# 1 Automation Built On 3 Different Platforms (n8n vs Make vs Zapier)

## Метаданные

- **Канал:** Nick Saraev
- **YouTube:** https://www.youtube.com/watch?v=MjIngZJHjBc
- **Дата:** 03.05.2025
- **Длительность:** 34:20
- **Просмотры:** 24,817

## Описание

Join Maker School & get automation customer #1 + all my templates ⤵️
https://www.skool.com/makerschool/about?ref=e525fc95e7c346999dcec8e0e870e55d

Want to work with my team, automate your business, & scale? ⤵️
https://cal.com/team/leftclick/discovery?source=youtube

Watch me build my $300K/mo business live with daily videos + strategy ⤵️
https://www.youtube.com/@nicksaraevdaily

All Make.com & N8N templates mentioned in the clip ⤵️
https://leftclicker.gumroad.com/l/xovbb

Summary ⤵️
In this video we build a youtube transcript automation using zapier, make and n8n, comparing the differences between the platforms in a practical way.

My software, tools, & deals (some give me kickbacks—thank you!)
🚀 Instantly: https://link.nicksaraev.com/instantly-short
📧 Anymailfinder: https://link.nicksaraev.com/amf-short
🤖 Apify: https://console.apify.com/sign-up (30% off with code NICK30)
🧑🏽💻 n8n: https://n8n.partnerlinks.io/h372ujv8cw80
📈 Rize: https://link.nicksaraev.com/rize-short (25% off with promo code NICK)

Follow me on other platforms 😈
📸 Instagram: https://www.instagram.com/nick_saraev
🕊️ Twitter/X: https://twitter.com/nicksaraev
🤙 Blog: https://nicksaraev.com

Why watch?
If this is your first view—hi, I’m Nick! TLDR: I spent six years building automated businesses with Make.com (most notably 1SecondCopy, a content company that hit 7 figures). Today a lot of people talk about automation, but I’ve noticed that very few have practical, real world success making money with it. So this channel is me chiming in and showing you what *real* systems that make *real* revenue look like.

Hopefully I can help you improve your business, and in doing so, the rest of your life 🙏

Like, subscribe, and leave me a comment if you have a specific request! Thanks.

Chapters
00:00 Introduction
00:34 Previews
01:06 Overview
02:12 Zapier
14:50 Make
24:10 n8n
33:14 Outro

## Содержание

### [0:00](https://www.youtube.com/watch?v=MjIngZJHjBc) Introduction

Cool exercise for us today. We're going to build the exact same system in three different platforms. That's Zapier, Make. com, and NAN. And I'm doing this because a lot of people have been wondering what are the differences between these platforms. Well, rather than just give you guys a bunch of my own and other people's opinions, might as well actually build a real system so you guys could see what the build process looks like. The system we're going to be building in particular is a YouTube repurposing system where I provide the URL to a YouTube video and then it just generates a bunch of social media posts for other platforms. And I'm going to be drawing out the process here because obviously it's going to depend on whether we're using Zapier, Maker, or NADN. But this is going to be the road map that we return to as I go through

### [0:34](https://www.youtube.com/watch?v=MjIngZJHjBc&t=34s) Previews

the build. Okay. And just jumping ahead a little bit, this is what the finished product looks like in Zapier. As you can see, we're doing it all in five steps linearly top to bottom. This is what the finished product looks like in make. com. As you can see, we are doing this in six steps, left to right. And this is what the finished product looks like in naden. And as you can see, we're doing this in five steps, left to right. The logic behind each of these does differ a fair bit and that's one of the reasons why I wanted to do this video just to show you guys how data structures change between Zapier naden and make. com which platform's simplest, which one's the easiest to get up and running with and which ones you might have to look for some gotchas in order to make work. So

### [1:06](https://www.youtube.com/watch?v=MjIngZJHjBc&t=66s) Overview

the way that the system is going to work is I'm going to keep it really simple and easy for everybody regardless of skill level by making the trigger an email received trigger. So when an email is received in an account, that's what's going to kick off this flow. What is the email going to include? It's going to include the URL of the video that we want to pump through the system. After that, what we're going to do is we're going to scrape the transcript of the YouTube video. The transcript, for those of you guys that don't know, is just the text version of it. After that, we're going to feed the transcript to AI. And then with AI, we're going to have it generate social media posts that are of similar form. And then finally, the very last thing we're going to do, and this is going to be optional. I'm not going to include it in the video, but I'll show you guys how to do it if you wanted to. We're going to post it on various platforms. Okay? What platforms might you want to post something like this on? I don't know. Instagram, YouTube, community posts. My guess probably do LinkedIn. You do X, although the X API has changed quite a bit recently. Tik Tok, right? A lot of options here. Okay, so without further ado, let's get into actually building this live. Okay, so let's actually build this system out in multiple platforms. In order to do that, what I'm going to be

