Marketing Against The Grain

there used to be a time where i would spend months and months building in these different ai marketing skills in claude and claude code probably much like yourself you're building hook generators were building skills to create content to do paid ads to create ad copy to help you with your newsletter each skill worked it produced a pretty good output but the content was always kind of not that great pretty average and i wonder if you are experienced the same thing it was confident it was clean it could have been written by any saas marketer at any company in the world but it wasn't incredible it wasn't world class and i couldn't figure out why i had good prompts i had good tools i was put in real effort i was putting in all of these hours every night to crunch through all of these skills and create something that i wanted to enhance my job to create world class output and then i found the answer and it had nothing to do with building skills it was actually something underneath skills that i hadn't even built yet and once i did everything changed about how i used ai cloud and cloud code i'm going to give you those secrets and better yet i'm gonna give you the exact things you need to build this for yourself stop me if you've heard this before your support team opens the q monday morning two hundred tickets before the first sip of coffee they already know what's in there password resets order updates the same questions over and over what if that part of the job handled itself hubspot customer agent resolves those repeat tickets using your actual crm data and knowledge base your team can focus on the conversations that actually need them and your customers can get the answers they need faster check out hubspot dot com to learn much more about hubspot custom agent now here's what most people build an ai marketing systems believe better skills equal better output you see it all over the place ai marketing skills dot m md files that you can download to little allow your ai to do something and skills done m md files are a skill and a text file that you can give your ai system something like a cloud or a cloud code you can give it that md file and then it can replicate that skill for you so we ob assess over skills we refine our prompts we test different models and we build all of these kind of elaborate workflows we spend weeks and weeks iterate iterating on these skills trying to make them better and the i put gets marginally better but it never gets distinctive better it never sounds exactly like that person or is shaped exactly for that person's customer it never stopped this scroll and it never made someone think why that person really got me they created this content just for me i know because this is how i worked and here's the kind painful truth the skills aren't the problem they never were they are probably good enough for you to get the output you want the problem is what the skills are actually build on top off now i love to bring this to life with a really cool story we just dropped a free guide that will turn claude into a full blown marketing machine it gives you four templates that will stop your ai output from sounding too generic each template captures a different layer of context your audience vocabulary your voice patterns your positioning and how your buyers actually convert the more you update them the better every ai skill you build will get get it right now scan the qr code or click the link in the description in ninety ninety five pixar made toy story the first fully computer animated feature film it really was a phenomenon it was a huge huge hit but behind the scenes something was really breaking pixar directors some of them the most talented storyteller alive kept running into the same narrative problems story structure kinda broke at the same points character motivation falling flat in the same ways right they kept running into the same pro they solve a problem in one film and then six months later a different team would have the exact same problem and the les ed cat took from that who were pixar president at the time was there was no shared system across all of these directors every film would actually start from zero the real problem wasn't talent and talent in your case is your skill but md file he already had the best directors what was missing was shared intelligence right away for the collective wisdom to actually flow across every project not just inside of individual people and so i wanna kinda reiterate that point pixar one of the best production houses for these kind of animated films had the best directors on the planet working on these stories and continually ran in to the same problems each and every time and those problems would be installed in one film and then would re occur on another and so what ed cat decided was they needed something called a brain trust and so the brain trust was this kinda of small group of senior creative who would screen work in progress together and so they would sit around and do this and it wasn't to get notes from the top here's what's working and here's what's not it wasn't to approve or reject anything again these were some of the best directors on the planet it was really to share context to build collective understanding of what worked what didn't and why across every film simultaneously and their results were things like to story two monsters inc find a ne the incredibles all of these incredible blockbuster movies that we have probably all seen and known and then disney acquired pixar in two thousand and and six and cat mode did something that proved the point definitively he exported the brain trust to disney animation and so at this time disney animation was really struggling and so they bought pixar and then they bought this creative brain trust and with the same directors and same budget the film that had built forget movies year after year did not replace the single director did not replace any single tools and they just added their context layer and what came out the other side was films like frozen big hero six so so same talent the exact same directors a new foundation that these films were being built on top of and a shared context created totally different outcomes and that's the exact problem we are all hidden with ai we are building the skills we are building the movies and we may be really good at doing that but we're not building the collective brain trust every single scale we build starts from zero just like the films that pixar were building they all start from zero so what do you do about that how do you fill in gaps so everyone is kind of replicating what picture we're doing before the brain trust they're building their skills that md file and they're probably really good skills and they're creating an okay ish content and the outcome isn't that incredible it's because these skills are starting from zero every single time just like the pixar directors with with that movie and the skills are kinda just averaging at the internet so the output you get in the content the product position in whatever may be is somewhat average because it's the average of the internet because you do not have your brain trust built you do not have your intelligence layer build i call this for us the foundational layer every single system you build and when you are building skills for anything you want to do to grow your business in a post ai world the foundation layer is much more important than the skill layer and i'm gonna tell you why i'm better yet i'm gonna give you four files that you should really have in your starter foundational layer to up level your skills so the output will be night and day in comparable to what you're getting today and so i'm actually going to share my screen and get in to really kinda show you these so we'll start off with a quick look at this so most teams are kind of building prompts get an output start and know this is how skills work they're really just prompts in text files what i'm arguing is that everything you build should have a foundational layer and in that layer are core dot md files that help describe who you are how you work and who your audience is and what they react to now the foundational layer can be anything that helps to up level your skills i'm going to suggest four starter files and better yet i'm going to give you those four starter files so you can build your very first foundational layer and i'm gonna take you through really quickly what's in that layer and so again this is most teams over here maybe this is what you're doing today this is what the foundational layer will do will actually help you up level all of the skills that you build on top of and i'm actually gonna tell you how they refer to that foundational layer as that's a really important part so the first one i'm gonna talk about is an audience delight profile most marketers have a traditional ic ideal customer profile and it really is like firmer graphics and and techno demographics and demographics it's fit and intent it's who is the ideal customer that would buy our product or service it doesn't really tell you what marketing would appeal to them what things you could create that with a lesson some sort of emotion connect with them get them to engage with your content with your brand so let me open up what this file looks like i think this is the modern kind of file that i would create you can see here not a traditional ic firm so basically this has some things like who they are in their own words so how they describe themselves how do their peers describe them look at this their person who has their shit together we right back to today's show but first i wanna tell you about a podcast i love it's called nudge it's hosted by phil ag it's brought to you by the hubspot podcast network the audio destination for business professionals and it's the uk's fastest growing business podcast what i love about it is that the nudge listeners love no fluff no bs evidence based marketing tactics they get in each episode you're gonna wanna listen because this is like an nba worth of insight in every single podcast and entrepreneurs you're gonna love the show because it's filled with repeatable proven studies not hearsay say not one off success stories marketers you're gonna love it because it discusses the psychology behind great marketing and what marketers are getting wrong listen to the nudge wherever you get your podcast this is for a profile that i had created for notion to illustrate this point so who do notion sell to at the company notion what makes them light up so topics they can't stop talking about you can see her it gives you like very granular details templates that safe in real time they love their notion setup people who are into notion really kind of wear they're set up as a bad pride they are kind of first movers love new features they love the feeling of replacing different tools they finally said different things like a around productivity they have like finally someone said at moment so these are the things that really capture their attention because they think these things but no one really says it this is pretty spicy confluence where documentation goes to die content they forward to colleagues this is really important this means that it's shareable what makes them reach out what frustrate them right you get into like the real granular details of what frustrate them objections and misconceptions what is their vocabulary i love this one so what are the things they say what are the things they don't say right they say second brain they say single source of truth they say wiki they don't say knowledge management system they don't say documentation repository what pulls them in versus pushes them away right real workspace screenshots here's my actual setup generic productivity advice plan your week on a sunday what i show and have so you can see this audience the delight profile is how i create things that will actually delight my audience okay so the next one is a creator style the creator style is what style do you need to actually bring this to life across all of the different things you create to delight that audience and so the creator style grounds the things that your skills create in this style for that audience so again we are updating all of the market and traditional docs that we used to have with these foundational files we used to have brand guidelines that's boring and now we actually have creator style much more interesting so this really tells us how to create content voice in one sentence it kinda talks a lot about what your voice should be the atomic unit again this is all built for theoretically if we were trying to market for notion tone characteristics the five defined traits so conversational not corporate empowering up patron patronizing and playful this is really good right not silly we have a personality with warmth occasional humor but we never sacrifice clarity for joke we're direct not verbose we have right in pot patterns we have our sentence structure we have our format happens we have our open end pattern we have our closing pattern this is my most favorite actually which is that we always do these things we never do these things and then you can add a lot of the kind of ai stuff in here where it's the m dashes or whatever you may be what signs like notion what doesn't sound like notion and we try to keep these files really short and powerful ai is a habit of getting kinda confused did you give it a lot of different context so the incredible thing about these four files i'm showing you is they all complement each other they do not overlap with each other i spent a lot of time refining them so they do not overlap with each other they all have a very clear purpose so if the skill loads two of them the two of them are complementary they don't overlap and give the skill mixed instructions so we have our audience delight profile this is what really is going to elicit emotions from your audience get them to react to the things that you put out this is the style we need to really be us to really market like us to really sound like us and that could be you as the individual that could be you as a brand that could be you who is not even really create content yet and once they get started on this path and can take a lot of content that you're inspired by so remember this you can say i'm inspired by a bunch of different creators you can create a single document of all of their content you can get this template that i'm giving you and then you can say claude can you recreate this template based upon this content that's how powerful this is and then you can create your own style market position in map so let me show you this one because this is super cool okay we are not doing a stuffy position in slide deck from last quarter that gets stuck in a google folder and never gets used this gets updated your foundational layer will get updated all of the time we have position and we have our strategic claim we have our competitive landscape you need to know this what they claim where they win where they're weak because if you're creating product marketing material or brand material you'd want the scale to be able to know this so again it would only use the parts of the foundational layer that it needs and again i'm gonna show you at the end of this video how to make sure your skill understands what md file from your foundational layer to load automatically make this all really easy for you so you don't even really have to think about it the cool thing is you can just get these templates tell a call to create them for you so you're not even gonna have to do much work once you actually have the template what we own versus what's contested so where do we win and where's contested territory for notion they're contested like ai powered workspace every competitor has a ship ai feature that's very true the claim is no longer differentiated on its own very very true all in one a very kinda contested area where is their white space for notion so cross functional workspace that's a really good one for them they should really lead into that cross functional is a bigger and bigger thing as ai class lapses functions into each other the anti sas is a pretty good one what's changing you would wanna know what the market shifts are and you would wanna update this every single quarter again short and powerful so now we have our audience the delight what elicit and emotions we have our own style now we have where we sit in the market and the last one we're going to do is a customer journey intelligence foundational file and this is one of my favorites which is how you market across the journey because you would want your skills to produce different output dependent upon where the journey someone is so let me download this one this is not a funnel diagram they're old and stuffy this is not a lifestyle cycle stage cr right this is a real living file that will continually be updated up level our skills so our skills are always producing great feedback so how do they find us so first of all if we were using skills to create marketing assets we would want it to understand where our audience find us for notion that's youtube they get pure recommendations they get tiktok reddit at product tongue they get google search what triggers first awareness how do you actually create first awareness look at this the emotional driver behind these moments live in audience dot md files you can see i have cross referenced the different foundational files with each other so they really are a live and breathe in context layer we have evaluation what they ask compare an objective to again you'd want to know all of these things because when they get into the buying decision they're kinda making decisions between you and someone else these objections are gonna be really good for when you're creating sales enable material when you're creating product position in work these conversion triggers what closes a deal the a aha workspace moment they build their first workspace and see databases link views so this is how notion really released sells a colleagues already using it a template that matches their use case a storage or a guest limit so this is very p driven proof points they wanna see the template library breath community size name company usage why do they stall and churn this is actually really good by the way why would they actually churn out of notion and why they leave team didn't adopt it and they found a specialized tool performance of scale context change and then how you can actually expand so you can see this customer journey is really how you acquire to actually expand to actually keep people using your product and how you actually sell in the moment when someone's considering your brand versus someone else again the context of the foundational layer is the core point of this episode and these files illustrate how you would bring this to life will give you these as templates that you can replicate or you might decide that there's things that you want in your foundational layer that is different from what i'm suggesting here now at my foundational layer is different per team i have a content team i have like different ai teams and they have different foundational layers that they will reference but here's the killer part alright k so let's say you end up with twenty foundational files that's your context there i've built like all of these different things i need my skills to refer to but your skills don't need to refer to every single file in that foundational layer before it runs they only need to grab the ones that are relevant for the thing that scale does so it has the context it needs it is not overloaded with context and gets confused so every foundational file has a header that declares what the skill should use it for and then the skill will load this when it's trying to do some of these things so it will say load this file when the skill writes any content blog post social copy emails landing pages do not load this file when the scale only needs audience data only needs competitive the position so for every skill they have this block that you run right at the start and what it's doing here is it's scan the foundational file and looking to see if that is one that it needs to run or not or include or not and then if it doesn't it does not include that one and so that's how you start to build a foundational file that really only gets used in the way it's meant to be used so to reiterate that the foundational file has a piece of text that basically tells skill here's how you use me and here's when you don't need to use me and then every skill that is referencing that file so in every skill the first thing it does it's just gonna scan your foundational folder it's gonna look across all of your dot md files and it's gonna pull in the ones that are applicable for the task that it is trying to do and in that way every update you do to your foundational layer improves your skills and that's the important parts now you have a context layer that these skills are built on top of a folder that says foundational it has all of these different md files and you update those md files every quarter you just tell claude hey let's update these and refresh them make sure that they're the most of the date the best that they can be and every skill that is referencing that foundational folder and the dot files that live in that folder will get better and better and better and the real upside here is if you are able to layer in performance data so you can tell claude here's the things that i produced this quarter here was the performance based upon this how can i improve all of the files on my foundational layer to get better performance these ones performed great this stuff didn't perform great i need more of the great stuff update my foundational layer with the patterns that would get more of the great stuff and i will do another episode just on how to improve the foundational layer with performance data if that is something you really want me to go through tell me in the comments but that is the pixar brain trust approach to build and skills don't ob obsessed so much over the skills ob obsessed over the context the skill is pulling in and make sure it pulls in the right context for the right task and that context is updated frequently based upon your performance what is the performance of the output you're producing hopefully that was of interest to you get your foundational files the start of pack in the comments and until next time this is marketing against the green everyone you know karen and i have been doing the podcast for a while now we've been at this for a couple years we love it we could not be happier to be doing this we wanted to take things to the next level we wanna level up the impact we're having with marketing against the grain so the next step of our journey is something we're really really excited about we're gonna launch the marking against the grain newsletter and marketing against the grain newsletter is going to be amazing if you are a marketing leader practitioner you're in the trenches do in marketing every day this is for you we're gonna deliver right to your email inbox and you're gonna get all the behind the scenes frameworks practices tutorials from us from guest we have on the show and from people even beyond the podcast that we think are gonna be helpful and really have an impact on your day to day week to week doing marketing you're going to love it it is something we've been talking about for a while we're really excited to have it out in the world we've already got a hundred thousand marketers who are on this newsletter please join it's completely free we'd love to have you as part of the marketing against the grain community and it's easy you can click the link in the description below or you can head to marketing against the grain dot com slash subscribe this data is wrong every freaking time have you heard of hubspot hubspot is a crm platform where everything is fully integrated well i can see the client's whole history calls support tickets emails and here's a test from three days ago i totally missed hotspot hubspot at him

