3 Boring AI Automations You Can Sell to Coaches for $1.5K

Here are three boring AI automations that you can sell to coaches, consultants, and anybody with an inbound YouTube funnel for 1. 5K a pop or more. I built them all in make. com and naden. I'm going to run you guys through all of these systems in a moment, plus a bonus CRM that I built specifically for

coaches. Our first system is a YouTube content repurposing flow built in make. It's a little different from most other YouTube content repurposing flows that you guys have probably seen because it actually has a unique twist where we have AI points out inadequacies in the source transcript and then suggest ways to improve the source by adding new ideas and essentially just changing things up, adding unique angles. So, as opposed to this being a parasite flow like everybody else's, what this is this is a parasite plus flow. It changes the source content and allows you to constantly be adding and innovating. Let's say we got a YouTube video over here. These eight AI trends will change business forever. get in early. If I go back to my flow and I click run once, now I've already cued this up. We've already done the scraping and click use existing data. What we're going to do is we're going to start by processing the transcript. Now, processing here is doing a lot of heavy lifting, but in a nutshell, all we're doing is we're passing this into AI and we're asking it, hey, I want you to take this transcript and use it to generate an outline. Now, the benefit of generating an outline with a transcript as opposed to just generating content directly based off the transcript is when you take the whole transcript and you funnel it into an outline. What you're doing is you're forcing all future steps to generate unique content, not the same content that a lot of other people are going to be doing. So, a quick little example of what the prompt looks like is we're generating content based on YouTube video transcripts. Your task is to take us input a full transcript of a YouTube video and output a variety of parameters like summary, like outline, and like suggestions. three ways we can modify the content to improve it including twists, different ideas, areas you could add more context, etc. And then spicy preferred that spicy preferred does a fair amount. When we feed it into the next module here, if I check the prompt as well, what you'll see is we are generating high-quality content based off YouTube video transcripts. Your task is to take us input an outline for YouTube video now and then implement the suggestions to make the outline more detailed. Okay, so what we're doing is we're not modifying a source transcript one to one like a lot of other people are doing. What we're doing now is we are having it change the outline and then we're going to take that outline and feed that back in AI in a moment which is going to allow us to generate significantly more unique content. Next, what we do is we generate a bunch of social media posts based off of that. So, we have guidelines for Twitter, guidelines for LinkedIn, guidelines for Facebook, guidelines for Instagram, and then we basically say, "Hey, go for it. " One other cool thing we're doing is we're generating a YouTube script with that. So, if you wanted to, you know, record a YouTube video based off somebody else's content or at least based off the idea or the packaging, but then you wanted the actual stuff inside of that script to be written differently and maybe framed differently, this is how you do it. And then finally, sort of the very unique thing about this is we then feed in that outline and section by section we generate the portion of the script that corresponds with the particular part we want to do. So, we're generating content section by section. Your current section is and then we give it the section. Now, the reason why this cyclic approach is so valuable is because most of the time AI models have to trade off depth for volume. And so, if you feed it in a giant list of headings or whatever, it necessarily will not really be able to produce a lot of content per heading. What this approach does is it combines the best of both worlds. We get to go super deep into every specific section. And then what we're going to do is we're going to have AI process it using one final aggregation step to weave through all of the issues in the transcript. add little intros, outros, connector sentences, and so on and so forth to turn this thing into just one giant, beautifully formatted piece of content. Now, I've actually built this entire system live on YouTube, starting from first principles. I always do my live builds in such a way where I have no idea how I'm going to put the system together before I do. If you guys want a more detailed breakdown of what this one looks like in particular, just check out the video that I've pinned above. Now the smooth out script module currently uses the best performing model available through the open API which is the 01 model as the time of this recording because that it takes a fair amount of time to work. We are feeding in something like 20ish,000 tokens at a time. So usually this step takes a minute or so. Once that's done all we do is we create a Google Drive folder then an Instagram post dock, Twitter post dock, one for Facebook, one for LinkedIn as well and we do some markdown formatting and then finally generate the whole YouTube script document as well. And then just as a final step, I was like adding things to some sort of database. So in this case, we're doing it with a Google sheet. When all said and done, we have it popped into Google Sheets over here. And then we have the Instagram post column, the Twitter post column, Facebook post column, LinkedIn post colum. Then finally, we also have the YouTube script as well. So let's just pop that open. And as you can see, this is very long. This is almost 4,000 words. So token-wise, it's quite a bit. This comes with headings and it comes with basically everything else. And as was saying earlier, in this video's case, it was like eight business AI trends. What we're doing is we're actually like diving really deep into every one of these. We're doing a historical context, current shift, AI's role in making product uh creation easier, practical implications. I mean like for eight sections, every section in and of itself is basically like a whole article, which is one of the really cool things about this approach.