### [2:12](https://www.youtube.com/watch?v=MjIngZJHjBc&t=132s) Zapier

doing is we're going to start in Zap here. I need to provide some sort of trigger first and foremost where when I receive an email into a mailbox, I start off the flow. Now, the way that Zapier does it in their canvas is they have this little trigger box. So, you can select the event that starts your Zap. Zap is just the term that they use for a workflow. So, what we're going to be using as our trigger is understandably email. Okay. So, the trigger event is going to be when we receive a new inbound email. I'm then going to continue. Now, it's going to give us a little box with an email address. And we can actually make this email address the end all beall deacto email of our flow in make. com and at an end. You'll see that I'm going to take a slightly different approach here. But we're just going to call this YouTube transcript system. Very long and unwieldy email. Okay. Once I'm done with that, I have it right over here. Then I'm going to click continue. And now they are basically listening to the trigger. So in order for me to test this, what I need to do is I need to send an example of it. Okay. So what I'm going to do is I'm going to go to my Gmail. I'm going to click compose. Then I'm going to add this in. And then what I'm going to do is I'm just going to insert the URL of the video that I want just directly in the subject line. Okay. So the URL of the video I want is I'm just going to use one of my own videos here which is these eight AI trends will change business forever. Get in early. I thought this would be a good video to do some repurposing on because obviously we have eight AI trends. So we could feed them into AI and you know generate eight posts. Now I see some additional metadata here like you know uh the timestamp and the channel about and so I'm just going to get rid of that just to keep my life super simple. And then I'm going to send this. Okay. Okay, so now that we've sent this, if I go back over to Zapier, then if I test the trigger, takes a second for Gmail to actually send the email. So, give me a second. Okay, great. So, it looks like we found the email. Email A, original record pulled on May 1st. And what's really cool is we have a bunch of fields here. We have raw from, raw message ID, raw, all this stuff, right? What we're really looking for is we're just looking for the subject line. Nothing different. Okay, so checking out the subject line here. And there we go. We have it. This is the YouTube URL link. Cool. So, I'm going to continue with the selected record. The next thing we have to do is we have to hook this up to an action. Now the action is I want to scrape the transcript of this YouTube video. There are variety of ways to do the scraping. The simplest way to scrape something like this is using a thirdparty platform called Appify. The cool thing is Appify integrates natively with Zapier and Make. Doesn't integrate natively with NAN, but that's part of the cool fun part. I'm going to show you guys like how to actually build out the HTTP requests and stuff like that. Basically, if we just type in YouTube transcript over here, you'll see there's actually a variety of scrapers that people have built for various YouTube transcript scraping purposes. This person Topaz_Sharing, obvious Naruto fan, is allowing us to scrape using theirs. Pinto Studio is another provider, Caramelo. Caramelo Insight API Labs. This is basically a marketplace. And what I've done is I found the cheapest and most currently widely available transcript extractor where you only pay per usage from the lovely Abla Kim Heaya, whose name I am hopefully not butchering. Okay, so what we want to do is we basically want to trigger this the second that we receive an email. And in Zapier, the way to do that is we need to click and then write ampify, select it from the top here, and then we need to choose our event. The event that I'm going to be doing is run an actor. And now what we need to do is we need to connect. So I haven't actually done this connection before. What I have to do is I have to go find my API token from Ampify. So I'm just going to go back to my account and then I'm going to go to settings. Then over here in API integrations, I'm actually going to create one and I'm just going to call this Zapier. I'm then going to copy it, then go back to my Zap and then paste this in and then click continue to Appy. This will now run my connection and it's now actually connected over to my account, which is pretty cool. Now, once I click continue, I'm going to get choose to the specific actor that I wanted to. So, let's see. This one here was called the YouTube transcript scraper, right? Looks like this one from DZ Omar. Give that a click. Then, I have the choice to run it synchronously. Now, run synchronously just means, do you want it to finish? Do you want us to wait however long it takes and then when it finishes return all the data and that's what we want. So I'm going to click run synchronously. Yes. Now we have a place where we can actually insert the videos that we want to scrape. Okay. So I'm