on today's show we're talking with kevin hudson for future media and we're walking through how you actually go from an ai novice to a master workflow builder we're gonna show you some use cases some examples and help you understand how to actually move and improve your ai fluency stop me if you've heard this before your support team opens the q monday morning two hundred tickets before the first sip coffee they already know what's in there password resets order updates the same questions over and over what if that part of the job handled itself hubspot customer agent resolves those repeat tickets using your actual crm data and knowledge base your team can focus on the conversations that actually need them and your customers can get the answers they need faster check out hubspot dot com to learn much more about hubspot custom kevin welcome to marketing against the grain thanks for being here thank you very happy to be here i know that there's a bunch of stuff we wanna build on today's show so maybe we kick off really quickly with how to think about where someone is on the ai journey and then let's talk about how we kinda move across yeah i can share about a couple little visuals here that may help by i kinda think of this as the seven different levels of ai most people start you know right on level one where it's the question ask basically where you've just replaced google with chat or claude just all the same stuff used to ask yours right in there so after that you start getting into prompting you just realize that when you ask things differently you get a different result and so when you dive deeper into that if you start putting in clear instructions add context examples and constraints it just changes everything so next to be the power user we've still just kinda focused on prompting there but each of the tools has so many other features that a lot of people don't go deep enough to find out like if we're looking in cloud for example just using projects where you can create a little folder where you can bake all of that context you can upload custom instructions upload files it could be like your brand guidelines sop anything you could think of and then it has that baked into every conversation you start so it's nice just for the organization and keeping all your chat in one place but then every time you ask it already knows what you're trying to do and you don't have to rep prompt every time on this kevin do you mean like would this be equivalent to creating in a cloud project the custom g a gem like a a maybe just expand why that's different from what be talked about in the previous level two yeah so a project is basically like a folder where you can store all of that so i have like these custom instructions here where it says what my channel is what type of content i like to do what type of voice that i use it also has the memory up top that cloud it adds to that every twenty four hours and over time expands on everything you've asked it and things you've critique like oh i didn't like this it just naturally baked that into the memory so every time i do anything in here it has all of that context i have a cloud project for every singular kind of project that i'm trying to do and i think of it as my ai second brain like ai is a real strategic partner because it has a second brain and it has the ability to kind of talk through those different projects with you in a really intelligent way whereas like level two you kinda talk to it in a chat and then you move on to another chat and mute kinda start all over again and then there's i mean tons of other features like cloud has skills where you once you've done something prompted back and forth then you can just say hey turn that into a skill and the next time it actually automatically will pull that skill in and just do all of those steps back and forth so you don't have to prompt that at all if somebody hasn't used skills before like explain a little bit more why why they should like show an example and then maybe like give us a couple of examples of skills that use commonly so that people could understand like the core use case so i evaluate like a bunch of skills like in cloud they're under customize here and so like this one's a content repurpose so you can just put in any sort of link to whatever that might be a news article or youtube video then it will take that and i i have a big list of things not to do that's common in ai writing so kinda like built that into the skill so it doesn't sound like it was written by ai and this is all like baked in every time the skill gets called like everything that i'm going through on the right and of course had claude helped me write all of this then it drafts posts for x and for linkedin yeah i have i think like four main skills i use when i'm scripting through youtube videos as well those ones i just use a ton know that's super ton of good ideas well i think where most people wanna get is is like into that workflow building stage yeah and and we wanna spend i think the rest of time kind of helping people understand what that looks like what some of the core use cases and tools are there most people will kinda just stick with their ll of choice but then around here is where you realize there's just so many other tools out there and it can get super overwhelming so i have like five main tools that i use a lot that are kinda pretty broadly applicable you people are gonna have tools depending on what they do but there's a few that kind apply to most people and then once you really get in depths on going over a couple different tools and working them all together most workflows can be cut down by hours we just dropped a free guide that every marketer needs right now it's called the ai skill ladder this guide gets you past the prompting stage and into the building stage with seven agent tools of framework and a cheap cheat to take your ai usage to another level plus it's free get it right now scan the qr code or click the link in the description few of the main ones like man so man is autonomous it's an ai agent that's really designed to act like a digital employee so it execute complex multi step tasks all on its own sometimes it'll send off like multiple sub agents to go do a bunch of tasks all at once and then just come back with a finished result it's pretty amazing it's one of the easiest entries into agents you just kinda ask it to do whatever that might be in and it just figures out weaves together all the multi tools could be generating images videos copy powerpoint points like anything it just kinda figures out what tools it needs and guess the job done what are some of the you know workflows when you use manage you think it's just much much better than a claude or a chat or things like that yeah so compared to like cloud or chat it can just do a lot more and then as far as the more advanced agents like cloud code or na n those are just a lot harder to learn like man most people can just jump in put in a prompt and it'll get it done so that's like the easiest entry level into that so i just start one i have this where i've got a youtube video about the seven different ai agent tools and i wanted a pdf resource that goes along with that that'd be like a lead magnet just covers everything that's within that video so it'll have like list of use cases starter prompts for each of those and have a cover page and then also have our like logo and branding for the colors so i can drop in our logo and branding and then just send it off and just from that it should have everything it needs to do the job so it's gonna go out and actually wash that video right on youtube then it'll develop all the copy it needs to put into that pdf and they can actually design the whole thing on its own using those colors i uploaded with all the that unique branding this will take a few minutes so i can jump into some other ones that i have in here i think one of the use issues are showing is super cool so you kinda give it a youtube video to go you say is it watching the video is it going fetch in the transcript or actually just going to consume the video and then turning in what it's watched into a pdf yes a it'll utilize the transcript primarily but it i can actually watch the video and find key points in the video model probably images from the video to use it yeah gemini i actually goes and watches the videos if you've ever dropped those in into gemini which is the main thing i use youtube gemini is just for because it's so good with youtube yeah the youtube connection like i think one of things you're showing coming back to like why man over some of the other agents like chat to or claude i think you're showing one of the core reasons why because it can switch between models for whatever task is relevant to that model so like claude is not going to go off and do this because it's not gonna connect to gemini and start to correct pull all the from youtube whereas mana is like it can just pull in the model that is best suited to that task which i think is a real advantage for it yeah exactly and then it can yeah pull a nano banana to generate images it can yeah have this skill to generate pdfs which is its own other skill look can do full websites it can go off it can actually create like google sheet and add a bunch of stuff to that sheet for you like i've had to go off and like find leads and it'll just scour the internet find a bunch of different leads based off of your requirements then you'll actually create the whole sheet with all their contact information go to their profiles and do all of that for you like ten minutes later you just have the whole list done and i've used it just for research before too like while this is working well yeah it looks like it's extracted everything from the video and then so like i identify the tools and now it's gonna actually go and research those in addition to what was already in the video so yeah it just kinda comes up with its own plan along the way like do you give away pdfs for your audience or something like that yeah so we actually did one for this video okay you know it's free pdf yep we collect emails of course but yeah so you made like an offer as part of the video yeah and the way we usually do that we use lovable on the back end so once that's downloaded use create a quick landing page and lovable it use basically ask it to create it we have a little template but it'll create a landing page to have to email collection and just do it all the once you prompt for it whole things there then we just embed that onto the site very cool and then for a one that will show how comprehensive these types of things can look in man this is the one just where asset it to go and analyze that's when you're like making a a video on something and you can just ask you to go off and do all the research for you and then build it in this interactive way that's just a lot easier to look through and part of the prompts i have it to as it's going generate things that could i could potentially use for b roll as well so it's like doing all the research and i'm reading through it then i can go down to like oh this is actually this the rules roll right here that's the smart idea yeah so that's all i yeah built in and i can like click through all of this as i go learn stuff i might not have uncovered in my own research and then at the bottom it has like the whole b roll gallery of everything created that i might wanna end up using in the video i think it's a pretty great use case you can use a bunch of videos that have done well in your vertical and use those as inputs and then come up with a different angle or a differentiated take on that topic so you can use that it's great research too actually so i think there's a ton of interesting use cases in kinda give minus youtube urls inputs and then having it create a bunch of things around that the other one i love to do but it's just too token and hungry is like create vertical dashboards for performance of channels and so you can just input a channel and mana will create you a dynamic dashboard to show you the performance of that channel over time yeah and even this one like this research i have it it goes and reads a bunch of different subreddit finds youtube comments across all these different videos and isolate pain points from all of those and that's cool that's actually overlooked use cases so man can grab the comments yeah there is like what reddit is actually saying it has specific comments from reddit that's cool is cool yeah there's like content gaps like what didn't cover you're finding where the arbitrage and the market is for what to create and then you're also using the same tool to kind of repurpose that content once you've built it and what's really cool in man like i mentioned skills in claude so this i went back and forth is the first one i did where i you know got the original and then asked it to do some more stuff in it and then at the end i just say hey turn that into a skill and yeah to make the process we used here into a reusable skill with skill creator and then now i added that to skills and then anytime i wanna research a new topic i just click it from the skills and type in the topic and it goes off does this entire re report for me we're right back to today's show but first i wanna tell you about a podcast i love it's called nudge it's hosted by phil ag it's brought to you by the hubspot podcast network the audio destination for business professionals and to the uk's fastest growing business podcast what i love about it is that the nudge listeners love no fluff no bs evidence based marketing tactics they get in each episode you're gonna wanna listen because this is like an nba worth of insight in every single podcast and entrepreneurs you're gonna love the show because it's filled with repeatable proven studies not hearsay say not one off success stories marketers you're gonna love it because it discusses the psychology behind great marketing and what marketers are getting wrong listen to the nudge wherever you get your podcast and do you do the same kevin with like updating skills like if you do that and you don't get a result that you want you may be i you know ask some different questions or run some different prompts do then go back and update that skills because i think that's part of the challenge with skills it's like managing and them and keeping them updated it too yeah absolutely i mean anytime i just does something you don't want you gonna just ask it right in the chat update this part of it like i didn't like how you worded this especially if it's doing like anything writing related yeah yeah that sounded pretty ai it just changed this wording and yeah it'll just update the whole scale and not do that in the future what do you think is the core difference that you need to have moving into this layer versus the previous layer to kinda build and real agent workflows like this really most of it is just curiosity and the willingness to do it yeah none of this stuff is really that hard to learn and you could learn any of these tools in a weekend yeah it's just really going out and doing the thing you mentioned you had minus and then there was another tool you mentioned yeah so like this actually this whole pdf was already done that from that prompts that we sent off so i can download this and something we'll do is use lovable yeah you can just generate also sorts of apps landing pages anything like that just five coating in general there's just so much cool stuff you can do i this prompt tier to generate a landing page and then i can drop in that pdf that man just created and you can see this a pretty simple prompt might take you a submit level but your creating landing page to host that offer as a giveaway yes yeah so host the offer and then you know standard landing page where it'll give you an overview of what's within that offer and then just have a button you can link have a little modal pop up to enter your email and then it'll take you to it like thank you page and download it and how do you collect the email addresses you have this connected into your back end yeah so this prompt is gonna show the landing page and then yeah you have to input your like bi behave or hubspot api to actually get it to collect their and there you have to collect you know connect to whatever tools but level can do all of that is what we actually use yeah yeah yeah and just five put in general you can just create such incredible stuff so easily it actually just yeah blows my mind i show like a different one that i made i use google a ai i studio a lot partially because it's you can do a lot just for free yeah it's unlike most people you can just go in and create any sort of internal tool completely for free and like never pay for it when you're giving stuff out as resources then they give you three hundred dollars in free credits when you sign up and i've not even come close to that so that's right that's why i kinda recommend google ai studio if you're just dabbling this is one i did where talking about like how to reduce hallucinations when you're prompting and you had a bunch of different prompts as part of that and when you click on it it has each of the prompts and then sometimes there's a field you need to fill out and as you do that it will just fill it in in the box right below and then you just click copy copy prompt ready to go copy your clipboard and you can go use that and again is that same thing you can just yeah give stuff like that that out for free yeah that's so much easier to kinda just give away like coded apps not in the way that we use to give way coded content you can even get more advanced on this you about a full piece of software just as a lead magnet yeah software code is essentially becoming disposable like you know you can give something away don't have to maintain it don't have to do all of the old software work that you have to do it's pretty incredible it's a good reminder of like the show don't tell like if you can give somebody an app and show them what the value you can deliver versus like telling them in a pdf for a slide deck you're gonna be at a better point of advantage i'm actually just constantly amazed like what you can actually build so i had a video where i wanted to compare all the different ai video tools or was like nine different tools and i wanted to do it like as a to tier list but i wanted to be able to actually play the videos live as i was doing all of that which there's just no software that it exists that's through so i would had to do that like in premiere pro after the fact factory it wouldn't have been live i was like at this time i'm just like not gonna be able to do like wait a second let's just ask google ai studio and see if it works and it literally just created this whole app actually yeah and when i actually did this in the video this is like a full like infinite canvas i had like a hundred fifty videos playing all at once on here and each time you just click batch media then you can drag a video into each of these boxes then it like pops them up on the side that's super cool yeah that would like they would just pop down onto the canvas then you can like just drag everything around and then this one you would just like spawn logos one of those things i was like there's there's no way can do this right and even if it does that has to break by the end of this video when i'm comparing this across like fifteen different prompt this was like zoomed out all the way full hundred fifty playing all at once and like nothing broke just the fact that you can create a tool like that just out of nowhere is yeah those it's super no matter how many times i do it and and i think that's the right example of like how to think about it you're going from interacting with these kind of assistance and augmenting largely text output for a lot of people some multimedia output to go building like full blown web applications or doing multi step workflows using skills multi model agents like mana and then you can just like have an idea of like hey i wanna visualize and show this thing and build a custom way to do that in a way it just wasn't previously possible to me it just completely changes the storytelling and how you actually bring stories to life which was a a big takeaway from the conversation today kieran anything for kevin before we sign off today i think the man use case you showed was great so like if you watching this and you wanted to kinda move into layer four i think using man as a kind of research tool to go look at these youtube videos and then you kind of layered in hey like go look at the video go research these other places like reddit at all of these other places and that topic they covered what are other things people are saying about that topic what are other things that are trending about that topic and tell me how i could create something about that in a really differentiated way and i'm sure it could actually create a pretty good script for you as well and so i think that is a good first use case to go from like the kinda of layer three into the layer four their workflow weaver and maybe go try out minus and start to use it as a great content research tool yeah definitely and you know everyone can kinda customize that to exactly what their use case is like once you see the power of it pretty much anything you wanna do you just add in a little bit of context and we'll just go off and get it done for you cool awesome well kevin thank you so much for joining us today we appreciate you and we'll see everybody next time i'm marketing against the green this data is wrong every freaking time have you heard of hubspot hubspot is a crm platform where everything is fully integrated well i can see the client's whole history recalls support tickets emails and here's a test from three days ago i totally missed hotspot hubspot grow ahead him

