I Would've Saved $400k and 3 Years If I Did This... (copy me)

hey there I'm Parker X and I Led Tech for a startup that did over $50 million in sales and on this channel I go through AI strategy and the things that I'm doing in the field because I started a bunch of different ideas after that exit that didn't work out but required me to be coding required me to learn all the AI things and I'm trying to push the limits of what you can do with it when it comes to actually making money not making hobby projects trust me I did that for way too long so we're on the road to 100K a month and I will be going through a couple different things in the videos so here we are I'm trying to get a better format for this what I think makes the most sense right now is we'll do some AI news we'll do questions and answers as people start to discover the channel we got a couple questions which is awesome we'll do yeah the AI news and not just like hey I'm another verbal newsletter because there's enough newsletters out there but just the stuff that's interesting if there's nothing that's interesting I'm not going to talk about it because what's the point and I'm fixing my camera while we're talking to each other um okay and then some strategy stuff and then actually building something so let's get into it the first one somebody asked about validating ideas and it's funny because I have a lot of experience doing this the right way and the wrong way let's see this is on the daily uploads Channel cool we got a number one out of eight video that means 13 views cool this guy XZ Jordy says helpful suggestion cover a topic on how to validate ideas Okay cool so I think there's different levels to this right like I was really really good at validating ideas when we we're when we were like going from 1.0 to 1. n and I'm assuming you're not talking about that but I want to just distinguish the two because if you've ever read the book which if you haven't it's kind of like a manual on making products but in the book 0 to one by Peter teal he covers a bunch of this about how to be different how you kind of want to operate as a monopoly if possible and how to validate your ideas so he's going to have a way better approach to it but instead of you know going and reading like a 200 Page book I'll try to give you my best rition of how I think about it but there are two camps so you have Camp one was 0 to one which we'll cover and then we have Camp 2 which is 1.0 to 1. n n being any future version now in this case we'll start with the second one because it's a little bit easier but if you already have something that's working working defined as let's say 10K a month and if you have something that is working then the easiest way to do that is in my opinion just to go and do a prioritization kind of exercise and it's called ice and each one of these stands for impact confidence and and ease and this is actually going to be the same thing over here you'd still use ice but you'll have less information because you don't have any customers but you can get information so there's usually some phases to this so in both camps you're still going to want to do some ideation obviously like you're going to be validating something you need to come up with the ideas and ideation should be inclusive of research which you could use Gemini's research and within the research phase you could be looking for macro Trends and that could be going to you know something like statista and there's a bunch of websites like this I've spent a bunch of my life just going on these websites to try to put together pitch decks to raise money I was able to raise like 51k off of just doing stuff like this but I don't know if you went let's let's say like CH for world statistics is a really good one as well but if you're trying to or our world data if you're trying to validate a problem from a macro perspective and to see if there's a growing Market because you always want to be in a growing Market I've done this the wrong way where I didn't do this exercise and I try to build something in music and what I wish I had done to save like 18 months of my life would be to do a little bit more research up front on the macro trend of realizing there are zero companies in music that are making money like that should have been like that would be how you validate it that would be a nogo forget it especially if it's your first time going after Adventure you want to do something that you don't have to go against the grain but like let's say you're trying to validate maybe like let's look at what they have for artificial intelligence so we can see like maybe some growth metrics and then in the research section you could see how much investment's going into artificial intelligence by type you can see 2021 huge spike a lot of the charts in terms of money will always spike in 2021 that's because we had this thing called the zero interest rate phenomena which was Co hits and then we print a lot of money and so we can kind of like be armchair macroeconomics people and just like say that that's just going to be a chop but you'd look at this and You' be like okay like if I'm building an AI like there's a nice little macro thing going on right so that'd be an easy one and then on your research phase I would do micr Trends so these ones can actually go against the macro Trends but the macro ones typically will matter more I'll use the use case of my startup that I did that was essentially just like Airbnb for musicians and the the micro Trend was that musicians are pissed are mad musicians get ripped off and I tried to apply my skill set of marketplaces because I had scaled one of those with a bunch of friends and we had done it you know successfully to 73 million at our Peak I always say 50 because that's around where we sold because we did go down due to not raising money and different Dynamics but you know if you have this micro Trend where like okay like musician you have the