just going to exit this out. And what I'm going to do is I'm going to press slash. That's how Zapier does field mapping. And then what I want to feed in is the subject. Right? That's logical cuz the subject is basically the URL. Then we have a bunch of additional settings we can use here. I'm not going to worry about it. The only setting I'm going to click on is transcript only mode. Okay. So now it's going to allow us to test the step. You guys can't see the test step. It's in the bottom right hand corner. But bear with me here. When I click that little test button, this is now actually sending a request over to Appify, which is that third party service. If I click on runs now, you'll see that something just occurred a moment ago. If I go back to my Zap, you'll see that we now have this message that says succeeded. And right over down here, we have data set items transcript text where it actually has the full transcript of the YouTube video. The way that we do business today is changing pretty fundamentally. Blah blah. Cool. All right. So, looks like we're making some pretty steady progress. After this, what do we need to do? We obviously need to feed it into AI and have it turn this into some social media posts. So, what am I going to do? I'm going to go down to Open AI. The event I'm going to want is the conversation API. So, give that a click. Then, I'm going to need to connect this to OpenAI as well. There are two settings here that you're going to need. You're going to need the Open AI API key, and then you're also going to need the organization ID if you have multiple organizations. So, I'm just going to go over to API keys. I'll go create new secret key. We'll call this Zapier. And then for project, I'll just go default project. Then I'll create the secret key. Copy this over and then paste this directly in here. Okay, it should be sufficient. We'll now continue. And this will now hook this up to our Open AI API account. Okay, from there we can now enter in a message, which is where we basically tell the AI model what we wanted to do. We can select the specific model. And then we have a couple of other fields here. Memory key, image. I'm just going to leave all of this stuff. I'm going to click show advanced options below. When you click true, you'll see that there are a variety of other settings here. Then what I want is assistant instructions. So I'm just going to say you are a helpful intelligent content creation assistant. Okay. Under max tokens, I'm going to write 20048. And under temperature, I'm just going to write 0. 7. As I scroll up here, the model that I want is not actually GPT-40- mini. That's an okay model, but I want one that's slightly better. GPT-4. 1. Currently significantly higher quality than GPT40. And now I'm going to give it a user message. I'm actually just going to write this live in front of you and then we're going to repurpose it for Make. com and NAD as well. So the specific user message I'm going to write is you receive transcripts of high quality YouTube videos and convert them into posts for let's just say we're doing this for LinkedIn. Your posts are written in a spartan laconic tone of voice and reflect closely theformational content of the source transcript. Okay. Rules. Identify highquality interesting parts of the transcript to repurpose. Generate at least five posts using the source transcript. Separate your posts using the vertical pipe character that is blank and then use emojis sparingly. Write as if you're Microsoft's content marketing at the Microsoft. Okay, cool. That looks good. Now I'm going to go to the run actor and apply data from before. The reason I'm scrolling down here is because I'm just looking for the transcript. Okay, so this is going to be data set items transcript text. Cool. So there you go. There's the transcript and we should be good to actually run this. Give this a test. So I'm going to click continue and now I'm going to test the step. It's now sending all this data to the model. And what we're looking for is we're looking for the data out field. That's how you do things in Zapier just so that everybody could see what's going on. Let me just make this a little bit bigger. Okay. So this is the data out. This is the big long transcript we fed in. This is the data out. Number one, AI shifting business fundamentals. Distribution now matters more than product. If you're building something, prioritize finding an audience before perfecting your offer. 