The second system is a YouTube trend detector. The way that this works is we basically have a Google sheet set up that has a few channels that we're monitoring. So in my case, I've just added two here. There's my own and then there's somebody called Leonardo Gregorio who runs a channel that I just like in AI and automation. We then have a couple of different tables with that channel ID. So this one's mine. This one is his. In this case, I'm testing this manually, but we can also schedule this to run once a day. What this is doing is actually pinging the YouTube API directly in NAD. I've set a limit of five just for time purposes here, but this gets a bunch of data about each video. It gets stuff like the number of views that every video has, comments, the number of likes. What we do then is we filter out all of the new posts, then we add them directly to this Google sheet, and that's one half of the YouTube trend detector. The other half takes that data from the Google sheet, and then it actually just sends it in the form of a daily digest. So, what this flow is doing is if I click test workflow, it's reading through that Google sheet and then it's just looping over before generating an HTML template and just sending me an email. This is what it ends up looking like. We have multiples down below. We have durations, which I've just put in seconds, but feel free to change this for whatever you want. And then, you know, at the end of the day, you have all of the videos sorted nicely. And the whole idea here is you, if you're a coach or a consultant, get a trending list of people in your niche that are doing things that you might want to keep track of alongside all the information that you find important. The

last system is a YouTube content idea generator that uses YouTube comments as the source for the idea flow. We're watching Appify actor runs. So, let me just pump in the same video to a different Apify actor, one called YouTube Comment Scraper. If I start this, what this is basically doing is it's going through the YouTube comments on the video that I provide. And then when it's done, we run the scraper. This flow then extracts all of those comments. It then feeds it into AI. And all we're doing with AI, to make a long story short, is we're just having it filter out whether or not the content ideas that we're seeing, or rather the comments are reasonable. So my prompt is really simple. We're looking for content ideas below a YouTube comment on one of our videos. Your task is just to determine if it's relevant. Okay. Return the results in the array. Okay, here's a bunch of examples where I essentially show you, hey, for a given comment, what sort of idea would you create? And then last but not least, we actually go through and we pass it through an additional AI step that takes that data and then uses it to create a hook and an outline. So, for instance, this is a comment that I received about leveraging YouTube transcripts using Ampify, very meta. And what I'm doing here is I'm making a YouTube video hook and outline based off of a comment that I scraped. The really cool thing with this is obviously then we add them to a Google sheet. I always like having some sort of database. And at the end of it, you end up with a massive content calendar that looks basically just like this where you have the date on the lefth hand side, the comment that you scraped, the content idea, and then ultimately the hook and then the outline. The last system I wanted to show you wasn't actually built on a noode automation platform like make. com or naden. It's built on a project management platform called ClickUp, one of my favorites of all time. And what this is this is