actual venues thems and then you have the booking agent you have these three parties going on and you have this trend where like a lot of the money just kind of like gets deleted between these steps and because of that Dynamic you can go on to Reddit you can go on to any like Forum or Community about musicians at any level like the basically the 99% that are not at the very tippy top and you'll find that they're mad so you want to get this and even if it validates your idea it's it could you know you have to weigh this into the whole kind of process of of validating it when you're doing your research so cool then you did the micro Trends again like if there's a macro Trend that's saying Spotify which is the biggest one is still not profitable then it would outweigh the micr trend so now you've done some research the next step that I would do is something along the lines of like there's a business idea prompt which we can test out this is like a newer way of doing it and it would save you a whole lot of time by kind of poking holes and things the other manual way of doing this is just write a one pager and it's called the fa pfq and I would pick one of these options so like you can do step one with be the research step two would be one of the following within two so 2C is the YC application y combinator if you're not familiar the most popular kind of Silicon Valley darling for accelerator accelerator and they give you 500k and just application alone is something that is kind of ingrained in my brain because I've applied seven times the first time I got to the second stage and flew out and pitched the founder Gmail now but the application requires you to really think through the idea and it's better to do all this stuff before you go and build it like there's a really good like a blink and basically says something along the lines of like if you were cutting down a tree and you had eight hours to do it you'd spend 6 hours sharpening and then two chopping so it's something like that but the idea being like okay if you're going to sit there and just whack away at this thing the tree being your successful idea in the marketplace and making money instead of having this dull ax AKA dull idea that has not been through the idea maze then you're going to waste a lot of time which I really really relate to so it's very very important and this is kind of what I did when I decided to go after making content as a means for marketing my communities and building personal brand and I think that that exercise alone like made me think through like who are the people I'm talking to like even looking at this whiteboard in front of me it's like okay I'm talking to a mass audience of people that are new to AI there's this massive Trend that's going on where Goldman Sachs is talking about 300 million people being displaced by AI by 2030 cool that's something to bring in that's a huge macro Trend the Investments off the chart the micro Trends are basically that every single Niche is trying to figure out what's going on the micro Trend that I'm going after is the developer Trend around AI kind of writing all the code how do you leverage it how do you stay on it and that's where these videos come from it's like I'm going to do this anyways so I might as well share it so then once you've done all this stuff then I would say you can go into kind like that ice method and when you're doing that you'll just take your spreadsheet and was this under 1.0 1.9 this will apply to both so we can just kind of this will be yeah this will be umbrella the ice method would then be okay you have your spreadsheet and it would be idea and it just is a oneliner that explains either the feature like what is it how does provide value how does it work and that one liner is something that comes out of the YC app so I am having a flashback to pitching my idea to the founder of Gmail and I was like so nervous and I went in there and they say don't use marketing jargon about your idea and I went in and I said we're building metal it's the you know I think this is in like 2019 or 18 or something I said something along the lines of metal is building the health dashboarding of 2020 so it's like super like that would be a good hook maybe for a direct response ad but it's not good for a technologist that's trying to Define what it is something better would be B Toc SAS BC Health Texas to answer all the questions you have about your body like something like that um because then it defines who it's for it's for consumers it defines what the niche is it's Health it defines what it is it's a dashboard for understanding your body so then you'd have your idea and then you'd write down the impact and all these are 1 through 10 you write write down the confidence and then the level of ease to implement it so when you're going through this exercise these first two ones are pretty easy and they all lad up to a goal right so in 1 Point N your goal like this is the real differentiator is like what the goal is so if you already have something that's doing you know I'll use Delivery Dudes as a case with food delivery if our goal is to reduce the deliver three times by 5 minutes because we knew that every five or every one minute meant like tens of thousands of dollars to our bottom line then you know we'd make a lot of money if we did this and then we can reinvest that Capital if that's our goal in the 1.0 to 1n then we'd list out different ideas like oh if we added the better notes to the delivery driver app if we added something to the restaurant app that had a notification to call them instead of just ringing the tablet because it's busy night and that might help there's a lot of different ideas that you'd list out and then you'd end up giving them scores so what level of impact do you think this is going to have going towards that goal so you know you have like idea one and you'd say I think out of 10 this is a nine and I'm only I'm not that confident