2 3 4 5 6 7 and then eight. So what it did is it effectively found all eight of the fields and now we have you know eight different posts essentially. I don't like how there are numbers here. So I'm just going to go in and I'm just going to edit it and I'll say you know don't do any numbers. So let me just go back to configure. No numbers or any sort of ordination scheme. Just put the posts just output the posts in between vertical pipes. Okay, let's give that one more try. Okay, looking at the posts. Awesome. We received a very similar output to last time, except this time it doesn't include the numbers. That's awesome. Okay, what we're going to do next is we are going to format this. So, I'm going to click formatter. What we're going to want to do is work with text to basically split these based off the presence of that pipe. So, I'm just going to go action event. We need to do is we should have a split string. split right over here. The input is going to be that big string that we just output from chatbt a moment ago. So it would be reply. Okay. So I'm going to click reply. Then the separator is going to be this vertical pipe. Okay. And then the segment that I want is all as line items. I'm going to click continue. And then I'm just going to give this a test. And now the output looks pretty good. We do have a bunch of these spaces which I don't really like. Kind of looks ugly. H. We should probably just tell the AI model not to add any spaces or anything between them too. just for future reference. So, I'm gonna go back here. No new lines or anything either. Output them all in one big chunk of text separated by blank. Cool. This should be good. I'm just going to retest this now. Make sure that everything is good to go. You don't actually need to do this with other no code platforms, but Zapier is a little bit finicky. I want to say some things are a lot simpler to do in Zapier. Other things are way harder. Okay, that looks a lot better. So, now we have all that reply text. So now when we go to the formatter and then configure it and then split it should just get everything on basically one line. Nice. This looks pretty good to me. Uh there are still some issues with like a space at the beginning but pretty solid. And then from here the question is what do we do with it? Well what's really cool about Zap year is it includes a bunch of built-ins that allow you to post to social media platforms pretty easily. So for instance if I typed um Facebook you'll see that there's a Facebook pages, there's Facebook lead ads for business. There are a variety of these built-in modules and a lot of these include the ability to create page posts, page videos, and so on and so forth. Facebook isn't the only thing you can do. You could also do, let's say, Instagram. LinkedIn, right? The variety of social media platforms, which is really cool. And that's one of the reasons why Zapier is so good for beginners because Zapier actually has a lot of these things just like naturally built in. I'm not affiliated with them whatsoever, so I'm saying this completely unbiased. One of the trickiest parts for beginners when they get up and running with a automation platforms is like dealing with connections with social media apps because typically their systems are just very complicated like you know you have to do ooth and graph APIs and there's just so much more that goes into this. What's really cool about Zapier is it just allows you to focus on like doing the thing not necessarily worrying about how to do the thing if that makes sense. So for here if you wanted to you know at the end of this actually publish to LinkedIn this is what you would do. you would add let's say a LinkedIn module. And now in my case, what I want to do is I actually just want to dump this to a Google sheet. So what I'm going to do is I'm just going to create or add a row. Okay, be add with line item support. And then I'm going to connect my own account. So we'll just go connect account. Then what we need is we need a specific spreadsheet. So I'm going to have to go and create a spreadsheet. Now why don't I just make this the same spreadsheet for everything. We'll go one automation in three platforms. And then over here I'll write zap year. Here we'll go make. Then here we'll go nad. Then we're just going to write post. And we're just going to keep it really nice and simple. There'll be post here for nad and make as well. Now that we have this, we can actually go and we can select this. So I'm going to click continue. The drive is just going to be my Google Drive. And the spreadsheet is going to be this specific one. One automation and three platforms. The worksheet I'm going to be dumping into with the Zapier flow is obviously Zapier. And the row is just going to be post. I'm going to go slash. And then what I want is I want this text formatter. Okay. So output. Now this is going to have line item support. So it should just dump in all of our rows. And we'll click continue. Then we're going to test the step. When I move back to Zapier, you can see we've now added all of these posts. Okay. So as changing business fundamentals distribution beats product, you'll see that there are what looks like there's a space between them because I guess the AI put a space between all of them except for the first. That kind of makes sense. You could fix this just by formatting the text a little bit more. So you could um theoretically add a formatter here. Uh you could go down to text and then you could do what's called trim whites space here. And this would just automatically remove the white space at the beginning and at the ends of all of the nodes. Okay, cool. So this has now done everything we wanted it to do. Let's move on to the next system which is make. com. Now, make. com is