okay we all know building ads is a critical part of marketing rep four just released a brand new ad creation feature and i'm taking it for a test drive on today's show it does some things really well and it does some other things that think it really needs to improve i'm going to walk you through my secret step in claw to make sure that you save yourself a bunch of time and money before you actually go into rep and build the ads that i'm gonna show you how to iterate and work on those ads in rep to get the best results for the product that you're trying to promote let's get show every sales rep knows the feeling pipeline looks full activity looks good but you're not sure which leads matter which ones are wasting your week and which call you should have made three days ago hubspot prospecting agent gives you a clear path forward it shows you who's actually ready gives you the details your reps need and points them towards the conversations worth of having so reps know exactly where to focus and what they're walking into check out hubspot dot com to learn more about hubspot prospecting agent on today's show we're talking about something that's getting more and more important and that's ads has tools like cloud code and rep and lovable have become really prevalent and it's easier than ever to like build an app or a product and get it out in the world to actually test validate and grow that you need to advertise and so how's is ai doing it actually generating ads we've had a few shows in the past but there's a brand new tool from rep that i wanted to test live with you on the show and so today we're gonna dive into what the product is we're then going to actually generate some killer prompts in claw to actually put into rep to try to build an awesome at okay so literally hot off the presses here and this is a fun promote tweet from rep sometimes you don't have a marketing team or the money and this is what i'm saying sometimes if you're just testing out an idea whether you work at a larger company or you're a solo entrepreneur you just wanna like get an idea out there and validate it you know karen and i write in this book we tested titles with ads to kinda understand what title might resonate best with our audience and so the easier you can create and iterate on ads the faster you can learn so rep has introduced this let's watch the video real quick for a second here alright what if you can make scroll stopping ads rep is making ad creation fast and easy okay high end luxury fragrance brand adds built your ads ready in minutes your entire ad pipeline one prompt oh so the value prop they're talking about here is you go from a small number of ads to a lot of ads which i do think is interesting but in this case sometime it's going to be just going from no ads to en ad or a couple of ads so the value prop here is simple they're trying to reduce the friction of creating ads the question is are they really reducing the friction are the ads actually good and would we run them that is a big big question the first thing i'm gonna wanna do though so i'm gonna wanna make sure that my prompt my input to create those ads it's actually good and normally i really like to prompt in cloth you can prompt wherever you want you could write your own prompt to put it directly into rep but i find that claude opus four point six is a really good model for prompting it helps things things through with me so let's give it a try i'm using willow voice as my voice to text so that you can hear what the prompt is versus me typing it awkward and so here's my prompt hi yes i want your help building an amazing prompt for rep for new ad making skill and feature this prompt should be to create three ads each for linkedin instagram and google adwords four hubspot customer agent product i want you to go and do deep research on that customer agent product understand the core selling benefits understand how you should position it understand the best practices for advertising on those three platforms that i gave you and come back to me with a deep quality prompt as if you were an ads master a leading marketer who spent all day doing ads and give me that full prompt that i can go and use in rep and willow then takes that and turns that into text and i'm just going to run that in cloud and see what claude comes back with we just dropped a free ai ad prompt kit inside you get a complete worked prompt so you know exactly what good looks like an adaptation guide to make it your own platform specific creative direction for linkedin instagram and google and a troubleshooting guide it's built around rep add creation tool you can go from a blank page to ads that are actually worth running in minutes get it right now click the link in the description i find to really do generative ai projects so whether that's images that's video things that are broader than just plain text getting that prompt right up upfront is hugely critical i want to underline that it's really important to take this step of working to refine and iterate your prompt before you go and actually create the assets because it takes a while and it takes a decent bit of credits and usage that you're paying for to go and do this i'm on the twenty dollar a month rep plan i'll burn through that pretty quick if i'm trying to just iterate on these ads over and over again in rep with bad prompts so i'm gonna save a lot of time and i'm gonna save a lot of money if i iterate on the prompt and text before i actually go and create the ads themselves the other thing that i find really good about ai it's really good at helping you go from zero to one meaning you might not know that much about advertising you might just have an idea you have an app you wanna test you just have something you wanna get feedback on you wanna spend a hundred bucks and ads to go and do that that's all totally fine that's why this prompt step again is so important i could habit it go and research those advertising best practices and i might not be an advertising expert but it can give me a good baseline and then i can iterate on them based on what i see in the actual ads themselves so that being said it is given me a prompt it's a massive prompt you are an elite performance creative strategist who manages fifty million dollars of b b saas ad spend across linkedin instagram and google ads promoting hubspot breeze customer agent it gives full context on the product competitive positioning brand voice guidelines it's giving you the type of ads to create and it's actually giving you the ad angles and so what you wanna do here and this is like the really important step that nobody does you wanna go through and look at each of these actual ad concepts they had a perfect world we weren't doing this on the show i'd probably spend a good thirty minutes just going through all of this and so for linkedin it's saying hey we wanna lead with an angle for ad one of fleet with a jaw dropping customer result that makes the buyer stop and think wait really use one specific stat from a named customer hubspot aggregate data pain point problem agitation okay let's say i wanna do something different i can just highlight this and say for this instead i want to make this complete screen stopper i want it to look very anti linked in and more homemade update this part of the prompt with that direction okay and so what you wanna do is exactly what i'm doing here which is pick the parts that you don't like and you know it's ai so it's gonna give you the over hyped pure genius you did a great job love that angle but it's like look this is the taste you have and one of the things i know about linkedin is that linkedin can sometimes feel a little boring feel a little slow feel a little too polished so what you see here it added some critical creative direction add musk a complete scroll stopper basically takes what i said and gave the instructions and you would go through and do this with each one of those ads and what's really great is that it's giving best practices and skills and quality standards all really good and again like i said you would go through section by section and do some edits let's say that we've had the time to do that then we're gonna go over to rep and one of things you're gonna wanna do is use a skill because when they say i've got a new feature it often means they've added a new skill to do work for a very specific task and you see a few marketing skills that it is put together here one of them is ag creative which is what we're talking about here and so you have clicked on that skill so it has the full context in addition to my prompt and i've now pasted it in my prompt from claude i've picked the ad creation skill so it knows that this is what we're working on and has a unique skill there and i'm just gonna run this guy and now we're going we're off to the races you're now on economy mode we've automatically switched you to our new economy which is up to three times more cost effective great one of the things with these tools is it does cost real money if you you're iterating on these again which is why you iterate on the prompts prompts are much cheaper iterating on the ads themselves through ai is expensive sometimes you might get some ads that are close and then go manipulate them at canvas or your design tool of choice and so why it's thinking here what is rep rep is what's called a vibe coding tool it's a tool that allows you to take an english language or whatever your native language is natural language and input that and it's going to actually take that think about what you're trying to do and write code computer programming code to actually deliver the outcome you want and sometimes that output is research sometimes it's a website in this case it is ads and what's really interesting about rep their new version rep four has a lot more design and creative elements to it and you're seeing this happen right now with this rep canvas feature where it's actually showing you that you can look at and manage and edit things here in the canvas tool what it's trying to do is merge instructing it in your natural language as well as being able to do what's called a graphical user interface where you can go and manipulate and move around and do spot edits and everything as part of that and so rep lovable base forty four there are a lot of companies out there whose whole goal is to help you go from an idea to build something very quickly in this case we're seeing how they do it at i've used several ai ads tools super scale etcetera i'm excited to see how rep does now that it's dot rep for with this new ads skills feature before we continue let me tell you about a podcast i love marketing school marketing legends neil patel and eric sue bring you daily actionable digital marketing lessons learned from years in the trenches marketing school delivers bite sized marketing wisdom you can implement immediately whether you have a new website or you're an established business you'll learn the latest in seo content marketing social media email marketing conversion optimization and general online marketing strategies that work today they just did a great episode called ten seo lessons that still work in the ai era listen to marketing school wherever you get your podcast so while it's building on our ads let's check back in at what we've done so far so far we've looked at rep new ad feature then we went into claude and we worked on creating an awesome prompt for our ads and claude went researched and wrote out our prompt we then made some edits and gave it some creative direction to make our ads even better and what was really great is we used willow voice to talk to claude to make that process really fast then we took our prompt we came over to rep we did a really important step we turned on to use the ad creative skill so it knew exactly what task we were doing and then we put in our prompt now it's building our ads assets and we're gonna look through them to understand what it well what it didn't do well and we'll talk about how we would iterate and make them better and i wanna remind everybody we had to do a pretty complex task which we wanted to create three different types of ad forward three different ad networks and again all these ads are for the same exact product hubspot customer agent and claude did the research on that product and told rep where to focus what the key benefits and background of that product are if you're gonna do great ads you have to deeply understand the product so that it understands what the value proposition to the audience could be so we've got this linkedin in net seventy seven percent fewer tickets zero to hires nut customer result i like the data of the graphic drives me insane that is a terrible terrible image right in theory the ad concept is good in terms of copywriting the creative is not and what i find with these ai tools is that the imagery and the creative and getting that right you really have to iterate on and it's not just a one and done where these tools have gotten very good at the copywriting side the creative side this is like a state of the art application is still not quite there like for example the next ad first of all this isn't the hubspot logo you heard me tell claude hey i need you to use the hubspot style guide and so clearly hubspot a sprocket not a steering wheel so well that's a whole thing thirty nine percent faster resolution breeze customer agent okay by the way i think this is what it was trying to get with scrolls stopping with this like real life whiteboard but the problem is it still made it look like a very commercial ad and it faded in if it had taken a different design approach this ad could have worked this sad not good i wouldn't run it it's pretty bad if your mental model of an ai chatbot bot is still press one for billing then you need to see what's changed you have the classic text on image issues let's give this a rev this ad is really bad the image has nothing to do with the product the copy isn't as strong as the other two you have an image overlap over the word support that makes it eligible the crm context and value propositions are in this weird blue text and like just hanging out at the bottom this is one of the worst ads i've ever seen please redo it and i'm not trying to be is it's just a bad at so again what i do like is that you can on an ad or different components you can work on iterating it there again this is not a hubspot logo responsive search ads now that's interesting this because it's google and it doesn't require all the visual imagery is gonna be a lot easier because it's more of a copy writing problem and this is actually really good i like the fact that it's not just giving me like plain text copy it designed what it would actually look like how it may appear in google search and i find that especially when you're doing these type of search ads getting that visualization is really helpful especially if you're early on and haven't done a lot of search advertising and this is a very different thing because it's copy only hubspot customer agent resolve sixty five percent of your tickets automatically set up in minutes not months ai support agent that works across chat email whatsapp and voice thirty nine percent faster resolution no code required start for free sign that's good copy it's very much like if you match that with an intent somebody who's actually looking for this type of agent you're gonna be really successful and that is much better and what's great it's giving me different headline options to test and it scored them right the winning is too many tickets too few staff teams support six five percent auto resolved and so that's like a very cool iteration and i love that it built be that in this web page i could see it all together and ai is really good at scoring and grading and helping you iterate so whatever search ad headlines i picked i would want to kinda continue to say like hey i want the best grade possible help me work and iterate through those and give it creative direction so it also didn't do any instagram ads one of the things that's nice is if you select to give an element you can see in this text editor that you know it still has that ad that i'm trying to fix selected so what i would do is say you forgot the instagram ads can you do those for me so redesigned our linkedin ad oh and now it's got these instagram ads over here this canvas feature is kinda weird so if we wanna zoom here this redone ad is a lot better not a chatbot your ai support teammate there's is a much better message much better copy the data proof points are down here the logo is still bad this is what i'm saying i think we could get to a really good linkedin ad here it's also just like a bad image i need to give it an image i need to give it a copy of the hubspot logo i probably need to do ten to twenty versions of this ad and rev through it before i get something really good and this is one thing i'll tell you there's a lot of people out there that are like oh you could do this so easy it's five minutes it's free no you're gonna have to iterate on this and even iterating on it you're probably gonna run through a good amount of credits to iterate on this might cost you twenty thirty forty bucks in credits just for this ad right and even on economy mode depending on how much you have to iterate on it so you can get there these tools are really great especially if you don't have any design skills or anything like that really really opens things up if you're a really good designer they might be slower than actually just rip it and doing some of this stuff in canvas so that is where we're at and like the current state of the models and the tools we got some cool instagram ads here the before and after i like that it's very instagram and i love to do these weird like post apocalyptic illustrations one of the things that i should have done that i would advise you all to do is give it a not do list like don't do these style of illustrations don't use any colors that are not on the brand style guide like this weird blue like where is this weird blue coming from and that will help you get better first rev and so what's my take on the new ad skill feature in rep for it's a mixed bag is it gonna get you good ads eventually it is but you're gonna need to iterate and work on that creative you're not going to get the perfect outcomes one shot unless you're giving it a ton of existing sample creative to build off of and you're really continuing to iterate your prompt what i suspect is gonna happen is to get a set of ads like these nine ads to get really good probably gonna take you a couple hours now once you have them and you test them and you have some advertising data the next batch is probably going to happen much slower and as you know karen and i talk about loop marketing as you run the loop from express to tailor to amplify to evolve the evolve is like how you actually go and think through the creative take the learnings and make it better the next time it will get faster and better every time but if you're just like hey i'm trying to get some ads up for this project i'm doing expect to spend a few hours to get ads that are really really good that would be at the level of standard that i would want them to be so shout out to the rep team for innovating for making ad creation easier and easier i find it to be similar and in line with other tools like super scale and canvas and base forty four and others that do a similar thing so i think a lot of it's gonna depend on where you prefer to work and how you feel about like this canvas feature and some of the what i'd call user experience and like subscription software features around the tools themselves but i'd love to hear from you drop a comment as to the ads you're building what you're building them and what's working what's not working please hit like and subscribe remember that we have done everything you need to do to get ads created and we've run that through from start to finish we started in claude we built out an amazing prompt we added that prompt we put that prompt in rep and then we showed you how to iterate on those ads in rep to keep making them better better till their something that you would want to pay money to run that was a lot of fun i really appreciate you joining us on this episode of marketing grain we'll see you next time hey everyone you know karen and i have been doing the podcast for a while now we've been at this for a couple years we love it we could not be happier to be doing this but we wanted to take things to the next level we wanna level up the impact we're having with marketing against the grain so the next step of our journey is something we're really really excited about we're gonna launch the marking it's the grain newsletter and marketing against the grain newsletter is going to be amazing if you are a marketing leader practitioner you're in the trenches doing in marketing every day this is for you we're gonna deliver right to your email inbox and you're gonna get all the behind the scenes frameworks practices tutorials from us from guest we have on the show and from people beyond the podcast that we think are gonna be helpful and really have an impact on your day to day week to week doing marketing you're going to love it it is something we've been talking about for a while we're really excited to have it out in the world we've already got a thousand marketers who are on this newsletter please join it's completely free we'd love to have you as part of the marketing against the grain community and it's easy you can click the link in the description below or you can head to marketing against the grain dot com slash subscribe