like if it did work with calling them then I think that would be a huge impact because most of our restaurants are really busy we get our biggest order Spike volume like you're basically painting this picture where it's like okay I know throughout the week you know Monday Tuesday Wednesday Thursday Friday Saturday I know Sunday I know that like we we're pretty much like this and then we go like this every week like really really high over here and then it comes back down so I know that if we're making the bulk of the cache right here that if we brought in this busy notification thingy that it' have really high impact but I'm not that confident in it because if they can't hear the tablet which is already loud why would they hear the phone maybe that I'm just you know I'm not sure so I'd give it maybe like a six and then this ease question comes into the like technical implementation of it so then that's when you're going you're like okay well I know we have for this phone thing it's like well we have an API and you know you'd have a technical person here if you're working with others we have our API it's written in go we've got the tablet app it's written in flutter because we were crossplatform this tablet what else we probably need like twilio do we need it to be in you know today's day and age would it be like like an 11 Labs voice do we need some sort of ivr are they able to call it back You' go through this kind of question and this is like called scope hammering so then you might list like what are the things to be able to do this such that the ease is high so when you go through that with someone technical then you end up with a score and you get the average of that and then you have your ice score right so at the end of the day like 967 let's just call like seven .1 or whatever I'm not doing public math here but if I'd go through and I'd say okay now I have this list of all the ideas I'd go through it and I'd rank it in descending order so it's going with the top one that should be the easiest that I have the highest level of confidence the most conviction around its impact and it's going to be the easiest thing to do so that same thing would be done for a new idea and I kind of as you do this more and more and more and more you really start to like ingrain this process of you know impact confidence and ease into how you think about ideas and then you know you'll start using stuff like scope hammering and and all those so I hope that was helpful and if you guys have other questions then we can do stuff just like we did probably yapped a little too long was like 15 minutes but you know it's the first question so and then that brings up this other topic which is just the power of compounding and everyone knows that investing always talks about this so it's like you know if I do a compound interest calculator what it means is that if you DCA or dollar cost averaged into a stock let's say I have a thousand or yeah let's say I have $100 and every year I add $100 and the interest rate is six 6% and I have 30 years to do that and it compounds 12 times annually because Let's do let's just do one basically the future value would be 8,954 which is a lot more than just adding $100 a year because that 6% rolls into the next 6% it compounds on one another it's the classic snowball and I wanted to bring this up because it applies to everything that you do in life it really does where it's like if I use that same concept of the DCA the dollar cost average where if you had you know 52 weeks across the year I'm not going to write you know one one time each I'm not going to do one through 52 but if you did it 52 times and you each time got a little bit better at whatever the thing is that you're doing like in my case it's these videos so I'm going to not just do it 52 times a year but I'm going to do it 90 times in the next 90 days then every time you get a little bit better and every time that you get a little bit better it stacks on one another and it goes from this like really really heavy climb which is you know you got your your sticks out and your your climbing Mount effers to get up there but then you start to get momentum and it becomes easier and easier and easier and I just wanted to bring that up because I'm noticing it too where when I made these videos maybe like the third time I tried you YouTube I think I've tried YouTube like a hundred times but it was a lot harder because I didn't get in the Cadence of doing the daily Habit to be able to do the thing and when you don't have that daily action that you have to take which for me right now that I'm trying to gain you know so-called compounding interest on like the daily compounding list which is like also comes from a really popular program called 75 hard it's like the most popular fitness thing and they say like 15 mens of reading two workouts at 45 minutes each one gal water what else is in that you do a daily like accountability photo because at the end of it you get to see how your body changed no sugar no alcohol and I think that's pretty much it and then the second level of that they actually say like to add whatever it is that's for your occupation or goal that you have but for me you know I've done this I think successfully like two times and then I've started it and then manipulated it like five times but the daily compounding for me right now is the daily videos right so I could hardly use this tablet I don't know a week ago and I had zero subscribers on this channel now we're at 40 and I'm learning how to write in public I'm learning how to speak in public the less times that I say the words across a 90 minute video 60 Minute video the better that I become I can kind of see that in real time because you won't hear these ums in the videos because I go to Premiere Pro and I do text based editing which is an AI feature and I can see that at the start I had 100 ums across an hour which is very misleading and confusing