### [14:50](https://www.youtube.com/watch?v=MjIngZJHjBc&t=890s) Make

similar to Zapier in so far that they have a lot of built-in nodes and built-in modules, but I personally find it to be like the right mix between simplicity and then sophistication, if that makes sense, cuz you have a lot more control and a lot more granular oversight into what's actually going on under the hood. They also visualize like looping really well. With Zapier, as you saw, we generated eight items and then we added all eight items to one Google sheet, but it was difficult for us to visualize what was going on under the hood. I think we could probably all follow along because it was pretty simple top to bottom, but make us sort of like left to right and I think it's just a lot easier of a thing to get used to. Okay, so what I'm going to do is I'm going to start off with a web hook and then I'm going to go down to custom mail hook. Then I'll click add and I'll say new YouTube URL received. What this is going to do is give me an email address just like we had before where if I send an email to it, it will trigger our flow. Okay, then I'm going to click run once. This is now going to wait for me to do the thing. Then I'm going to go here and then let me just go down to that YouTube video again. Just uh copy my ugly mug. Paste in my ugly mug one more time. And I think I forgot an S. There we go. Click send. Now, if we go back to make. com right over here, we just receive the data. And the way that we receive the data is using this little modal where we can get the subject line immediately. And it's also really easy to map, which I like. Okay. All right. So, same thing after this. this we have an ampify built-in module. So we can just go apply then we can go run an actor. Now I actually already have a connection. Um the connection logic is very similar on all sides. All you do is you click add connection type appy or I guess the appy recommended one. Click save and then you'll actually log in. So this is now like an ooth 2 sort of deal which is pretty nice. And then you can like hook it up allow it full API access to your account and so on and so forth. I personally think the oath is a little bit easier. Then what do we want? We want YouTube transcripts. So it's that one by DZ Omar. I'm going to click run synchronously and then we just have some input JSON build timeout and memory. We have some additional fields that we didn't have in Zapier. So, as I mentioned, Zapier is a little bit more simple, whereas make. com is sort of like the middle ground of complexity. I find what this input JSON is if you go back to the Appify actor and you click on JSON, you actually get like this object here and this is in JavaScript object notation, which can be a little bit tricky for people to understand if it's the first time they're looking at it. Basically, this is just a simple and easy way that computers can communicate with each other in a little bit more rule-based of a manner. So, there's this item here, the URL, that I need to replace with the actual variable. And I'm going to do that just by selecting subject. Okay. Now, what this is doing is the same thing that we were doing before. It's just we're actually sending and dealing with all of the JSO and the JSON. Okay. Now, one other thing that make. com sort of does which can be a little bit frustrating is it doesn't store all of the data. So, we don't actually have access to this data. If we were to just run it, it would just wait for us to resend the email. I personally don't really like this. This can take a fair amount of time because then you have to go into the habit of like resending emails every single time you want to test a flow. With Zapier, as you saw, you just made a couple of adjustments and then retested the specific step you want. NN does this, too. And I think it's just a matter of time before make gets their act together and does this on their end. But what you can do instead is you can just rightclick the module, press run this module only, and then manually fill in the variables. Okay, so that's what I did back there. Now, in order to actually get the data in make, you need to go one additional step. And so, it's not enough just to run the actor. What you also need to do is you need to get the data set items. And the data set items are always going to be hidden in this field called default data set ID right over here. So, I'm going to do is I'm going to copy this now. And I'm actually just going to run this step here manually just to show you guys what the data really looks like under the hood. If I rightclick this, and again press run this module only, you'll see there's actually no h why is there no data here? Okay. So, I'm just going to run this whole thing now. And then what I'm going to do is I'll go back to my email address. And then I'm just going to go and I'm going to paste it directly in again. So, as I mentioned, this is a little bit more annoying than doing it automatically like Zapier lets you do cuz you can cach the previous response, but all automation platforms are made a little bit different and that's the purpose of this video. Okay, we've now received the mail hook. That's pretty cool. Now, what we're doing is actually running that actor. So, this is the YouTube URL as we see back here. And now we just finish the execution. All right. So, it's pretty neat. I always like to double check this inside of Amplify. So if I go back to the specific run and actually click this button and you'll see that I've actually received all the data inside of Apify. The question is how do you get that inside of make. com? Well, unfortunately make doesn't do all of it in one step. So