hey everyone i'm pumped we've got matt wolf back on marketing against the grain he is a friend of the pod k and i have done several episodes with matt and he's helping us by diving into something that's really cool google released a new design tool for vibe design called stitch and really what's cool about it is let you go from prompt to design concept for landing pages campaigns everything that you might be building and doing in your marketing and it's really especially good if you don't have access to a designer developer to do these polished mock ups and designs for you can really act as that for your product or for your company and matt's gonna walk us through exactly step by step how to do it let's get into the show here's a quick word from hubspot hubspot helped tumblr solve a big problem they needed to move fast to produce trending content but their marketing team was stuck waiting on engineers to code every single email campaign now they use hubspot customer platform to email real time trending content to millions of users in just seconds the impact three times more engagement double the content creation wanna move faster like tumblr visit hubspot dot com alright so on today's episode i'm gonna show you a really cool new tool i'm gonna show you how to use ai studio from google as well as stitch from google to create an awesome website it's gonna design the website for you and then build the website for you and if you're a business or marketer and you need landing pages or sales pages or checkout out pages this tool and this sort of process i'm about to show you can do all of that for you it is really really cool let's check it out so this is what google's calling vibe design right obviously playing off the vibe coding trend where you just talk to an ai coding agent and it will code up software for you well apparently this is that but for designing for designing websites for designing web apps things like that i did play around with it i had it redesigned the future tools website for me and was actually like pretty blown away by how well it designed it this is a tool called stitch from google and if you take a look at what stitch looks like it pretty much looks like fig it's google's version of fig let's just call it what it is so it's like this giant canvas tool where you can drag your own images in you can do your own designs and your own design tool and then bring those designs in but now you can actually just prompt your own designs into existence this is obviously great for businesses marketers things like that you know coming from the digital marketing world this is probably gonna be the new way you build landing pages right we used to use the click funnels and the optimize presses and the lead pages of the world and it's like all the stuff we used to do we would spend a week on a landing page right and now it's like i can have that same kind of thing done but probably even a better and more modern design and like a minute it's so wild when we used to do split testing with click funnels and lead pages and all those kinds of tools you built a new page basically from scratch every time now it's like i literally can speak with my voice and say create me a split test version with a couple different headlines and it'll just be like alright here you go this is the one that i made with a demo on my own youtube channel right i basically initially prompted it designer a website that helps curate and filter all the latest ai tools the same way future tools dot io does so i just gave it my future tools website essentially you could see that it automatically pulled in the website here says i've analyzed future tools create a synthetic dark design it actually figured out like the color schemes and the fonts and everything just by looking at the website so it built like a whole styles sheet essentially leaf for the existing website with the colors and fonts and all the things right and then after it did that it went okay here's your new site design and this one's a fairly close replication of what my existing site looks like it's not too far off so it started by designing the exact kind of replica and then i went and said alright generate three variations of the home slash discovery screen and it created this bold dark version this warm energetic version and this cool professional version but each one is like a completely different layout different color scheme different design and all really good modern professional looking designs but like if we wanna continue to iterate off of this one i can select this one and i could press this button here and just like talk to it so let's say hey what are we designing on this home discovery cool professional screen today can you give me two more variations but change the hero image up on the top right on both of those variations on it she's on it mh great i've created two new variations with fresh hero images take a look i could see down here that it's still generating them so she got ahead of herself with saying that they're ready while they're still generating but here we go now we can see him populating yeah so they changed the hero image up on the top and left everything else the same i mean i actually think the original one looks better but you know it gave me what i asked for i can split test different headlines give me two more variations but change the headline on the top i'd like to split test between them working on it yeah and we got two versions with two new variations of text at the top matt just dropped his vibe design guide and he's giving it to you for free it shows you how to build a high converting landing page in under an hour using google stitch you get his exact prompts the pitfalls he hit and the full workflow from first prompt to publish page get it right now click the link in the description let's try to create something new here let's do another web app what i was thinking is some sort of dashboard that compares the latest large language models let's build out a stats dashboard that compares the top ten large language models that are available right now for text generation make sure you pull actual data from the internet to find out what the best models are for this dashboard ready to dive in what's collision it's working on it but it seems like maybe there's a little bit of like bugging in the text like it sort of responds too quick or sometimes starts on the thing but doesn't respond but we can see i'm moving forward with a sleek electric tech design system deep charcoal backgrounds dashboard blah blah blah okay so it just came up with our color scheme here now it's generating the screen and the real double check is did it actually ground the information in the screenshot from real web data i just love that it just like comes up with a styles sheet when i was trying to do my own designs with like say click funnels or whatever i was always bad at like color schemes i don't know what colors go good with other colors alright so here we go here's our dashboard it's called mer core so here's the stats dashboard i mean it's not up to date as of like now because it's showing gp five omni so here's our web preview version and it's not totally up to date as of right now today i can tell you that much because it's showing gp five omni and cloud four on it obviously on five point six on it gp five point four we're on gemini three point one so i mean like the data is definitely not up to date so it's pulling from its existing training data so it didn't look like it grounded from the internet we can see this is all from twenty twenty four which is probably when the sort of pre training run was done on this model yeah it doesn't appear to be able to pull information directly from the web to build out your dashboard but we did get a dashboard design so the links don't seem to really work i think you have to go and design each page individually so let's see if i actually give it a website to find the data from will it pull this data into our dashboard so this looks more accurate i mean it's got three point five on it it's got gp four o twenty twenty four so it didn't seem to be able to pull the data from the website and i don't think there's an api here either i'm not seeing any way to like export this i mean i can obviously like copy and paste but so yeah it's not able to seemingly like actively pull data from the web which is interesting that it was able to sort of like match the color scheme of my future tools website by looking at it but it can't actually get data from this page by looking at it i was saying this in my news video you can actually probably have your open claw make designs for you now right like you can say hey go to stitch and design me up a new website for this your open clock could come to stitch give it the prompt to design something throw it into ai studio for you code it up into ai studio for you and then come back to you and be like hey here's your live website i made it for you let's go ahead and actually export it to ai studio real quick because ai studio just got an overhaul this week as well let's go to build with ai studio it pulled in all of these details our image it pulled in an html it pulled down a markdown file it looks a little bit more like an ide you know like a visual studio code or a cursor or that kind of thing so just very very simple integrated development environment when you're using something like visual studio code there's like terminals like you can open up and run a terminal you can install extensions you know like wordpress plugins but for code there's just a lot more that you can do in it where this is just like trying to keep it simple for you so here's our coded up version on the right it seems like it made a few like slight design tweaks so it added some like little extra details but stitch is clearly designed to be your design phase it will do a very very basic basic basic like html code behind the scenes for you so that you could click this preview and see what it looks like but you wanna throw it into something like ai studio to like start dialing in the functionality so it's generating a predictive heat map people pay good money to get predictive heat maps now you can just do it directly inside of this tool so look at that it created a heat map of where it thinks people will most likely click i mean how many websites have google analytics on it where google can sort of understand where people tend to click on websites what it happens if i do instant prototype okay create with ai describe a new screen to add to your prototype build out the models page and then once you prototype the whole thing you could probably dump the entire thing over to ai studio i mean i'm fairly hard to impress these days when it comes to ai stuff i feel like i'm kind of des i'm like alright new ai models is cool new video models cool yeah that's nice new large language models yeah looks a little bit smarter that's good that's good this was one of the things where i'm like i'm really impressed by this hey there we go although interestingly so it kinda tweaked the design a little bit well you know still got some limitations it's still pretty impressive that you can just like make a design like this right click export over to ai studio and have it just like a coded working prototype and quite honestly if this was me i would probably just build the initial design over in stitch and then build out the rest here because like if i was just to give it a prompt of like now build out the models page on the ai coding dashboard i think it would do it a lot better from the coding standpoint because it's not trying to redesign anything it's trying to just build another page on top of the existing scaffolding that had already built alright we'll test this this will be our last test here before moving on it's working because it says it's working alright so i i've built out the models page for the dashboard so let's see if i click on models boom models that's how i would typically do it is you get the initial design in stitch once you have this initial design in stitch export over to ai studio and then once you're an ai studio this is where you start telling it okay now we've got this page now let's build out the compare page let's build up the history page it will keep all of the existing sort of layout and aesthetics and everything and just build the next page for you let's see where does it actually publish what does publishing look like your chat history code will stay private your app will be accessible via a public url get started select a cloud project okay you have to have like a google cloud account but i'm not gonna go through the whole process because i don't actually want a published version of this but it would publish it on google's cloud platform for you you technically don't even need to connect it to a domain name it would just be like a long weird url to start and then when you're ready to connect it to a domain name you would do that from inside of google cloud a quick recap here so you give it an idea for a design once you have the design you like you right click on it you export it over into your ai studio once you've exported it ai studio you could pretty much continually just tell it to build out the next page build out the next page build out the next page there are other platforms out there you've got like lovable and v zero in some of those tools that do similar stuff but from what i can tell you can just go into google's stitch and do this i don't think there's any cost involved at least not yet maybe there's some like limits that you'll hit if you try to use too much but for what i can tell you can just go in and do things just being totally real totally be honest keep it in a hundred as the kids say i mean this is one of the more impressive things that i played with later hey everyone you know karen and i have been doing the podcast for a while now we've been at this for a couple years we love it we could not be happier to be doing this but we wanted to take things to the next level we wanna level up the impact we're having with marking against the grain so the next step of our journey is something we're really really excited about we're gonna launch the marking against the grain newsletter and marketing against the grain newsletter is going to be amazing if you are a marketing leader practitioner you're in the trenches do in marketing every day this is for you we're gonna deliver right to your email inbox and you're gonna get all the behind the scenes frameworks practices tutorials from us from guest we have on the show and from people even beyond the podcast that we think are gonna be helpful and really have an impact on your day to day week week doing marketing you're going to love it it is something we've been talking about for a while we're really excited to have it out in the world we've already got a hundred thousand marketers who are on this newsletter please join it's completely free we'd love to have you as part of the marketing against the grain community and it's easy you can click the link in the description below or you can head to marketing against the grain dot com slash subscribe

so today we are going to tell you step by step how to build an agent in complexity computer but not just any agent one that helps you work with creators which is a huge marketing trend right now and most businesses should be partnering working with creators so we're gonna merge creator best practices with building a really cool agent we're gonna do all of that complexity computer let's get to today's show okay on this episode we're actually going to live build a tool that most marketers would benefit from i think we are we we'll see how far we get because we're doing this live with no prop we're doing it a lot zoom live we're gonna compare and contrast per complexity computer which kip us super excited to buy cloud code which both of us are super excited by we're gonna each take one of these tools and we are going to build a creator hub i think this is a channel that has become an ever more present and important in b which is how do you distribute your product or services through creators there's some good tooling for that but we're gonna show you how you could potentially build a lightweight creator tool to help you do this for your business or service and at the end hopefully have something that we can actually share alongside the episode so kip you ready for some live building let's do this let's do some building okay so you're gonna take cloud code we're gonna put the same prompt in both right so i think here you're about to work on a prompt and you'll share with everybody watching kind of what we think the core steps of building a creator happen we say creator have what we mean is a tool that's gonna let you identify creators to partner with do the outreach probably propose like the economics and rough contract details so that you could actually sponsor different creators in your industry right and kieran you said before we actually started recording you're like hey i've kinda have an unusual workflow but i think it's really important that you share your workflow with people so why don't you tell people kind of like how you approach a problem like this so one of the ways that i would do this is everything for me starts with claude even when i'm using other tool so when i'm building in cloud code i weirdly use claude opus to kinda go back get work with and then i prompt cloud code i do think it saves you tokens from what i understand it's like cheaper to go claude versus clock we're in the era of token maxing you know ever trying to spend all the tokens you know well if you are a customer of mana which i am you can spend an infinite amount of the tokens they are a pricey tool all of the multi model tools pre computer you use a lot of tokens steal taxi alright so we're gonna just live so i always get like kind of embarrassed by my prompt and skills when we're doing live because it takes me a few iterations to go back and forth the cloud are you there yeah i'm here sorry sounds like something's not working the way you expected what isn't doing what give me a bit more context than health all accent it's i mean alright claude you are gonna act as my cto for this tool we want to build so you're a elite level cto you know how to build lightweight tools for non technical users and make them incredibly simplistic to use using best practices from the best point solution tools in terms of ux and time to value for this tool we are going to build a creator hub for market tiers and that creator hub is going to have the following functionality it's going to be a creator hub the marketer can come to the creator hub they can enter their domain once they have entered domain the tool will research that company the product and services and it will establish who the audience is and who the buyer of those products and services are once it has done that it will then do another piece of research to figure out what platforms are most relevant to that buyer that could be reddit twitter linkedin it could be a range of different platforms and will stock rank them in order of importance where those buyers go to consume information once it has done that it will go and find the most notable creators for those platforms people who speak about topics the buyer is interested in relevant to that company's product or service that have a real audience and real distribution leverage with that buyer it will show you the audience size they have for the different platforms relevant to that buyer claude can you start by turn in this conversation into a functional spec that we can get aligned on and break it into step by step parts that we can then individually build sequentially we just dropped a step by step guide on how to build a custom marketing agent in per complexity computer it documents the exact five step process we used you'll get the functional spec framework the context files that make your output actually useful for your business and the prompts we used to build a working creator research tool and it's all free get it right now click the link description right brilliant so you want me to architect this as a proper functional spec with phased build out let me get this straight before i write it up when you say the tool researches is the company are you imagining it pulls from your own knowledge base or web research in real time and on the creator discovery piece are you thinking it services creators algorithm academically or is there a manual curation layer involved it will use web research user or web research tool and it will automatically show creators there is no manual creation layer although the user can delete creators it does not feel our applicable got it so fully automated discovery with user override to remove irrelevant ones that's clean let me redirect you here since this needs a proper structured functional spec with sequential phases detailed api requirements and technical architecture decisions this really belongs in the text interface where i can lay it out with proper sections phasing and clarity i can absolutely do this but you'll want to see it formatted properly rather than me trying to speak through a multi part technical spec shall i pivot you over there or did you want me to give you a quick verbal summary of how i'd break the phases down first alright claude you can pivot over to text base please his accent is like a bad guy ritchie movie it's pretty incredible i have my family yet have you pivoted yet called oh it's he's working what's happening here claude i think i just have to pivot it you just type in pivot to text it's that's wild hud the light use voice you're a big voice user i love voice but and i love ai all these ai tools are still really really buggy i don't know like the amount of bugs and stuff on the edges like we're wondering why there's not mass ai adoption that's certainly one of the reasons if you're watching the show this is why willow voice or whisper flow these dedicated ai voice dictation tools are really awesome they're super cheap i use willow voice all the time and cannot live without it i actually i'm gonna start using whisper but i might use willow because you're an investor so i am i'm sorry i i'm a big willow of voice fan shout out allan team they're great alright so we're doing this live i will say when i say live like we i just came up the functions back as i was talking about it while it's creating the file maybe break down like your thought process from that prompting journey because you had a bunch of interesting little tidbits there that i think most people wouldn't have said like you told you're cto you told it to break it out in phases maybe give some of your like logic behind how you gave it instructions yeah i work back to the front i think one of them best things you can do when working with these models is to first go back and forth until you align on something so whether you're create a post whether you wanna create a product position in whether you wanna do anything go back and forth and say can we align on how we do this and then actually move on to execute that a task a lot of people just jump in and start to do the thing but actually it's really good if you go back and forth in the line with it i find that that makes a very big difference the other thing is yeah i it a role this is just role based persona some of these are just habits that i have and i'm not sure with the later models if you have to tell it explicitly to act in a certain way anymore if they're just good enough to say well i need to be a pretty great cto to put it this together i try to give it some parameters i don't think i did the best job there but like trying to use best in class ux principles to have short time to value to have simplicity within the app and then the functionality again i think you wanna break out have it line on the step by step sequential way that we're billed this and we probably might wanna just shot it i actually think for fun we should launch shot it like oh definitely one shot it again i've old habit where i will just go in parts because when the models were not that good i built an entire app this time last year with claude and lovable and really it would go aw if you give it too many things to do but i just think the models are so much better now that we could probably try to one shot this so it's back with it's functional back sweet let's see what's it's done here you see the new cloud function i love this the new cloud functionality where it builds with line it doesn't even need the rf artifact anymore start to build imagery in line all this kind of stuff in line it's just so cool yeah and you also saw that they just released pew use when we're recording this do a show computer use they have dispatch which allows you to control it from your phone well one of the tools we're gonna build it in and is per complexity computer and one of the things i love so much about complexity computer is that it has a mobile app at a desktop app and they sync perfectly and can access everything in the cloud and it just like works in both environments where so much of ai is still very desktop heavy yeah exactly i'm just looking here got a real spec here screen come on now claude automated