for listeners you're trying to be authentic you're trying to be an authority and someone to listen to and when you hear someone say well you've lost interest they lost their confidence so for me it's the daily videos it's the upwork Outreach it's the blog posts and then it's the addition of new AI automations addition or changing of so I wanted to bring that up because if you're trying to do anything in life the power of compounding definitely applies to it it's allowed me it allowed me to do what I do now which is I had a job and then I invested all the time half my paycheck every month and it rendered me a lot of cash to be able to do stuff another idea that I saw from cat Cole who she started no she did not start she led marketing for yum brands which is gigantic Hospitality food holdco I think it's Taco Bell Cinnabon all these things are horrible for you but she did a great job marketing them and she's gone on I think she does athletic greens now but she brought this thing up of the hot shot rule which is the AC of thinking of someone that they that you admire and then pausing and reflecting and asking what they would do in your situation and then answering what they would do and then acting on it and that has to do with a lot of different areas of life where people that are successful they don't bring emotion into business they don't sit there and think about the call that they had or the email that they received or the rejection that they had over and over and over analyze it they just move to the next thing so that's something I'm trying to bring into what I do because it allows you to move faster so thanks for the question I know I went on and on but the next is on news so everyone on Twitter in the little Twitter bubble has been saying mCP a lot it's funny because it came out in November why is it so popular now well it's allowing people to connect different tools to their models within their AI editor so it's becoming more and more popular and there's also this trend of lm. text so I think both of these are going to be foundational to the way that we develop software and frankly the way that other tools Beyond development tools communicate with llms it probably means that a lot of companies that jumped on the bandwagon of adding co-pilots maybe didn't need to do that and what I mean is if you have something let's say I have a documenting tool I wrote a new or no let's use something that exists already let's say I have click up oops I'm a product manager click up I got to refresh I have this pen and I accidentally click stuff and then I have to refresh but let's say I'm I'm click up and this may not be the best example because it's gigantic but if I needed to add AI stuff it would make sense to have a lot of it be internal such as writing slash editing docs or tasks but if they start building in custom Integrations I think that's where mCP or some sort of protocol will replace it which is great because integration software sucks to write and it doesn't do anything that unique there's really no value in the actual IP it's just requirement to be able to bridge two apps so right now in tools outside of AI editors you have to write these custom things that are the bridge that keep things either bir directionally so a two-way where they're watching one another and it's in real time or unilaterally where it's just one way so I think that mcps are going to be the foundation for another protocol Maybe will be mCP but I don't think so where you can just bridge the gap between things but there's a big problem that's going on with these and there's a big problem that's going on with another kind of step in the right direction but not the ultimate kind of way of doing things is with the lm. text so if you're unfamiliar with those I pasted some here where when you're developing software you obviously have the phrase R and D A lot of people forget about this one like this R but it is really important and so when you're going to do something you know you used to a year maybe three years ago would go to stack Overflow and you had a question or you go to the documents and you'd spend a lot of time doing that and then you go develop the vibe code thing totally destroys that idea and that's why there's so much crap getting built unless someone's actually good at coding begin with but that's different topic but the lm. text basically says okay I am super base provider of a software toolkit or service in this case it's a backend as a service it has authentication databases all this stuff well database with postgress but the problem is when you go to their site there's just a ton of different documentation and it's not that that's inherently a problem it's great that they have all this stuff but when you're trying to get something done and you want the llm to do it these documents typically change pretty often and you may have issues when something changes in here because it is going to reference the wrong Source material so when you're looking at this you need to know okay well I'm actually not doing database work I'm doing storage those are different sounds similar but they sure are different and I need to do standard uploads and I need to do standard uploads in Python and I need to do standard uploads in Python for a content type of jpeg Well turns out I just had to navigate from docs to guides to storage to uploads to standard uploads and then query against what type of language it was so you can by all means just paste in docs SLG guides and into wind surf or cursor but it's not going to do that great and that's because it's just a lot of information to go through like if I looked at this I have 1 two 3 four five six seven eight I have eight different areas to begin with and then in this one this storage one this is really small if I went to database within those eight I don't even need to count these that's a lot of stuff and even if I click into one of them then I have even more stuff