you actually need to both run the actor and then after get the data set ID. So that's kind of annoying. I'm not going to lie. But what we're going to do is we're just going to trigger this manually so that it's a little bit easier for us. Just go down to the default data set ID and I'll show you guys what it looks like when you trigger something manually. We're just going to paste the data set ID in there. Click save. Right click, run this module only. Because we're not using any variables in this, we can just run it and then we get access to the full transcript text right here. Okay. So, exact same thing as before. All right. Now, in terms of adding an open AI node, the way that you do this is you go over here to OpenAI. Then what we want is we want the create a completion. And you'll see that there are variety of more settings for make. com than there are for Zapier. And that's just because again, it's a little bit more complicated of a platform. And I mean this in the best way possible because the added complexity just gives you more room to play around with and do cool things with. I'll go to developer system. I'll say you are a helpful intelligent content writing assistant. I think that's what I said for the last one. And then I'm just going to copy over the prompt that I got from my Zapier node. Let's go to configure. And here is the text that I'm going to be feeding in. I go down to user and then text content. Paste all this in. Okay. So you receive transcripts of high quality YouTube videos and convert them to post for LinkedIn. and your post are written in blah blah blah here's how you do it now the thing is make. com actually allows you to control the output that the AI is generating for you and so you actually get a lot more granular than we were doing before so that's what I'm going to do here instead of doing all this vertical pipe stuff okay what I'm going to say is output your response in JSON in this JSON format we're going to do is we're going to say posts then we're going to make the post an array I'll Okay, first post, second post, and then I'll go etc. And the model will understand that this is what we want to get. Okay, and I'm also going to say no new lines. Now I'm going to go show advanced settings, response format, JavaScript object notation, parse JSON response. Then I'm going to go down here and add one final message that says user. Inside of this user message, I'm actually going to be putting in the whole transcript. Again, this is just a lot more freedom, I would say, but the cost of the freedom is obviously you have to know a little bit more about what you're doing. But this to me is the perfect blend. Okay, so now that we've set all this stuff up, let's actually click run once and let's wait for an additional ping. What I'm going to do is I'm going to copy this over. I'm going to Oops, that's not what I wanted to do. Paste in the subject line again. Paste in the two again. And then click send. What this is going to do now is this is going to actually grab our data over here. That's then going to feed that into this actor. Once this actor is done, we're then going to get the data from this module. And after that's then we're actually going to feed that right into AI. So, we just finished with that. Now, we're feeding this into AI. And let's see what the output is. You can see it's taking a little bit of time. That's cuz it has a lot of work it needs to do. If I click on the result object, you'll see it's buried inside of a post array because we've actually built a format inside of make. And as you can see, we have all of these posts right over here. Very cool. I like this. Okay. And so what do we do now? Well, what we do now is we post this to that Google sheet we had. So in order to do that here, all you do is go sheets. And then what you want to do is you want to add a row. You need to select the right account. If you haven't done this before, just click add. And then it's a very simple and easy sign in to Google. And then from here, I'm just going to select the specific spreadsheet that I had before. One automation in three platforms. Should have said on three platforms. Then I'm going to select the make sheet instead of the zap year sheet. And under post I need to select my post. The thing is since make gives you a lot more control over the data format in order to iterate through this array of posts I need to go and do one more step and that's I need to add what's called an iterator in between the two. Now anything that's buried in an array like this aka anything that has these square brackets is ripe for iterating. So what I'm going to do here, so I'm going to go into this green flow control section, click iterator, and then drag it in as follows. Now under iterator, what this is going to do is it's going to get my array of posts. Then over here, what we're going to do is we're actually going to cycle through the iterator output right over here, which is going to be I think just value. And I don't believe we have access to this right now. Like we need to go 13 value. Okay. All right. So, now that we're done with this, going to actually give this a go. I'm going to run this one final time. Again, we're going to copy over the specific YouTube URL and go straight to the specific recipient. Going to paste that. Okay. And if we go back to our sheet, you see that we are iterating over the array. And then if I go to make, we've now pumped in all the posts just like we did before. Just this time we didn't have to deal with any weird text formatting issues because all of the stuff was pretty native. Okay. All right. Now, let's move on to the final approach which is in N8N. N8N obviously