company research by we scrape blah blah blah one to two weeks what i can take one two weeks come on come on alright let's see one go this is gonna be a pretty wild tool if we manage to build this so it's got the company intelligence and buyer persona yeah so it's gonna build that research into persona it's gonna look at platform relevance i think it's gonna to have trouble with linkedin perplexed we might do okay linkedin i think i'm curious cloud code one of the ways we will have to do this slightly differently is you're gonna need to set up api keys so that might be a big pain the ads so maybe we just make a judgment call here and just do perplexed that's computer and then rebuild and cloud code because we will have to set up api keys and that's gonna be a little bit of time because you're gonna have set up x api we can always export the code from per complexity computer and have cloud code yeah ref it and build it so it's gonna do a creator discovery automated create research and rank from one platform e linkedin it's gonna search creators discuss some relevant topics score by relevance blah blah camp them and then show creator discovery extend to all platforms so it's gonna do single platform then multi platform then it's gonna do a deep dive rich creator profiles i like that actually polish and optimize success metrics persona accuracy create relevance low time user mps tech stack okay and we want to build this as an mvp using per complexity computer i think one of the things that i find so interesting about kira use of ai he use a psychological warfare against the ai do you so you do it all the time it's a very interesting approach i i don't know what's gonna care that i need my bonus you need the shed up you should've have been like my daughter needs new shoe you should like really guilty today day i was gonna say i need a dope watch then wouldn't cared about that obviously you need to tell the platform you're building for that was a bad prompt i could've just said to start with go research per computer and then craft a prompt for that tool i should give it that to begin with but let's see what it comes back with what i hope is interesting about this is that what we're trying to just is take you behind the scenes of like the steps involved to build something if you're not like a traditional engineer right and it's not just like you go in and you write a few sentences and you get the thing you want right there is still like a real process and a process of iteration that happens to get to an outcome that is good enough for what you kinda had in mind and was in your head yeah i think this is can you get to mvp of a pretty useful tool in a podcast episode so like if we can do this in under forty minutes but to really bill something that's tailored for your needs does take some amount of time so one of the things we did here was we asked it to figure out how to make this better so it's suggested how to made the app better and then you actually ask it to research the app you're gonna build an in which we said is per complexity computer and then it will give you some gotcha around building this in per complexity computer and in how should condom now we're gonna truly vie build us because we're under time pressure and when they say yes but if you're were really building this karen it would take you like a week oh yeah we're gonna be the honest ai folks and tell you this is a mvp show you how to get there hopefully you'll learn some things but this is not the x like i want shot at something and look it's changed the world you really do have to like work on this stuff and go back and forth so this is really an mvp but to your point here and i think that's a very valuable step i was in san francisco last week spent time with a bunch of companies much startups the clear cultural shift that's happening with most of these companies is that they are just prototyping and mvp stuff very quickly so that everyone has something to look at to say like is this actually what we want exactly how and if so then we'll go invest the time in building it versus like it used to be you'd have a memo where you throw around some ideas and like this is what we're doing essentially the first step is like hey this is what it would look like and in theory what it would do and if we like all that then great let's go and build it yeah we are doing the real show don't tell so we would be saying hey we're at mark tier we think creators are a great stream of distribution to add into our marketing mix and to do that we think we need a little custom tool that allows us to manage all of this stuff within a create hub and we are going to now get our prompt i think for per complexity computer hopefully this time it comes back i do think the bug we ran into was because it was using the legacy model all opus all the time as much opus as you can get that's my life motto all i did use a chat this morning for the first time because i had something in how'd that go well it has something in its memory that i wanted to like recap on yeah but they need to get out of the it's like it was a really mundane personal thing and it's not important it's not a ted talk it's not gonna change the world it's a very one thing i had this like issue in my freaking downstairs living room we talked about electrics what plug should i get yeah the best plug or that plug and it comes back and it's like you're definitely right to ask this question it's not about this plug it's about a whole different way of thinking about electricity within your house it's like they ted talk everything back to me and like please please those a lot of people are using codecs to build and code and really like it yeah i have spent some time with people building in codecs it looks unbelievable and not that is on my list the problem is there's too many tools right way too many tools computer cloud i'm spending all my time i'm trying to like really become incredibly proficient in cloud code but i'm actually probably gonna start getting into practically computer a lot because of all the stuff that you've been doing on that so let me switch to a complexity computer we have our prompt how many tokens are we going to use here let's we have a big credit pool don't worry about it alright do some token maxing right now we call this rip credits right now we just ripping through the credits looking through their credits what else would you be doing what's interesting though here and if you hit the arrow over in full screen it also will show you the usage and credits for this specific task now the file that shows you what files you're creating in that task the usage just for that task like like like that i like that which is good and then sharing and then to dos which i think is pretty awesome like the user experience is good oh they've they've got some cool ux stuff for actually they've got very cool ux stuff i like this fix and deploy okay yeah and so you can kinda see how long it's working credits the the files it's creating it's like from a ux perspective it is about a hundred times easier to understand the cloud code why don't they just integrate and allow you to host do they like host for these apps can i just put a website live or put an app live i think that's coming but it's not live yet and if they did that i would probably never use any other tool that's what i'm thinking through is like are they gonna go after that vibe code true vibe code market i i hope so i need to talk to the folks at slash if your complexity and you're watching this like come on the show talk to us about it come on this we'd love to to kinda work through some computer stuff and ask some questions around kinda how can computers is gonna evolve or show my tab for a second not because what i think i'm doing is better but there's an interesting point so i've been in computer here and kieran one of the things that i love about computer is that you have files and if you go to files you'll see you have all these files i have like hundreds of files in here right and what's really cool is i can call and access any of those files really easily oh wow and so in the prompt that i did here and you and i are are writing a book called loop and people are gonna wanna read it and pre order is gonna be out soon so more information on the show soon but i have all of the draft chapters in computer and we have a whole chapter on creators you can just refer to and i literally just said go look at the creative chapter in the loot book have they add collaboration features yes would you ever use google drive again no not if it was good at collaboration editing no yeah and by the way on all this you can literally if you're using the comet browser have them turning this into like google docs like it can manage all of your google stuff for you and we'll actually create new docs work god i've got a whole new tools gonna sound so i didn't have to put much in karen and because it had the chapter in context this is why context is so important it actually built an entire skill around creators and it's like create a researcher discovery loop creator fit score so it made a creator fit score based on everything we talked about in the chapter contact contact discovery partnership proposal email composition and send draft via gmail and so now i can just run this skill in complexity computer not even need a web app right but then i said wanna you go and turn this skill into a web app and that's what it's currently working on should build an app that allows people to input kindle book and extrapolate it into skills is based on that book no so free startup idea for anybody who wants to start a start and i would potentially do some funding for this is do the kindle marketplace but for allowing books to be accessed by ai because karen you don't get pdfs of those kindle files like you actually can't do what you say said i wonder if you can do through an api yeah you're right because over why would an individual get an api for their i kind of kinder books that's interesting for a lot of things actually how can you give agents access to like courses would be another one oh hold on i have a but real idea you're ready for this and we could maybe even do a show on this or maybe you would i can put on the background what if we figure out our way to do this with all the books are out of copyright so i just blew was mind so there's i actually used to spend all my time on that marketplace the marketplace of things that have just gone out of copyright what's eighty years or something because like like a hundred years i forget it's a long time so what if you took that on copyright it and like had ai rewrite and refresh like update them but imagine just having a marketplace for those books are at a copyright kieran and just charging a dollar yeah for easy access per book into your ai tool of choice where like you do all the data management and parsing and everything so it's like really good and you break it down on to accommodate the token limits and everything like that but i guess what i'm saying is one of the things i've learned from writing this book karen and whether this makes show i don't know one of the things i've learned from writing the book is like wow if you spend a bunch of time really thinking through stuff really deeply and then you have un b access to it in your ai tools the extension is incredible one of things i love about computer kieran is that you can literally just shut your laptop right now and it'll just keep building does you don't need to have your computer up and then you download the mobile app and you can just check on it in your mobile app anytime you want and all of the tasks sync across pretty awesome i think i'm gonna start using this for when we get this connected into hubspot hubspot project management slack oh the snowflake mc game over you can just make it into your hub that's what i'm planning to do the people perplexed are gonna love us super probably gonna be the best case study in the world yeah yeah using all the tokens on top of it yeah we're gonna be paying them an absurd amount of money literally token maxing it to the degree okay look at this it's deploying we might actually get this done which is pretty amazing let me clean up the player it's gonna play right session why is it gonna play oh it's doing a what is scrape k here's hope is live baby okay let's see i mean if it's done even a half decent job of this it's pretty insane alright let's open it up and check it out okay oh there you go alright you used a weird font but i want talk hubspot the comments is probably me one the hardest ones so let's see what yeah let's see what it does for him com i'm sorry this is sick by the way what's it added in here the import export oh it allows you to import export your json of course it does because who doesn't the wanna export a little json in the morning you know okay here we go alright buyer profile i mean it's doing a pretty good job for like you're impossible x exit five yeah shout tell yeah mart tech blue this is all very good they listen to the palette baby let's go it caught me out of the podcast fuck man computer don't do karen like that i mean come on think i i would actually just assume i'm first enough alphabetical order so that's why i'm gonna go and check my access my company feels may have been that live on show alright your top platforms i mean if this is good it's pretty insane this is done on a whim as we just gone on to start figuring out what we wanna do with this episode we were literally slack last night be like oh let's just build like a crater hub on the pod yeah and let's try to do it live and by the way we would actually be a lot farther we had thirty minutes of boo dog problems with claude and everything else yeah we've done this in way short amount of time we've really had thirty minutes given the problems oh look at this alright let's do it oh linkedin is the top yep newsletters okay click your platform of choice look at this and fits for alright pick one of them and let's see what it does so we'd recommend linkedin so i'm gonna just go over shrek yeah let's go let's see what linkedin is the hardest yeah so let's see i think it's gonna struggle linkedin but let's see because now look if this is mvp what will we do in a quick iteration we would give it access to apis yes of the x connector these different connectors that allow it to pull from different platforms but i kinda feel like pretty good i could probably do a couple of rev and sell and you made it in in thirty minutes of professional computer thirty any minutes where you and i were messing by wood figuring out if we should actually update winning mean we literally also took a side journey of yeah you know what let's just great modern versions of winnie the poo created by ai much which i still don't think is actually a really good idea one week we should do that and just for fun because i man maybe i'll do that today oh here we go actually ashley oh let's go oh failed some peeps here hold on kieran flanagan again on this i know it doesn't allow me to just search actually ashley why am i not on this come on i'm my baller here i'm like that just scroll all the way like make sure you're not i'm not on these people have tiny audiences is for to me complexities to screw any against you as they come live i'm more followers and literally everything single pick one and we gotta finish the demo alright let's to finalize my shortlist fit good recent content look this let me just wrap the outreach i mean yeah might getting you to switch from cloud code to computer alright let's look at the outreach let's see if it's good your mp post and have so audience your post mt based leads going head hard especially in the department that's an exactly conversation our audience so if rev ops managers vp vps aren't sales we're both reaching the same people i'd love to cobra juice with a fifty minute blah blah blah done i mean not bad for a literal prototype for a prototype for an mvp and you would use this prototype to send a few of these emails yeah and by the way like you can copy the clipboard but like it's easy enough to do the gmail integration with this look at this update i've reached drafted it's keeping the status i would actually have a little search bar here so i could search through and i could query who's on the list but yeah again i think the reasons picked all these is because of our profile this is pretty like i can remove some of these folks this is pretty dope alright so yeah that's our first iteration of our and by the way it built me a totally different version let's see okay i can show it to you see it made this much more a web i like that actually that's actually pretty cool that you have a campaign you wanna run and then find yep and then you can do create a research and so basically it'll go research everything and then you can add creator i'm gonna test this live with a campaign we're about to run after this oh i love this and this really good crater scoring which you had contact discovery one the things i found karen is perplexed is pretty good at finding contact information that's really good and then i have it drafting a actual full partnership proposal and then we'll draft the email i like starting from campaign i think that's better i do think that's best i i actually think overarching that app is much better actually the way it has the nav but but i think if you mush them together you have a great app and then when we connect it into house hubspot it could auto pull end campaigns it can if we have any of the creators in our database to pull those but i think it's interesting kieran is why i think they were so different is because of the context i had that whole creator chapter use that to build the skill and we basically put our blood sweat and tears into twenty pages about what we think you need to do with creators and it took all of that and basically made it into an app very cool which is pretty awesome so we covered a lot on the show today we talked to you about building with complexity computers we talked about some prompting clog some of that failed we actually walked you through a few prototypes of creator outreach and creators and partner with creators really important components of any campaign and we're gonna iterate on these and maybe on a future show we will give one of these creator apps away for everybody to use if you would like that hit that comment and like button and let us know that you wanna see more of the creator stuff and we'll be back real soon on marketing it screen

most marketing teams are scaling with ai right now it is the ai explosion but the output is getting much much worse for these marketing teams not better it's more volume it's more channels it's more agents run autonomously in the background doing the marketers work but the problem is nobody owns what comes out the other end the prompts these teams are using are inconsistent the agents are going off script the brand sounds like everyone else because it's just ai reg the output for each of these marketers your buyer is asking ai what product should i buy and your company is not even part of the answer and the creative you're creative is fast but it's instantly forget and here's the real problem everyone is having a debate externally about what marketing jobs will ai eliminate and it's really the wrong debate that more important question is what are the ones we actually need to make sure ai does an incredible job for our company what ai jobs will exist in the next three to six month from market tiers that didn't exist eighteen months ago and i'm gonna tell you the five roles you need to have in your future marketing team to make ai successful for you and your business all of that and more on this episode of marketing against the grain here's a quick word from hubspot hubspot helped tumblr solve a big problem they needed to move fast to produce trending content but their marketing team was stuck waiting on engineers to code every single email campaign now they use hubspot customer platform to email real time trending content to millions of users in just seconds the impact three times more engagement double the content creation wanna move faster like tumblr visit hubspot dot com why is this happening now what's change well i think three things are really coming together at once to kind of set the scene for why i think a lot of marketing teams are struggling with ai i think we are in the thing that i call the messy middle everyone can use ai to create things but they're not using ai to create things of any real substance now first ai went agent we move from tools you could prompt once to agents that run autonomously in the background twenty four seven you see it all over your ex feed all over your linkedin feed marketing team has went from using ai to deploy an ai and that brings a whole set of different problems ga says that forty percent of enterprise applications will include task specific ai agents by the end of twenty twenty six so it's not a prediction it's already happening you likely are using some of these agents yourself so within the tools you use there's much more autonomous agents being used the second one is we know this we've talked about this a lot on this show ai source traffic has really exploded from these ai platforms like chat tea and complexity it's increased anywhere between five to six hundred percent across the last five months we see this within our own data the amount of traffic we're getting and from these answer engines has grown over time the way buyers discover brands has fundamentally changed and most marketing teams are still running their seo strategy like it was built for twenty twenty third the content volume problem has hit a wall right eighty eight percent of marketers now use ai and their day to day work but only a third of organizations have moved but not beyond experimentation into something that's actually scale in right the gap is not tools it's how you have people use an ai there's just this huge skill gap there's this huge gulf in how you need to set up your team to be successful with ai our team put together an ai career strategy guide that uncover the one hundred thousand dollar ai jobs that no one is talking about it also includes an insider roadmap from ai experts and market data that shows one hundred fifty percent job growth in these emerging roles get it right now for free so we're gonna go through each of these five roles i wanna start with the prompt strategist rule number one this person is really the person that owned the entire team's prompt infrastructure right they build the shared library or prompts they train the team on how to use them they're accountable for the consistency of the ai output across the whole function you wanna think of them as the person who stops your team going rogue where prompts now the interesting thing is the deli between someone who can prompt really well on your team and someone who cannot is huge so if you have a prompt strategist that actually up levels the entire team provides them with a prompt library of how to get the best out of each model for the use cases that your company truly care it elevate the entire team all at once now here's real world proof companies using structured