so the way that the companies are trying to solve for this that provide all this information is with these text files so the way they're doing it most companies that have it is you just go to their rout and then do a for lm. text which means in their public folder you can publicly see this and then you'll see they're broken apart by the area that we just saw they don't even have all of them they just have the sdks so when I go into here now it's taking this JavaScript reference and it's basically a very large and detailed markdown file with Snippets in it which is content that is optimized for AI readability very different you and I wouldn't want to look at this this is really dra doesn't have the cool gradients it doesn't do it for us but for a machine it's going to love it right so a lot of people just don't know that that exists and it doesn't matter how many YouTube videos or Twitter posts go out about it it won't go to the masses until it's a lot easier and that's something I saw from a Dev tool Creator I think his name is Dax he makes this thing called SST that's besides the point but he mentions that his tool is getting bad reputation because people that are trying to use it instead of the old way which is if we storyboarded this it would be okay the dev is here and he's reading the docks he finds out about this new thing he checks the docks if he's lazy he used to just go to stack Overflow that's no longer the case if he is Savvy he would go to GitHub and search a an operator with maybe like typescript as the language and then the name of the package so you could see someone that implemented it on an open source perspective but no one does this anymore at least the vast majority of new people and he wants to take advantage of them he wants them to know about his tool and use it and be happy but instead they go in and they you know there's no more docs it's just either chat do or an llm that's running right here and so what happens is this may know about the package that's called SST I think it's I think that's what it's called but the problem is the model OKAY the brain that's running this is It's trained on Old info so the cut off might be October 2024 but guess what that going to be on like version 1.5 and we've gotten from 1.5 to 1.6 to 1.7 to 2.1 to 2.5 we might see like a whole jump in how the package works and so when they go in and they're yapping and they're like this thing's broken well guess what it's because you're on an old version and even if they had the wherewithal to go in and add it into cursor you still run back into that problem I was talking about where it's like you know you have the things that are just so deeply nested so that's what this tries to solve I think somebody will build something that basically fixes the distribution problem it's not a technical problem I think that this sure can get better where if these were existent for every single Dev tool out there which the ones that are on top of their stuff are doing this then there no longer really a technical problem it's a distribution problem and my forecast or prediction for this is that you'll see the idees do this eventually but it will first be like I think this is the kind of end game is the IDE will be able to recognize it'll say hey like you have these packages and we noticed that there's not only lm. Tex there's tax support for it but there's also an mCP for it and then it'll be some sort of little native popup that most people ignore but if they're smart off they'll hit cool and that'll probably happen in the order of cursor and wind Surf and then GitHub V sorry vs code will show up like three years later more realistically because we've seen this before is in the next call it three months we'll see it in an extension called Klein this is the one that leads the pack and everybody copies so I'm guessing they'll be the first to do this which you know the feature would be mCP plus. txt scan SL apply scan and apply so it' go through it says oh whoa you've got super base guess what we noticed that it has these things we just applied it for you it probably be an auto op in because this would be so valuable for so many people that it should just be in the Box I think they'll do it first and then shortly after you'll see a fast fall from cursor so who knows between these two I think it'll be cursor first to follow and then later you'll have a vs but that is a problem if you guys didn't know about that start using those next up is some builds let's do some builds we're 37 minutes we'll try to fly through this so This fifth grade writer thing I want to build something that makes me a better writer that makes other people better writers and I watched a video just before this and it dawned on me that the best communicators in the world right at a level between third and fifth grade which is crazy you look at people like Steve Jobs and you're like wow this guy invents he is so smart he must have this like crazy technical vernacular where he's saying stuff you know like like Hertz he's saying Hertz he's saying gigabytes he's saying all these well guess what he doesn't say any of that instead he uses simple simple language the classic oneliner of a thousand songs in your pocket which went Uber viral and then even more Monumental was the 2007 release of the iPhone and if you take his speech and you transcribe it you can run it with whisper and then throw it into something that analyzes like Hemingway editor which I needed oh I do have it great if you take his and you put it into here it will tell you the readability of it in the grade and I think that Hemingway it says no AI well it does have ai now I'm gonna have to check that out after this but I want something that does this in my content pipeline because I want everything that comes out on the internet to be at a level that everyone understands and it's more memorable so other examples I don't care about your politics but Trump is it's not even