### [24:10](https://www.youtube.com/watch?v=MjIngZJHjBc&t=1450s) n8n

gaining a ton of popularity recently is one of the best ways to get up and running with no code. I will say the reason why I demonstrated this stuff starting with Zap here, moving to make. com and then finally finishing with NAND is I consider Anaden to be the most technically complicated of the platforms just cuz they show you a lot more of the guts of automation. They show you like JavaScript functions and so on and so forth. But we're still going to make it pretty simple and easy for you. Okay, first thing we have to do is we have to add a Gmail onssage receive node. What I have to do is I have to add a credential. This credential is going to be an OOTH credential. I'm going to click sign in with Google and then I'm going to sign into the specific account that I want to be watching for emails on behalf of. Okay, so we actually have to sign into our own account here and we have to set up our own account as the watcher. So, what I'm going to be doing is adding a simple filter and the search is subject and then YouTube. This is a specific filter that will only get new emails that have come in that have the subject line containing the term YouTube since obviously we're going to be sending YouTube posts. Keeps it pretty simple and straightforward for us. What I'm going to be doing now is I'm just going to email my business email with a YouTube link like this and click send. And then I'm going to wait for Gmail to actually send it. That's what happens when the undo tag leaves. Then what I'm going to do here is I'm just going to click test workflow. Now when you test a workflow in NAD, you end up with an output that looks kind of like this. Okay. And it looks like we got it. This is the subject line right over here. One of the really cool things about NAD, which a lot of people don't understand initially, is it allows you to pin outputs. So just like Zapier earlier allowed us to pin the output without really telling us what the whole pin thing looked like. Naden allows you to do so just by clicking that little pin tab. But anyway, just like in Zapier, you had things with key names and then values and in make when you had key names and values and it then represents that data the same key name and then value. And what we're going to be doing is we're going to be pulling from the key name subject uh and then getting that YouTube link. Now, if you think about it, go back to our road map here. The next thing we have to do is we have to scrape the transcript. However, if I had search for the term ampify in here, you'll notice that there is nothing to be had. This is unfortunately one of the biggest issues that beginners have with naden because they don't have anywhere near as many built-ins as zap year and make. com. It can be really tricky to figure out exactly how to construct a flow. And usually what you have to do is you have to turn to their API. Now API stands for application programming interface. And this is basically the under the hood code version of those nodes and modules that I was talking about before. And it turns out on amplify the one that we are specifically looking for is this one here. Run actors synchronously with input and get data set items. Now, you're going to probably be pretty intimidated by this cuz I mean, we're seeing a bunch of code. It says post, there's curl. What the heck does all this stuff mean? The good part about NAN is it makes large chunks of this pretty easy for you. Although, I find you still need to have some technical background in order to really, you know, understand it. Basically, what this is this is the URL that we're going to be making the request to. Anytime you have a colon, usually it's a place where you're supposed to put the ID of something. Now, in our case, it's going to be the ID of the Appify actor. What I'm going to do here, okay, is I'm actually just going to go to this curl body. I'm going click copy and I'm going to go to NAD. I'll go HTTP request, import curl, and then paste this curl in. What this is going to do is it's actually going to map that curl request into NAD. So, it's going to do more than half of the work for us. But our work here is not yet done. We still have some things we need to do. Notably, we have to replace the actor ID. Now, if you didn't already know what the actor ID is, it can be pretty tricky to kind of figure this out. But usually in any service that you use, the ID is always the thing in the URL that corresponds to that particular thing that you're doing. So, I'm looking for specifically actors. Notice how the URL says SL actor/ something. Then it says slashinput after. Well, if there's a custom input, it'll probably look like this. This probably means the actor. So, this is honestly the same way that I figure stuff out. I just try to paste in an actor and I give it a go. from here. Now, we have a couple of additional things we have to do. Under authorization, you have to enter the API key yourself. Again, you're not going to know how to do this unless you've actually run through maybe a couple of API connections. So, don't sweat too much if this seems like magic to you. What I'm going to do is I'll just write NAD. I'm going to create. Now, I'm going to go down to NAN. I'll copy this over. Go back here and then paste my API token. And then down below, what we have is our body content type, JSON. Specify body using fields below. What I'm going to do is I'm just going to specify using JSON. This is the same thing as that um input JSON that we saw in make. So I'm going to go back to the actor. I'll go JSON. I'm just going to copy this whole thing in. Paste it in here. Okay. So now we have the YouTube URL. As you can see, we're currently hard coding this. We don't actually want to hardcode this. What I'm going to do is click on this expression tab. Open this up in this window. This is now going to allow me to import variables. Then on the lefth hand side here, I'm going to go and I'm going to grab the specific subject variable here, kind of like make. And on the right hand side, you can actually see what the filledin value is. Left hand side is with the variable JSON. Right hand side is the variable actually mapped. So this is the value of the key name subject. That makes sense. Don't sweat the JSON or the dollar sign or these two curly brackets. These are just um NAD's built-in formatting. Okay. So this is what we're going to be feeding into API essentially, which is kind of cool. There's some additional settings here. Follow redirects, max redirects, so on and so forth. I'm not going