in the prompt. Write a nice personalized email. Then we can add a chat trigger. And you can see it's taking the information from the Google sheet in the email based on this. Pretty simple and easy. But let's try another example now. Let's say we want to build something with Twitter. We can put in an HTTP request. In fact, let's try Twitter just to mix it up. Let's say we're targeting content around AI agents. We want the latest content, maybe the top 10 tweets. We can do that. We'll grab the API key. Grab the endpoint. Copy that. Go back. Post URL. Inside the send body, we'll select using JSON. Then we'll grab the JSON. Copy this. Go back. Plug this in. Hit test step. That's running now. Then we can add a new HTTP request. We're going to link that in. We'll select it on get. Then we're going to take the URL from the API. Let's grab that. We'll get the endpoint. We'll scroll down. We'll get the last run data set items. Copy that. Then we're going to go back to N8N. Paste in the URL. You don't need to do anything else. And you can see the node is executed perfectly. And we're good to go on that. So it's really easy to set this up. Once you get the hang of it, once you get your head around it, it's super simple stuff. The cool thing about N particularly is that you can self-host this and then it's free as well. So at that point, you're not paying for N.

anything. All of this is pretty much free. Now, I want to address something important. You might be wondering if this is ethical or legal. The answer is yes. As long as you're scraping publicly available information, and you're using it responsibly. Google Maps listings are public, social media posts are public, company websites are public, you're not hacking into private databases or stealing confidential information. Now, if you're serious about scaling your business with AI, you need to check out my AI profit boardroom. It's the best place to scale your business, get more customers, and save hundreds of hours with AI automation. We currently have over 1,000 members who are all learning how to use AI to grow their businesses faster. Inside the boardroom, you get access to my most advanced automation workflows and templates. You get weekly coaching calls where I answer your questions directly. You get a community of like-minded entrepreneurs who are all on the same journey. Link is in the comments and description. You can also book a free SEO strategy session with my team. We'll look at your current situation and show you exactly how to get more leads and customers using SEO and AI automation. No pitch, no pressure, just pure value. Link is also in the comments and description. And remember, you can get the complete N scraper template for free inside my AI success lab. Plus, you'll get access to over a 100 other AI tutorials and use cases. With 14,000 members, you'll be joining a massive community of people who are all learning how to use AI to grow their businesses. You see how I show a checklist of 100 different tutorials that are given away as freebies every day inside the community. You can get all the video notes from there and all the other stuff that you get along with all the trainings in the AI success lab. Julian Goldie reads every comment, so make sure you comment below and let me know what you're going to scrape first. I'd love to hear about your results. And if you found this video helpful, smash that like button and subscribe for more AI automation tutorials. I'm dropping new content every week that shows you exactly how to use AI to grow your business faster. Until next time, keep automating and keep growing.