Up For Debate he's one of if not the best communicator on planet Earth right now because he's able to get a majority of the country to buy into his ideology he wouldn't even use that word and I think it's a scop where a lot of people that are smart will want to use these words to impress others there are definitely times and places for complex language because it actually saves you words there's one word that all of us together know what it means and it's going to save us going through stuff like if I said Rag and you're an AI you know what that is but I'd never use that for writing in public because no one's going to get it unless I know I'm talking specifically to those types of people but Trump uses really simple language his speech when he announced running was kind of the quintessential example of this where it's written as a third grader there's nothing complex in it but the topics are crazy he talks about how politicians are not suit suited for the job how he's going to fix China how he haven't done anything impressive in a long time and the same thing is seen with Mr Beast largest you YouTuber writes like a fourth grader ER another popular YouTuber his most popular video written like a fourth grader so I want to build something that does this and I think there's no better opportunity than to kind of maybe just I'll just sketch it out what would this look like I think the version one in terms of going back to earlier with the ice framework like I can just shorthand this where I know that I can run the Cal in my head roughly so maybe the first version of this is just a keyword expander or key keyboard text expander these will work for oneshotting small amounts of text so the use case would be that I'm writing an email or a message or a YouTube comment or something that is one to four sentences and I want to have it be stronger so I could write something that would require some sort of small prompt an example of this is when I hit the text key and I go like this and I do GPT concise it expands and it says I want you to edit my writing I'll share an excerpt with you to edit as follows make it more clear and more concise use simple words give me three variations and here the excerpt so I have that and it's super helpful I use it dozens of times a day and what I do is I just pop this open and I do that and then I paste in the piece of text that I'm working on so I think this would be the easiest version of it but I don't think that it would be what I'm going after because it doesn't cover my goal which is I want every piece of my writing to be graded oops to be graded because I want to see it before because I want to become a better writer I don't want my muscles to atrophy so I want to be able to get that feedback get the smack on the hand saying wow you Tred write for PhD grad student now I want it to be graded seen then updated and sent and this is for a Content pipeline so doing this keyboard expander would actually not solve my goal so while I have a high level of confidence I think that would be really easy in this world the impact zero confidence is zero level these is 10 so the score is not going to be good whereas if I said I want to add GPT module in make.com that adds see adds the prompt which we' need to meta prompt it so that's really two steps it does that I know how to do those things those are easy it has examples in it so the job speech maybe a Mr B script snippet of it and it notifies me before it goes out so I could get some sort of pause with the grade so you get a grade plus HL HL means human in the loop it means I need to do something before it continues not sure how I'd get the grade so I'd actually be curious how Heming way editor does it I'm building a or sorry you are a helpful software development architect and are responsible for answering questions about software and how it works research how the user's topic is implemented in the real world find specific examples outlining the technical details and the algorithms used or libraries used languages used you might find this in podcast interviews or YouTube videos and return a product requirements document that outlines what it is why it exists how it works any specific technologies that were used and the user stories required to be able to build the given topic and as I was saying that I why I paused was I forgot that Gemini deep research is better than grock now because it has access to YouTube and it's insane so paste that in there and let it run but that's besides the point I would then take that research I'm not going to let this hold up the video I'd take that research and then I'd bring it into the ice score so that's a good example of how you would do that and then what I do know is that this would be a you know a 10 out of 10 on going towards my goal I'm really confident in it so I'd give it about an eight and then three is TBD on the Gemini response so that is something that we can think about building the Gemini assistant for agency work this is just a use case I wanted to share because I just talked about demiz release yesterday their updates and I used it for real life use case so when I type in someone's name let's say John carac see how when I type in his name he has more stuff than if I just typed in some nobody well that's because he has a Google Knowledge Graph and these require you being either famous or technically Savvy and I'm the second one so how it works is Google needs to know who you are needs to know an overview of who you are they needs to know information about your birth your marital status there's a bunch of different things in this overview panel and then you also have this people also ask thing if you're really famous but we wouldn't focus on that but you get instantly legitimized when you have one of these let's say you have a client or you have someone that is just curious about who you are if you typed in your name and you had this inst still legit it's actually insane