to worry too much about that. What I'm going to do next is just click test step and I'm going to see what happens. Now, what I like to do is if I fire something off in a no code tool, I always like to go back to NADN or Ampify, sorry, and then see that it's actually running. And in this case, we actually did run it. Actor succeeded with one result in the data set. If we go back to NAN, you'll see the node just executed successfully. We actually have access to that full transcript text again up here as well as some additional settings like timestamps. Okay, so we have a little bit more data, but the one I'm curious uh the one that actually matters for us is this transcript text. Now, we're going to pin this data. And now you'll see that this node just transformed purple. This will allow us to propagate the data further and further. What do we do next? Well, luckily they do have an open AI node. So, we can actually just go straight down to message a model. This will recreate the same functionality that I had a moment ago. And the way you connect the credentials, pretty easy, too. You just add the API key similar to how you did um before for both Zapier and Make. I already have one called YouTube, so I'm just going to do that. The specific model we're going to be using, if you guys remember by now, is probably GPT-4. 1. That'll enable us to select the model. And then I had a system or developer prompt in make. I think it was called the assistant prompt in Zapier. Well, here it's the system prompt, too. And what I'm going to do is go back to now my make scenario. And I'm just going to copy this stuff over one for one so we can bootstrap it and cut a little bit of the time. You're a helpful intelligent content writing assistant. Let's paste that in there. I'm going to add another message. And then I'll go back here and then copy in the user message. Okay. Boom. Paste. And the thing is naden and make. com both give you a lot more control over the data type and the format. So I'm going to keep this JSON format in here. All right. And then down here there's an output content as JSON field which I could use as well which is what I need to click in order to do that properly. Then lastly I just need to feed in the transcript text down here. Okay. All right. Now I'm going to click test step. We're actually going to send this over using NAD now and we're going to see what sort of output we get. Awesome. That looks pretty good to me. We now get all of the data buried under this message roll content posts object. And again, we have 1 2 3 4 5 6 7 8. Very cool. Very cool. All right. What I'm going to do is I'm going to pin this output. And now what we want to do is we want to dump this to a Google sheet. Right? The thing is all of the stuff right now is buried inside of this one item. So it's similar to how in make we need to add an iterator. Well, in nad what we need to do is we need to split these out into individual items. So the way that you do this is you split out a list inside items into separate items. Then what we want is we want the content. posts field. This one right over here. Okay. If I test this, what you'll see is instead of having, you know, one item, what we're doing is we're actually outputting what looks like nine items. Okay. Now, if I pin this and then if I go here and I now I try and add to my Google sheet, the function I want is appended row in sheet. I'm then going to add my own credential. But again, the credentials are very simple. You just sign in with Google, which is pretty neat if you're using the cloud version. Then I'm just going to select the one automation and three platform sheet. Down over here, I'm going to go nad. And then I'm going to map the columns. Post is just going to be message. content. posts. But you'll notice that there's this use the minimize API calls option for greater efficiency if your sheet is uniformly formatted. Well, the reason why they do this is because NAD has a pretty poor integration with the Google Sheet suite as of the time of this recording. And so you're actually likely to get rate limited if you try and dump a ton of rows into a new sheet. So in order to minimize that, you click down here, select minimize API calls, and then you click test step. Now, if we go back to our sheet under NAD, you'll see what it's going to do is it's just going to dump all them in simultaneously as opposed to make, which looped over one by one, and then Zapier, which also did some looping under the hood. Okay. So what have we done now on three no code platforms? So, we've received an email, we've scraped the transcript, we fed into AI, we've generated social media posts, and then, as I've shown you, you guys can post to platforms using both the built-ins, or if you're using something like NAN, might be a little bit more complicated, but you can still get it done. And just check out my video, sell this AI content system for 1K, which will show you how to do the graph API connections, the LinkedIn API connections, and so on and so forth.

### [33:14](https://www.youtube.com/watch?v=MjIngZJHjBc&t=1994s) Outro

Hopefully, you guys found that video super valuable as a comparison of the current major noode platforms out there. As you see, it's not necessarily that one platform is always going to be easier than the other, but really they have different things that they're really good at. And some things are unfortunately annoying in one platform versus the other. And that's just sort of, you know, what you have to accept if you choose to put all of your eggs in the basket of one platform over another. But as you guys see, if you learn a little bit of the fundamental logic of how, you know, data flows through systems, you can pick up more or less any code platform in record time just by implementing some uh pretty straightforward first principles. Thank you so much for watching. I love putting together videos like this and I also love working with people on developing not only no code flows but no code businesses. If you guys have found yourself on the fence wanting to start a noode or AI automation business, then definitely check out my community maker school. It's a 0ero to one daily accountability roadmap where I run you through everything that you guys need to do in order to take one of the systems that we might have built today, then actually put it in front of a customer and have those customers buy it. Other than that, please like, subscribe, do all that fun YouTube stuff that gets me to the top of the algo. And I'll catch you on the next video.

---
*Источник: https://ekstraktznaniy.ru/video/12155*