basically a coaching CRM, a COMM. Essentially, the way that a lot of coaches and consultants work are they'll have some sort of like hands-on program where when you sign up to their offer, they pump you through a system, assign you to a coach, and then you essentially have like some sort of daily or weekly or whatever the cadence is call where they walk you through, you know, how to get better, how to solve your problems and stuff like that. So, this is a very simple instantiation of what that might look like just turned into a project management pipeline instead. The biggest issue that I came across when I started working with coaches and consultants was I just noticed how disorganized they were. They usually found like a winning offer in the market and then they exploded their revenue. But as a result, they were usually juggling like a hundred different coaching clients, not knowing exactly where to allocate their resources, not having like any simple standardized way of pumping them through the flow. So, I distinctly remember I had this exact situation with one of my old clients that scaled upwards of $300,000 a month. And the way that we solved it was through a system like this. The way that the CRM works is there's this intake stage up at the top. So, the idea is when somebody signs up to your offer, maybe there's some sort of form they fill out, they get automatically added to this. There's an onboarding stage next, a week one stage, a week two stage, a week three stage, a week four stage, and then a completed stage. And these weeks are totally arbitrary, by the way. Every coaching program is different. Some break things down based off a week. Others of stage. But the important part are the columns. So what I've done is I've included a Google Drive column here. So the Google Drive column is just where you store all the assets that you are working with a client using. You then have an assigne column. So I mean, in this case, I've just assigned myself to all of these, but as I'm sure you can imagine, you're probably likely to have multiple coaches in the company. You can just assign them that way. You have a next check-in date, which is important. This next check-in date can be manual or automatic. In my case, it's manual, but I'm sure you guys could see a simple way to make it automatic. And then there's just a one-click Zoom meeting room. So, if you're hosting all these things through Zoom, you can imagine how when you add a new record, you can automatically create a specific Zoom meeting room for every record using make. com or nad. The idea here is if you are a coach, all you do every day is you just open up ClickUp and then just click that one Zoom meeting room button when the time arrives of your session. There's a few other things like session count, days, and stage, and some other tertiary fields like email address and stuff like that. But let me run you through how you might want to take a pipeline like this and then actually automate it. Okay, so what I've done is I've created two simple make. com automations. One called send onboarding email and the other called create Google Drive. The way the create Google Drive flow works is basically if somebody new is created in our pipeline, what we do is we just make a Google Drive and we update the field. So a quick example of this is let's say I'm adding somebody here. Let's say their name is Leonardo G. If I press enter right now, what happened is we're now receiving the event inside of make. com through ClickUp web hooks. We're then creating a folder and then editing the task and updating it with said folder. So, what we have now is we have a custom Google Drive folder specific to the person inside of a coaching CRM Google Drive. So, I'm sure you can imagine this isn't rocket science, but stuff like this just helps keep companies a lot more organized. In addition, let's say you want to send an onboarding email or just trigger something when they change the stage. What you can do is assuming that we fill in the email address over here and assuming that you know we're filling the rest of these stages, we can set something up so that when somebody moves from intake to onboarding, what we do is we trigger a nice happy onboarding email. The onboarding email is something quick and simple. In our case, I think I'm just saying like, "Hey, welcome aboard. " or something like that. Let's actually go check the email. Welcome aboard, Leonardo. Thanks for joining and we're excited to have you. Our goal is to help you unlock your potential. Here's a bunch of information. Looks like I forgot to put a tab there. Hopefully you guys see the point. You can take a system like this and then extend it, make it arbitrarily complex.

But every one of these systems is something you could sell for at least $1,500, if not more. Hopefully you guys appreciated the video. If you guys like stuff like this, definitely check out Maker School. It's my day-by-day accountability program where I show you how to get your very first AI and automation customer in just 90 days. I believe in this so strongly that I actually guarantee that you will do this or your money back. Our most recent and probably standout win was from the lovely Eric Misho that just replaced his whole six figure job. His monthly retainer surpassed his corporate salary. He was able to submit his resignation and then do this sort of thing full-time. We get wins like this literally every day. If you guys want to add to them, then definitely join. I increase the price every 100 members or so just to reflect the added value. Otherwise, please like, subscribe, do all that fun YouTube stuff that gets me to the top of the algo. And happy automating. See you all tomorrow.