and it's the same thing with companies and it takes a while to get if you didn't take the steps that I'm about to explain so one is obviously you need to be on all the different social media platforms you need to be everywhere you need to use the same name don't spell your middle name in one and not in the other because robot's going to be reading this and trying to figure out if you deserve one of these other steps include getting backlinks so that's where this gem could come in and be helpful so if I go back in I'm going to start the research I'm going to open a new tab and I think that's the same tab but whatever and what I did was the Second Step of it with getting that Knowledge Graph is feeding it information about yourself and asking it to go and find every link about you where you're mentioned for high domain ranking websites so I say I'm building out my Google Knowledge Graph and I need to get all the links that are high domain ranking that I appear on read the following context of who I am then conduct research to complete a to find a complete list of all the websites I appear on I give it context about who I am mimicking what we just saw here right and once I give it that then I give it the links that I knew off the top of my head so I had written a bunch of articles in 2018 on a website called hacker noon and those are good they're high quality links because Google is just going out and it's trying to figure out is this person legit and they rank it in a bunch of different ways but links is one of them they index like billions and billions of pages so you need to help them and it goes off and it does that and then it comes back and it shows me all the different links and this is something that people do fulltime for living is part of the SEO process so this is just really helpful I wanted to share it you can see that it has a bunch different things you can go and read its thoughts if you want to you can see the different websites that it came across and I think that that would be a good gem so let's see if we can't take it and make it and then I can post it on the website so if I go in here I want to create a new gem and I'm going to call it SEO deep research link gatherer and let's paste that in now instead of this because I want people to be able to do this I'll put uh I'm actually going to use XML just to beef this up a little XML is the language that's used when these things scrape enter in a flashy but not over-the-top two to five sentence explanation of who you are what you've done why you're important include as much Social proof as possible example if you're a writer include I wrote an article in Forbes or I have a best-selling Amazon Kindle book if you're a programmer you might write something about an award you won or some sort of accomplishment cool so we have the context links all the tag there close it out links that's not it let's go down to the bottom cool so we have here we have instructions purpose or goal then we'll close I'm building out my new Google Knowledge Graph and I need to yeah actually we can just include all that angle close that out if I wanted to beat this out I would do step-by-step instructions that would look something like this so I do instructions open I do instruction open and then I proceed to give it step-by-step instructions but I think this is fine cool and then I can add files to reference that's great so if I needed to I could add some files and then let's create it when I hit save see if I can't share it deep research how do I share this gem share I can't share that's too bad ohuh oh sorry about that you can copy the prompt what else I think I'm yeah I'm going to wrap the video here no I'm not I'm going to mention one last thing just about the content Playbook you can check a previous video but you're going to see this in real time this is why watching this video if you've been checking every once in a while wow I'm red I need it do I have a filter on this thing what is going on didn't realize was this red oh crazy but you'll be watching me in real time implement the stuff that I talk about on this channel across the entire internet so on my other channel soon Instagram soon medium LinkedIn Twitter Pinterest Facebook Tik Tok all these channels so the thing that I'm thinking about is how do I do the highest leverage stuff least amount of effort highest Roi kind of like the ice method so one of them I'm thinking about is well with Instagram I hate the idea of Facebook but maybe I try something like this stupid dinosaur where could I put interesting stats about or news about AI that stands out this would require way less work I can use a perplexity prompt that I wrote to generate the backgrounds and then the text is really easy because then you can use canva and this is an example of somebody doing this that that's something I'm thinking about with this content Playbook that I have which is in here and I outline every single step of what I'm going to be doing and the tools and then how I plan to grow to millions of people learning about AI so that they can learn more and be more successful that's it for the video today happy Saturday these go out a day after so it's Sunday when this is getting uploaded but I love making these videos I love when you guys write comments I got one and I was telling my fiance about it like I got a comment and it was my birthday two days ago that's why that's there so drop a comment below if this was helpful and I'll see you in the next video oh and all the resources for this video like the Whiteboard and the prompts and anything that we talked about will be on my website just Parker x.com blog see you oh I got to make a prompt thumbnail the name of this video is going to be how to validate ideas and I'm going to be looking at a list of icons that are right here that's one option or looking like I'm pale and dead because I didn't validate the idea which was me like the last three years okay see you