prompt engineering they report forty percent fewer hallucinations and sixty percent better brand alignment in their ai comps so that's not a small number when you're run ai across air team you know the failure story that's happening in your team right now is every person on your team they're prompted differently your content agent is using different instructions than your campaign agent your social copy sounds nothing like the email copy because there's different people prompt in different ways and the volume is going to keep going up and so you start to see this huge disconnect between all of the output your team is producing and the customer experiences all of that and it's incredibly jarring the prompt strategist is the person who really is going to fix this before it cost you dearly in bringing that to life in your customer experience so they would maintain and enable people on your prompt library so you can see here this is a prompt library we have different marketing and prompts we can scroll around so your team can come in here they can come in and they look to see what prompts they have for their different use cases if i'm an seo and au i can come in and say well i want my au content optimize there's a prompt the prompt engineer will refine this and improve this as they continue to see the input we can see the model we can see who owns the prompt we can see how many times it's been used when it was last updated so in this way your prompt strategist they own that prompt library they own the integration of all these prompts they own the improvement of their prompts because they can see all the output they can make them better and it's pretty cool you can start to see how your team is using these things across all of their work so that's number one the prompt strategist let's get into number two number two is a big one your agent ops manager and now this could actually be marking and wide or go to market wide and this is really the most underrated of all five as marketing teams they're gonna deploy fleets of ai agents and they're gonna be agents for content for a prospecting for campaign management for ae someone has to manage them i'm telling you from ex experience and someone has to manage them if the monitor performance they have to catch these kind of failure states they have to onboard new agents think of it as your chief agent officer right in today's word we have a chief product officer in the world tomorrow we're gonna have a chief agent officer and i love this quote here from joe mara who we've actually had on the show the ceo of crew ai he says just as devops reshape software deployment agent ops will reshape ai operations in twenty twenty six a very important quote mark ops is for the agent era right that's what this role really is now let me give you a real example so unilever is actually one of the most advanced examples of this play out in the enterprise at scale they've built what they call a brand dna system it's a train and repository that ensures the ai models only source from approved brand voices values and visual identity so their beauty ai studio produces assets up to thirty percent faster and critically they double key performance metrics like video completion rate and click through rate at the same time so that's not just speed that's a managed system so that's unilever have actually integrated ai and agent ops into how they actually do their marketing so without agent ops your agents are gonna be running all over the place no one's gonna be watching them via a dashboard they're gonna have inconsistent output then probably gonna break and ways you can't even imagine so that's the second role agent ops let's get into our third role and that's an ai specialist this is the one we've covered a lot on this podcast before and this youtube channel for how buyers find you has really changed forever so in twenty twenty you would do this search hey best crm for b b saas companies and you get your blue links today in twenty twenty six you go to an answer engine and it gives you an answer we know this is an important thing right chat has eight hundred million daily queries we've seen ai traffic grow from answer engines by five hundred twenty seven percent we know forty percent of old searches that i start with ai eighty percent of brands have not even started their ae story this is a really important role to have on your team you need to appear in these answer engines to even be in the conversation for the product or service that you offer so right now the opportunity for you is only twenty percent of organizations have even started implement this that is a huge gap and you can still be a first mover to actually integrate ai into your marketing and into your business strategy before we continue let me tell you about a podcast i love marketing school marketing legends neil patel and eric sue bring you daily actionable digital marketing lessons learned from years in the trenches marketing school delivers bite sized marketing wisdom you can implement immediately whether you have a new website or you're an established business you'll learn the latest in seo content marketing social media email marketing conversion optimization and general online marketing strategies that work today they just did a great episode called ten seo lessons that still work in the ai era listen to marketing school wherever you get your podcast so that's number three ae number four is ai content strategy this one is huge and another one that is a must have right now content via ai is an incredible winning in strategy if you know how to do it at scale so you still need a cra person to do this rule but wow if you have craft person that really knows ai you can scale your content engine like a large team even if you are just a team of one i really believe that i think one person plus ai can actually be a content team of ten twenty people right now that's how good these models have gotten but you need to be a cross person you need to have real domain expertise within content so the ai content strategy it works with these ai agents to build your brand's entire content strategy that's gonna define the ryan style if you want to see how this works look back at my latest video in cloud code where i showed you some of my agent content system built right into cloud code so it builds you a right in style it actually builds you a style guide for your brand that the agents can read and follow it's gonna build you win in profile platform so think of a view want to get really good x articles or really good at subs or really good at blogs or really good youtube transcripts you can actually build win in profiles for each of those platforms and the agent can follow that profile whenever it creates content for that platform because what works on linkedin probably doesn't work on youtube probably doesn't work in email so you have to have these win profiles per platform and then the important part is you edit the agent's output to make it distinctive and that is really important right you are not copy and paste them what the agent does and that's what a lot of marketing teams are doing today and it's why all of their content signs is same you're getting first draft you're getting ideas and then you are building on top of it but because they agents have become so good and the models have become so good they can do seventy percent of it and you do thirty percent but you have more time to make that thirty percent pretty incredible and then because you are ai native you take that edit you take what you did and you feed it back into the agents and the agents continue to get better over time so the ai content strategy is making sure your agent team is unique to your brand does not sound like every single other person and this is not a niche thing coca cola are trying to go oled ai for a lot of their content anyone saw their twenty twenty five holiday campaign it was a lot of ai integrated into it they obviously got called out in twenty twenty four for doing something pretty bad and it was really ai generated it looks really bad but the coca cola own content team have really described the problem of using ai for content perfectly the challenge and that's maintaining brand voice at scale knowing one when the ai i put is technically correct but emotionally wrong right trying to really find the craft within what ai is creating and that's the job of the ai content strategist coca cola is actually still trying to figure out the right balance the brand that hires this role can be ahead of large brands like coca cola who were still trying to do that and you compare that to unilever who i talked about earlier on their chief of growth and marketing who said this isn't just about pushing at more content and that's what a lot of marketers are doing anyone can do that it's about starting with a deep deep understanding of people's needs and desires if you go back to that video i mentioned the cloud code i talk about building and content profile for your audience to create content for them so you have to start with deep understand that the people's needs and then you have to have the creativity to create campaigns that cut through that's the ai content strategic mindset let's do the last one ai creative director and they are going to direct their agent team to create great creative for the marketing team they know what models to use for video or for copy or for image generation they're gonna maintain that distinctive voice for ai when it's handling the execution because just because you can create more content creative assets for ai doesn't mean you should right you should slow down and make sure the quality is really good i'm going to talk a little bit about unilever again they built an internal design tool called sketch pro it's an nice design tool using v three now this is a huge brand imagine all of the process you have to go through to get something done in unilever and they were able to use this tool to bring a concept to life for consumers in two hours not two weeks not two months two hours and the reason it works is because humans are directing it the ai creative director decides the concept to brief the aesthetics and the ais executed that and it's fine tune for their needs and what they found is content was created sixty five percent faster with up to fifty five percent cost savings and their engagement metrics have actually gone up and so they built their own internal creative studio without an ai creative director you're gonna get the coca cola version go look to see their ad swap campaign it's volume without any soul so just because you can do fast doesn't mean you can do better the key for this is if you have an ai creative director you can do both speed and quality at the same time and that's gonna be really really important alright so here's the five rules that i think are going to exist these are the future marketing team i think actually these five rules could probably do a lot of your marketing right now this is a great quote from andrew yang ai won't replace workers workers you can use ai will replace workers who don't i think we probably heard this before but i think that's where we're going and these five rules the prompt strategist the agent ops manager ai specialist ai content strategist ai creative director so i think these five rules are gonna be fundamentally important critically important to marketing teams of the future in fact i think these five rules can probably do a lot of your marketing and in every single one of these roles ai is doing the large part of the execution in the human oversight is making the agents in ai better and better and better and so the roles exist to make sure ai execute well and we're going into the era of where people are going to be skilled and managing an ai and making the from ai better so that is the future of marketing teams the future roles that we're going to see become more and more important i hope this was helpful for you and i will see you all on the next episode of marketing against degree everyone you know karen and i have been doing the podcast for a while now we've been at this for a couple years we love it we could not be happier to be doing this but we wanted to take things to the next level we wanna level up the impact we're having with marketing against the grain so the next step of our journey is something we're really really excited about we're gonna launch the marking against the grain newsletter and marketing against the grain newsletter is going to be amazing if you are a marketing leader practitioner you're in the trenches do in marketing every day this is for you we're gonna deliver right to your email inbox and you're gonna get all the behind the scenes frameworks practices tutorials from us from guest we have on the show and from people even beyond the podcast that we think are gonna be helpful and really have an impact on your day to day week to week doing marketing you're going to love it it is something we've been talking about for a while really excited to have it out in the world we've already got a hundred thousand marketers who are on this newsletter please join it's completely free we'd love to have you as part of the marketing against the grain community and it's easy you can click the link in the description below or you can head to marketing against the grain dot com slash subscribe

we got a special show for you today today we've got barbara back on the show barbara is one of our most popular episodes of all time and she's bringing the heat she is walking us through how to build a content engine in claude code and this is taking a single webinar transcript we all do webinars and turning that into blog post linkedin post email sequences everything you need to have a full content strategy and content plan around basic event like a webinar this is something a lot of people can use she is an ai and automation pro she's a non technical builder so we're really excited to have her come back and not tell us what she's doing but show demo walk through so that we can all build and recreate this ourselves let's get to today's show here's a quick word from hubspot hubspot helped tumblr solve a big problem they needed to move fast to produce trending content but their marketing team was stuck waiting on engineers to code every single email campaign now they use hubspot customer platform to email real time trending content to millions of users in just seconds the impact three times more engagement double the content creation wanna move faster like tumblr visit hubspot dot com hi everyone i'm gonna be showing you a really use workflow one of my newer ones it's an entire content engine that can be generating content for you after every webinar podcast episode or anything that your company wants to promote this is a workflow that i use in my own agency for all of my clients who are holding webinars we have this workflow kind of down to a tee so that we can make sure that we are planning the webinar building that promotion engine hosting the webinar and then taking advantage all of the content for promotion afterwards which means that when you do that entire ecosystem you have this really great content engine for your leads that you can kinda keep them warm that you can start converting them and you get that engagement in rs vps and so this is something i use every single day for my clients from planning the webinar to organizing everything to hosting it to dealing with the content afterwards which is arguably the most important part the live event is the least important thing everything that you do around it is what's important for keeping those leads warm and making sure that you are slowly converting them over time especially when you have something like a webinar series so after the webinar when you have your content your transcript your recordings here's what you can very easily deal with them to make sure that you're immediately promoting that content and making sure that you're reaching your target list with that and taking advantage of it so this is a workflow that i use for all of my clients who are doing webinars so we work from step one to step three in step one we have the basic planning of the webinar that we use ai four as well so figuring out who their audience is taking all of their current ic figuring out what is the best topic for them how do we need to structure this event where do we need to get rs sv vps what is our kind of starting point and then building that promotion engine based on it we wanna make sure that we are posting everything on social media we make sure that there is a distribution because just posting on social media is not really gonna get you those rs sv vps unless you have like thousands and thousands of followers and when you kinda have that promotion and content you can easily find ways to distribute via email linkedin and then after the webinar is done you need to then take that content take that transcript take that recording and make sure that you're keeping those leads warm so everyone who showed up for the webinar and everyone who are sv but my night have showed up you still have their email you still have to keep that lead warm and this is the part of the content engine that i will be showing today the post webinar content engine and how you can do it very easily with a very small team and basically immediately after the event and have that ready so if you wanna build this for your own team go to the link description and it will have everything you need to build this custom ai agent for yourself be creative be original tell what you need and tailor it to your own workflows click the link in the description box and you will find everything you need to build your own custom ai agent so in terms of the tech stack that i use i outlined everything here for all of the steps but we will be focusing on step three today we're using claude for everything that we need to do with ai and i will be showing how we created and automated the content engine we are assuming that you are using riverside for webinars you can use anything that you might want zoom is great riverside gives you a nicer recording so that's my personal preference it gives you a really nice landing page for webinars as well for rs vps so it's a really good solution to kinda handle everything we used to use a zapier integration to take the transcript from our webinar and then automatically just generated all of the content that we need so linkedin post blog post newsletter linkedin articles any outreach that we might wanna do after the webinar it automatically just gets generated for me based on the transcript this part we now automate automated it even more with a custom agent that we built so i'm gonna demo that agent today and i'm gonna tell you exactly how i built it so that you can build your own or you can just take the markdown file that we will leave in the comments or in the description and you can just recreate the exact thing that i'm gonna be showing you on my laptop all of the files can be basically generated and stored wherever your team works best so whether that's notion google docs whatever you guys might use so that everyone on your team who needs to look at the content before it goes out they can see it easily if you have different people for social media email they can see everything in one place and just prepare those promotions and send them out so it's really easy to sync across teams and for email and for linkedin automation we usually use hair reach and instantly because they have an integration together so hey reach might be a bit of a newer player but it's definitely by far the best linkedin automation tool i've used so that's kind of our tech stack for this third part so there's a few ways you can do this and it depends on what works best with your current workflow and how big is your team and what is generally the best way for you guys to sync on this content but you can use zapier you can use n a essentially what riverside will give you right after the recording is a transcript that transcript is everything that you need to generate the rest of the content so you should go into riverside you should download a transcript so i didn't automate this step because i already do it organically but look at your own workflow and if something needs automating do it if it's gonna work well it should be custom to what you're doing then you can use a zap that then takes that transcript and transforms it into these forms of content with claude that you need so you can outline exactly what your team needs for me that's usually linkedin post linkedin articles blog post newsletter recap email that gets sent out to attendees and then i also want two separate emails that are for cold outreach because you should be using that webinar and any upcoming webinars to reach out to people via a cold email it's a really good strategy for anyone who is maybe a lead or an existing warm lead that you have in your pipeline and claude would generate all of that for you now the other way that you can do this is a bit more custom but i think it's well worth putting a little bit of more effort in with claude it is basically your own custom agent so this is the one that i built specifically for my use case and for my clients we technically build each of these for each of my clients which might not be like the most effective or scalable way but again takes you like maybe two hours to build it so i don't think it's that big of a lift so this is what my content agent looks like it's fairly easy i build it through cloud code i am not technical but if you just go into cloud code if you describe what you need it will help you build exactly what you need so for example this does exist on this domain and i can share it for everyone to test it out to use it just so you can see what you can do with cloud code but i highly suggest you or anyone else on the team go into cloud code and explain what you need what your current workflow is if you've already been hosting webinars if you already have any kind of transcript that you need turn content just describe what you need and what your final result should be and it will tell you how to do that again i'm am non technical and i i've built this it just tells you what you need you need to have a little bit of patience you need to have a few you know back and forth if never one shot but if you have any issues worst case scenario you just take a screenshot of what you're seeing if it's an error and you just put it into clot and it will tell you how to resolve it it's a very creative and organic way to kind of communicate with ai and if you find cloud code a bit intimidating go into cloud chat explain what you want tell it to you draft what the tool you're trying to build should look like and then just download that as a markdown and put it into cloud code so you kinda have what's essentially like a prompt for it to start with but my favorite part about all of this is that it is essentially just a conversation so you can just go ahead and explain what you need we had a very clear idea of what we need to happen after each webinar so that was pretty straightforward to us but this is a test that i built just a demo here let's take it up for a spin as it was building it it was also telling me exactly what it needs so this is not something i would be able to figure out myself it just claude told me what to do and what to input and it kinda guided me through it so let's upload a transcript here from one of my recent webinars it's a transcript straight from riverside and this is also something that you can automate so for each client i basically have an agent that does the same thing and it already is familiar with who to speaker is who to host is it can pull that from the transcript as well because riverside gives you name and everything and it's also familiar who the target audience is what the brand voices is but obviously that changes sometimes for webinar to webinar so i like that i can kind of input it and control it that way and then we're just waiting for to doc to generate takes about two minutes and then we will show you the final part in the meantime you will see the google doc that generates as a result most of my clients work with google docs versus notions so that's kinda what i defaulted to in notion you can also have it set up basically an environment that you always use so like a full notion page with you know specific pages for linkedin posts for blog posts like anything that your team needs to kind of align on things and for me honestly google doc is easier it's easier for companies to share that you can really tailor it to exactly what you need and this is why i think everyone should kind of try to build their own versus just adopt whatever i give you as a download you can play around with it but essentially you should be able to create something that's custom for you because that's going to be the best automation for you as well and the benefit of automations with zapier and whatever is most teams know how to work with it at least in tech they're very easy to use their ux is great but it's not the simplest workflow either it looks very nice and simple ones zapier but all of those steps have to go in and be configured zap sometimes break you might not be aware that they were broken for a few days then you have to kinda go in and figure it out before we continue let me tell you about a podcast i love marketing school marketing legends neil patel and eric sue bring you daily actionable digital marketing lessons learned from years in the trenches marketing school delivers bite sized marketing wisdom you can implement immediately whether you have a new website or you're an established business you'll learn the latest in seo content marketing social media email marketing conversion optimization and general online marketing strategies that work today they just did a great episode called ten seo lessons that still work in the ai era listen to marketing school wherever you get your podcast podcasts so there it is content package ready let's open to google doc so here is the test that it outlines linkedin post newsletter recap short form video scripts and email follow ups sequences now that's usually what i need and everything else depends on the client but this is kind of what i wanna see as a baseline linkedin posts use newsletter recap because i wanna send that out the same day short form video scripts that helps the editor know how to take short of videos and then you have the email now one note if you build your own agent if you build anything with cloud code it is connected to the folders that you have on your computer which means that those folders can have all of the context stocks that you need and that ai needs to make this from a pretty whatever generic ai output to something that is very very usable for social media so for example if i show you my folders for clients i have writing examples of the founder i have case studies i have blog post guidelines i have ic list marketing blacklist words demos ic writing examples for both the founder and the brand and that has all of their writing in it so when you're building your agent that's going to take the transcript and generate this content for you if you have a lot of context stocks in that folder it will pull from it and the output is going to be much much better this is basically working with just the transcript so that's kind of the difference between really great quality that you can truly take from this google doc and mostly use it as is and you know kind of generic ai writing so take a second make sure that those folders have a lot of context as much as you can give it this will matter both for writing design anything you're trying to build the difference is truly staggering and then additionally to the context stocks when you're building an agent with cloud code you have what's called a markdown file where basically everything about that agent is stored so you can see the implementation guide that basically describes exactly what this agent is now you can take this file and you can put it into cloud code and you can tell it to build what is in this markdown and you will have exactly what i have here and you will be able to use it on your laptop it will take you thirty minutes with cloud code every single agent has its prompts and references now if you look at this prompt you can see that it's very very simple just take the webinar transcript run you through four separate your calls each with the specialized system prompt linkedin post like this is not a substantial prompt what i would do now if you have this built in claude simply continue to have a conversation with it so say i would like you to follow this writing style maybe you have like a full list of don't use m dashes don't use this sentence structure like just tell it what you need it to fix and just tell it to deploy that and the next time i refresh this page it will account for that so if you keep tweaking it the better and better the output will be what you can do is just kind of one shot at and give up on it because it didn't do great on the first try but honestly this took what two minutes to generate that entire dock i'm sure you have a few minutes to work with ai a little bit to be like okay i like this i don't like that let's lean more into this let's not do that you know work with it a little bit because this gets you to a really good seventy to eighty percent we're just in transcript in two minutes and this entire agent took me around two three hours to build so in terms of time and energy spent i think it's a pretty good output so this is what it looks like when you have a really solid folder with a lot of contact stocks both for writing and for visual elements of the company so a lot of brand guides all of the brand elements colors fonts everything like that and if you upload a presentation that you use during the webinar which it can directly pull some of these visuals from so this is what it looks like webinar series part one zero five and this is the blog post that it was generated it did both the text and the graphics i didn't touch anything i had to make very minimal adjustments and i just have this blog post ready to go and in the same breath it also generates a linkedin article for me which i've instructed it that it needs to be a bit more concise than the blog post the format is a little different optimization is different and it needs to be a bit more engaging for linkedin so in one go i get all of the graphics that i need for the blog post itself i have all of the text and i have all of the linkedin articles and i know that everything is optimized and then additionally you can keep working with it and telling it what you need so for example i want to have multiple touch points via email for cold outreach or for warm leads and i instructed it what kind of format i like i said what kind of links i need to be included there and then it just follows that format every single time and basically adjust it based on the latest transcript that i have so after the webinar i immediately have linkedin post that i can start scheduling i have video scripts that i can give to the editor to make some short videos added the full webinar it basically takes snippet of the most interesting parts that would perform really well on social media after each webinar you can also take social media posts from your linkedin and show the analytics and just put them in the folders so that ai knows what post performs best and can kind of optimize all of the following post based on that and i have the newsletter and i have email outreach not just for people who attended the webinar but for anyone that we wanna start targeting via cold outreach we wanna start targeting via linkedin outreach and basically any single target audience if you have a webinar and you know that domain attendees of that webinar were three different roles within three different industries you can also tell it to tailor an email response for each of those target audiences because you might have different landing page that you want those people to land on there might be resources that you wanna share then you can really tailor your approach there and your communication without spending a week of you know coordinating with a team and trying to extract all of that it just does it by default and then you get you really use the momentum from that webinar and kind of do the outreach and promotion and make sure that you're staying top of mind to those leads that just saw you on the webinar and then this is how the linkedin article this is also all generated with ai cloud code works much better for graphics especially when you have really good brand guidelines because if you generate a graphic you can just downloaded it as an sv spg upload it to your fig which means that you can move all of the elements that was an issue before with chad gp you kind of get an output but you can change it and if someone leaves feedback on that that you're like well i can't really change it i have rep prompt it you don't really have to do that anymore fig now has an integration directly with gl as well so it's becoming a more and more seamless process and then you can kind of load all of the social media post to go along with the short clips when you create them there's also different tools online they can help you with that riverside as well helps you generate some shorter videos and clips from the webinar itself immediately after it's recorded i work with pretty regulated industries and most of the time ai doesn't really give me what i need with short videos so i do often edit them myself but i added them in cap cut and has a lot of really good features so it's not really that big of a deal but i am excited for ai to remove that part as well so this is your entire content engine you can basically automatically generate any content that you need after a webinar a podcast recording you can even upload an inside call with your founder here if you need to generate some content from that and it will do everything for you and make it very easy for your team to collaborate on it it will land directly in your google docs or your notion wherever you might want it and you can very easily kind of take the next step and do whatever your team needs to do with it if you wanna automate any of the next steps you can do that as well as soon as the linkedin post is done it could be scheduled and am is great for that i personally like to review most of the content before it goes out but it just depends on want you and your team need and it creates a really good content engine that you can just keep rolling if you have a webinar everyone or two months this is a really beautiful kind of content ecosystem that you can sustain pretty easily with maybe just one person on your team so yeah if you wanna get this exact webinar content engine that i'm using here we will have a markdown file in the description that you can download and put it into cloud code and just tell it to build it it will guide you through the steps there's maybe a few things you were gonna need to help with web and give access to but you can have this exact thing on your laptop just tell me what you want fixed tell what your use case is explain what you're trying to build and you will have a really really useful tool that you built within an hour or two that is completely custom to what you needed i really encourage you to also try to build it from scratch because you know your workflows best currently and you know what you need so my solution might not be the best for you but it's a really good starting point and we will dive into the entire webinar workflow in one of the upcoming episodes and keep you updated with everything that we are building with cloud code and automating wow that was awesome huge thank you to barbara if you wanna go and build that yourself if everything click the link in the description below get hooked up with everything to kind of build what barbara showed you please hit like please hit subscribe we're doing more and more awesome videos to try to help you build an ai first marketing and business we'll see you next time on marketing instagram hey everyone you know karen and i have been doing the podcast for a while now we've been at this for a couple years we love it we could not be happier to be doing this but we wanted to take things to the next level we wanna level up the impact we're having with marketing against the grain so the next step of our journey is something we're really really excited about we're gonna launch the marking against the grain newsletter and marketing against the grain newsletter is going to be amazing if you are a marketing leader practitioner you're in the trenches do in marketing every day this is for you we're gonna deliver right to your email inbox and you're gonna get all the behind the scenes frameworks practices tutorials from us from guest we have on the show and from people even beyond the podcast that we think are gonna be helpful and really have an impact on your day to day week week doing marketing you're going to love it it is something we've been talking about for a while we're really excited to have it out in the world we've already got a hundred thousand marketers who are on this newsletter please join it's completely free we'd love to have you as part of the marketing against the grain community and it's easy you can click the link in the description below or you can head to marketing against the grain dot com slash subscribe

i give cloud code fifty one of my subs posts and in five minutes it told me exactly why some of my content works why some of it doesn't and then it generated an infinite number of new ideas that map to my winning in formula that that's just one skill of a much bigger content system that i've been building over the last number of months it's an entire ai content team built in cloud code it's actually a eleven skills across five different layers of content it can research your audience analyze what content works draft posts for linkedin news newsletter youtube transcripts and then it can actually improve itself every single month on this episode i'm going to give you a look over my shoulder on how i use cloud code to create content how i use my content team and i'm going to give you one of the best skills for free let's get into that and more on this episode of marketing ending against the green here's a quick word from hubspot hubspot helped solve a big problem they needed to move fast to produce trending content but their marketing team was stuck waiting on engineers to code every single email campaign now they use hubspot customer platform to email real time trending content to millions of users in just seconds the impact three times more engagement double the content creation wanna move faster like tumblr visit hubspot dot com so we are going to give a look over my shoulder as i use my ai content team to create content up it's going to be awesome so what is this this is really for people who wanna start to see the possibility of cloud code and if you are a system thinker a system thinker how you can use cloud code to be your super weapon how you can use cloud code to super charge anything that you're doing in work if you can think about how to build systems with it and in it and this system is all about content so i'm gonna give you a quick visual of the system that i'm going to bring you through today this is over on my subs for people who want to check it out and so this is the system we're gonna go through today the little starred one here is when i'm gonna give you all for free and so hey it has an orchestra skill so the orchestra skill is going to actually use all the other skills which makes it really easy to use the system you can kinda just thigh with the orchestra skill and ask it to do things and it goes and uses all of the other skills for you it makes using these systems really really easy you can build a content audience profile which basically you wanna figure out who you're creating on the content for build a profile and make sure all of the content is tailored for that person the system does that for you you can build any number of writing in styles right i build a right and style for me because i create a lot of content but if you haven't created much content before you can create a right style around someone you like and then adjust it to your taste and your audience it has really great research and so it caused these content ideas skills it has a talk point extractor it can go off and find viral talking points from trend and topics or you can give it content you like a youtube video a pdf and it will actually turn that into talking points that they can create content from it has a lookalike content skill my favorite and one of the ones that i'm gonna give you today you can just give it a data dump i give it fifty one of my subs posts it creates a profile of winning content patterns and then goes and finds other content ideas that map to those patterns show you how it works there's a lot going on here it has content skills this post and richer is so good the compost postal richer is i give it an existing post and then enrich reaches it with different parts of content and that works really well in different platforms maybe a data point maybe a story maybe a case study content creation linkedin use other an x now youtube would be in here but i only added that midnight last night i have no life i need to actually really think about that and then feedback loops this is so important if you're a system thinker most people stop here i actually have built an app the app captures all of the content that the system has created and then actually allows me to input performance data and then it takes that performance data and every month i run a review and it actually makes the skills better right just been thinking we just dropped the eleven ai skills i personally use to build my own content team i'm giving you the full architecture it doesn't just generate content it studies your best work it will research your content in real time it will make drafts across every platform and it even updates itself on what's actually performing get it right now for free click the link in the description i basically went into cloud code and then i say let's start creating content and that triggers your orchestration skill and so the orchestration skill allows me to see what's in here so it picks up an audience profile this is for the lookalike research skill basically this is fifty one subs post and it's created a profile of winning patterns in my subs so i can actually go find other ideas that map to what my audience like it's picked up my rhino style i've already created a random style for me but i can create an infinite number of writing styles it's already picked up content research that it's done previously it's picked up eight linkedin post created i'm gonna show you that because then the app picks up all those posts with performance data and actually improves the system the api keys are x per complexity fire crawl and open and then in the automatic i have another agent which is claude is managing all of this automatic means basically is when i pick this option the system would just work itself so i will just come in every morning and the system will have created a content using all of these different skills and i don't have to do anything and that is what i'm building right now is an agent team to just do it all for me it tells me what you actually need and a previous episode of at cloud covid james who's really great at this the board marketer this is really like vi i i know it's was never a big fan of that kind of vibe market and where it workflow tools because it's not vi you have to actually drag and drop all the workloads together this is not software so i have a skill that builds this audience profile and basically says hey these are three things need you don't need those three things because you can just iterate and refine in real time and i didn't want the client to domain i just give it a target audience i give it a primary content platform i actually wanted x articles because i think x articles i'm pretty interested in starting to quit more content on x because of x articles and so i wanna actually start to do that because i really like long term content and so it's building me that profile i ask it to actually validate the research before it builds a profile so you can see here it's asking me is this right and actually was really good it picked up on this is a growth practitioner who's really obsessed with ai and make an ai work within their company super accurate and once you confirm it creates this audience profile now this is hard to actually see it's creating an d file so it's gonna create that content audience md file that and the right in style power of the entire system because every other skill looks to see what content profile and right in style that you want loaded to create content alright we're gonna go ahead and see the html i purposely create html for show how to do these things the system does not need the html html file but it's easier for me to show you all what it creates this is my ai growth practitioner this is specifically for x practitioner this is the industry and you can see here it's created a audience identity jobs to be done the kind of pain points pilot hell i love this by the way this audience profile i have perfect this this is really good it took me twelve months to kinda go back and forth so i have some really dope stuff here so the vocabulary library i find really good where you can say what they do and don't say all kind of the way they speak the emotional register and like validation hooks it's like really good this is not like an ic right this is actually content they react to and it's all based upon research and engagement data with pretty lengthy it's pretty awesome and it's specifically for x articles this persona on x articles you could see the amount of context that i'm given this system but just when i wanna create one type of content for one type of persona now this system allows me to have ten twenty thirty rhino style ten twenty thirty content profiles and it can load anyone i ask for and so it's really flexible we are back in the orchestra scale the next thing you would be doing during onboard is what ryan styles do i have to use so you could create a writing in style so the way the right style skill would be used is i basically wiggle go create a right in style for myself or for any creator and what it does is it goes and scrape the web creates a file for that person and then creates a write style now i've already created one for myself i have a founder voice model after da one of best writers there is i have my own one here so let me again show you maybe just how that looks so you can see how this is all kinda shaping up so i have a style card right and so it creates me this whole right in style it goes and scrape away for my content creates a cool write in style captured from existing content tells you what it's sample primarily platform is a subset it does struggle with linkedin and so what you could do for your own system and i have done started doing this is you could just have to go and like export your own files to upload because it tries to get linkedin from per complexity fire doesn't have a great time get linkedin post so it has a lot of stuff like why hasn't it not capture like x because i don't create much or x at the moment but i actually wanna change that because i do like the articles so this is like primarily focused on subs so if i was doing this i would do a arrive style per platform right that's how i would think about it so this is like my subs one so that's that right so we've done content profile and so what does the talking point one even look like so i'm gonna show you how it works and i'm gonna do that before and after because i already have some talking points created what i'm gonna do is show you how i run it and then i'm gonna give you the example of the output but the output won't match the ask i've just made because i've done that it's a prior one that i did last night but let's create some talking points for marketers who are trying to adopt ai now i could also give it existing content it we'll take existing content and do the same thing but i'm gonna show you how it looks i've talking points and content ideas in separate folders this is your lookalike alike skill that we're gonna show so this is talking points and so we're gonna say hey we did some of these already so you can see it has viral talking points it dates them so let's open this one which was really around cloud code for marketing and so again look at this right it maps to the audience so at this point it pulled in a vp of the demand generation and then it looked across reddit and looked across x it looked across per complexity and it pulled up these kind of pocket points and so it has here like your marketing team is building the tools it used to buy it has y skilled beat prompts for demand and work this is a good potential post if your team is still copying and paste and prompts from a notion doc you are a one model update away from losing in half an hour of output quality every time cloud changes and then it gives you a court insights some of the reasons it believes that's true gives you the application of this it basically categorized them so these are educational posts things where you teach someone these are data nuggets so james you come on the great v marketer he has a post over here where greg is talked about the ability to create agents to do a bunch of marketing talks about the action the context the value these are all kinda of data points but i wanna show you down here spicy take your job title marketer is split in two in twenty twenty six there are two types of people in marketing departments go to market operators who in systems and everyone else who is still waiting for engineering tickets so that's a pretty cool one right like that's a good spicy take this the impetus of a good post so these are just like creative sparks for me i like to go through this and i'm like oh these are you know interested well your function is mostly the most rep replaceable thing your demand gen budget and then we have these kind of story sparks story sparks are like bring to life something through a story so you can see here and this is kinda built around a story at ten pm she give cloud code one directive by eight am thirty four new landing pages built live on a site i would tell you if you use that hook on social that will do really well and so this is pretty cool that's the output for you on the talking points so that's viral talking points go scour of the webs or take existing content and turn them into talking points so let's go back here and we'll run the skill that i'm gonna give away because it's also pretty amazing before we continue let me tell you about a podcast i love marketing school marketing legends neil patel and eric sue bring you daily actionable digital marketing lessons learned from years in the trenches marketing school delivers bite sized marketing wisdom you can implement immediately whether you have a new website or you're an established business you'll learn the latest in seo content marketing social media email marketing conversion optimization and general online marketing strategies that work today they just did a great episode called ten seo lessons that still work in the ai era listen to marketing school wherever you get your podcast podcasts now let's look at the other one that you're going to get access to which is the lookalike one which is like one of my favorites the way the lookalike one runs is i would go and say here's a data dump of my linkedin post here's a data dump of my subs post here's a data dump of my top performing x articles here's a data dump of my best thought leadership post on my blog and i would upload that messy file to this skill that you're gonna get access to and what it does is it takes that mess and if it has performance data in there it will look at the top thirty percent of your best performing post because it wants to extract winning in patterns from the best performing posts if there is no performance data in there it will just look across the data you have given it because it can't tell which ones perform better than others and it first of all creates a profile because it wants to work from a profile and so let me show you the output again in html the system all works in d files but for the purposes of this video i wanted to be able to visually show you how this works so it creates my subs content profile and so what it's broken at is it's basically said hey like this is the wit formula this is the top performing clusters of content look how cool it is here tells me the different clinic of clusters of content across my subs it talks about the structural dna of my subs so like typically how lungs opposed to different sections the structural type paragraph style embed prompts work really well you'd haven't checked that my like you should it's awesome the hook formula the emotional playbook and format that works always really important to say like what doesn't work and so from that messy data that you have given it it creates that profile and then it goes to take that research and gives you a bunch of ideas and so we're gonna go back here into our content system we have these ideas so you can see here i ran this last night for my subs look at this how good is this right these are ideas for subs again always tailored to an audience so it picks up a content profile and we'll basically say what content profile do you wanna use if you remember the content profile is the audience and then it will say this was the winning in formula and here are ideas again i create content i use a lot of the content system for idea education research first drafts and actually i'm i have a pretty cool skill to show you that actually helps me enrich some of my content and so here it has one about meta prop because when it looked at my topic clusters it could see that that was a really great topic for me is like people really still want to understand how can get more of these models the ai boss era or how to run a fleet of agents across our marketing team this is a good actually your ai has been in green with you too much here's how to fix it pod models wow i replacement marketing teams this is a really good one we were in the era of functional teams and now we were in the era of pod models and i think you have to structurally really rethink the way you think about team structure the six x gap why some marketers get insane results and most don't study of nine thousand ai users find a six x productivity between the effective and surface level wow this is super cool wow data hook so it tells me which they are by the way now you can see that kind of frames it up for me ai frameworks it's a data hook it should be a four step tutorial and this is the emotional driver how to simulate your customer's reaction before you launch it i've done that one so that's interesting that's picked up that one why your linkedin prompt fails in every other channel okay yeah that's a good one actually what it's to saying is that there's like this quirk what you say create a great post for linkedin to the model and the model goes and says well this is what a great post looks like linkedin and linkedin for the most part sucks well that sucks but there's a lot of bad content on there so you're gonna get a ton of bad content from that so you can see how good these are i'm actually reading through some of them myself i only read through the first five the most unused ai technique in marketing and every ai put you very seeing was a first draft the ai never reviewed its own work beforehand handing i mean i kinda feel like this is better than me it's come up with so much good stuff before we go to market push i fed the entire strategy into my a model and ask you to find every way fail that's very good i find six problems we seeing here are the exact process and then it does like a nice little summary here but you can see here gives me the title the key pattern right i always like to do pattern matching i think a content is kinda modular it's patterns tutorial and then the trend and signal that's pretty good and so where's the one i really like pods replacing teams future of teams framework and helps hubspot story yep pod defaults for high forms yeah wow so much great content too much great content but but as we said we are building an adjunct team and the agent team is going to do this autonomously and because you're a subscribed to the best youtube channel on all this stuff you will get that first when i built it so let's go back to our graphic i'm gonna like show you the post and richer because it's dope this works the way you think it would work but i will do a separate video just on content creation because too many people will use these cut and paste that's not how you do that right so i think i will do a standalone video just on feedback loops because i think that's pretty important but given i wanna make sure that you can get through this and there's a lot to it we're gonna go add a sequence here in terms of how the system should be used but again i keep talking about this this is not like using software it's flexible and so when you've extracted talking points obviously the system would then say here's your talking points and you can basically just point at one and say okay like turn this into a draft linkedin and post turn this into a draft youtube transcript turn this into a draft subs turn this into a draft x article but what i wanna do is because i wanna show you the post richer i'm gonna try to see if i can just give the talk point and ask it to enrich that post we have data enrichment i have built content enrichment tools and i wanna show you how they work and i think i will do a standalone episode i've have just decided to do a stand with an episode on content enrichment because it's super dope but i'm gonna show you this alright let's see which one of these we like education educational workflow redesign is a real eighty percent adoption domain expertise speech to prompt and skill your job is now ai agent and management spicy take roi gap is a coordination problem not a measurement problem agencies need to be brand proactively really reality gap standards data and they'll get ninety one adoption versus six percent zero employee ai mark team the room where nobody uses ai let's see what stories spark is okay so let's do a first draft of a linkedin post for the story spark zero employees using the linkedin skill and then we'll use the post richard because i'm kinda doing that sequence so i wanna show you the content enrichment tools need a first draft to enrich and so we're trying to enrich just the idea is kinda hard alright so it's on a post here's your linkedin post so remember this is a story spark so it should actually feel like a story also you this is a draft in idea and then you make much better he didn't hire a marketing team he onboard one so it's actually taken someone's real story documented this in favor he built the full market function with zero employees for ai rules fully integrated from each other adds analyst producer blah blah each role had to find inputs i outputs it hand the whole stack run so if i was doing this i think they've cool version of this is marketing in teams of won and gets big multiple examples so for example there's one person in claude running a lot of their marketing so i'm going to get my transcript that i've done here and actually start to like format this into a kind of marketing team of one story enrichment so it does have enrichment but i'll show you how this works the party that expanding simon he didn't just pick four ai tools he structured each agent like an employee in board and doc a defined scope yet so she tells me to go that's the gap between marked and one blah blah alright so like it's a pretty good first draft again first draft when you look at this i do not use this to post content and so these are all ideas and now it's pulled in my enrichment here so when enrichment should actually get me modules case studies stories examples enrichment let's make sure we use the post richard skill here we go alright we're back so this is the beginnings of an enrichment tool that i'm building so it looks for ways to enrich that content twenty twenty two jeff sent an internal memo that changes how amazon built software every team had to expose their data and capabilities through clean surface interfaces divine and blah blah twenty years later the same principles are showed up in ai market stacks the agent can that run reliably in production are the ones built with precise world very cool connection the failure mode isn't ai quality it's the same failure bezos with solve in twenty twenty two undefined rules that don't scale whether the work is you're in very cool gives you a natural case study here but a platform running autonomously talks about the connection and then gives you a pretty good quote polaroid right sir who runs the ai marketing institute next gen marketers know that in order to deliver personalization experiences modern consumers expect marketing in most smart blah blah okay so you can see at the enrichment tool and i can easily just pick one of these and it will enrich that post alright it's not finished it's saving that post but i can show a different one so remember i was really interested in that marketer rolled being split i thought that was a pretty good post i did do a first draft at that one the marketer job split into this year most people haven't noticed very cool it's actually done it as a linkedin post it's so good spicy take again audience profile hook pattern contra punch enrichment type because it all my stuff gets enriched enrichment source actually tells me where it was word count talking point so it toss about five i pull from from on one that as well so you can see what it looks like it's really good by the way right the mark jo to this year most people have noticed on one side go to market operators they build systems inside cloud code hello i've got engines running on eleven apis ad creative generators produced in a hundred variables blah blah blah on the other side marketers waiting for the next platform to do it for them and chris walker put the logic plainly in twenty twenty four instead of having ten hundred k doing it what if we had two hundred sdr str augment a people with ai blah blah blah blah very cool and so it saves in there the other ones in here as well he didn't hire a marketing team here on onboard one i have a hook skill and so i can improve the hook with that skill much like i have a content enrichment skill so obviously i would never ship this i actually think the more interesting one is the rise of the one person marketing team but there's some like really good stuff in here and here's the enrichment that we did now the last thing i wanna close this out in the last minute is where people don't understand the importance of feedback loops because they're not systems builders but systems builders do and so what actually happens here is the system picks up all of the content in these apps this one sucked i was when worst performing post people did not like a product position they're not answer it and so it picks up all the posts that i've created and then i can go here and i will be able to add the link api at some point but i don't have it added at the moment and so i have to manually add the data in here but once i add the data what actually happens is that i can just go and run a skill and the skill does review analysis of all of the performance so we can see what it was created from the system it can look at the performance and it actually updates all of my skills so it's a skill that improves my skills to everything i showed you will get updated and improved each month based upon what's doing well what's not doing well and so that is cloud code content system your free skill that you get for free which is one of my favorite actually it's pretty cool the lookalike content audience and if you don't have a lot of content then you can go collect other folks content that you like and give it that as a data file and it will go find those with patterns you you don't have to have your own subs your own linkedin your own x you can go get that content from elsewhere so that is the content system there's a lot in this i'm gonna do some follow ups on the feedback loop gonna do a follow up on cloud code for content creation and how you could iterate really really fast all of that and more on a future episode of marketing and i guess green so make sure you have subscribed and i will see you then everyone you know karen and i have been doing the podcast for a while now we've been at this for a couple years we love it we could not be happier to be doing this but we wanted to take things to the next level we wanna level up the impact we're having with marketing against the grain so the next step of our journey is something we're really really excited about we're gonna launch the marketing against the grain newsletter and marketing against the grain newsletter is going to be amazing if you are a marketing leader practitioner you're in the trenches do in marketing every day this is for you we're gonna deliver right to your email inbox and you're gonna get all the behind the scenes frameworks practices tutorials from us from guest we have on the show and from people even beyond the podcast that we think are gonna be helpful and really have an impact on your day to day week week doing marketing you're going to love it it is something we've been talking about for a while really excited to have it out in the world we've already got a hundred thousand marketers who are on this newsletter please join it's completely free we'd love to have you as part of the marketing against the grain community and it's easy you can click the link in the description below or you can head to marketing against the grain dot com slash subscribe

in ai ai day skills and building skill files are one of the most important things you can do to up level the results you're getting from ai and today we're giving you a twenty thousand dollar ai transformation consultant in the forum of a deal we're gonna walk you through how to use it how to get the most out of it and bonus that the end of the show we're actually gonna give you a master skill template that's it's gonna make it really easy for you to build any new skills as you want for yourself or for your team if you want the files we talked about and can click the link below we'll get them right to you let's get to today's show here's a quick word from hubspot hubspot helped tumble solve a big problem they needed to move fast to produce trending content but their marketing team was stuck waiting on engineers to code every single email campaign now they use hubspot customer platform to email real time trending content to millions of users in just seconds the impact three times more engagement double the content creation wanna move faster like tumblr visit hubspot dot com okay so kieran recently on the show we talked about some new data from ant that there's this big gap in what's possible with ai technology and what people are actually doing we kinda tease that we wanted to help people solve that and i'm here to say that i have spent probably ten hours and i don't even know how many credits of compute to try to come up with the best solution to that problem and i wanna go through that with you in real time today get your feedback and then it'll be available in the link below for everybody to go and try this out themselves but we've been talking a lot about on the show skills and how skills are real unlock of how you use modern ai assistance and agents and i've spent a lot of time building what i call an ai transformation skill that helps anybody take the work they're doing and transform it to an ai first way of doing it and you kinda talked about this i think you have a dream scenario of like how this should work so tell me what you think the dream scenario is and let's let's see if i can deliver on it yeah i think that the way this is gonna work for most teams is that you can record yourself just doing work and that rate in your work and you can give that to an agent and that agent will build skills that help you replicate the kinda workflows you were doing on a day to day basis minus it's the easiest way for the average person okay i agree with that so here's what i did just for the sake of the show karen i posted in here a transcript it's two minutes long i did a very short video recording of myself walking through something i have to do all the time which is review and synthesized information in slides c that feedback and then send that feedback in an email something you do all the time something i do all the time something i think everybody watching the show probably does all the time right right so that's what you're looking at here is literally just a transcript and what i've done and this is in claude i've attached the skill file as well as the examples and reference data so if you're doing this you just literally get like a zip file folder of these markdown files and you can just upload them to cloud to chat gp to complexity whatever you're using right and you can tell it to use these skills so it's literally as you're seeing that's all i've done right i literally said run the skill against this transcript so i have not seen this done this way before you're just adding the files right in the chat and tell that to use this stuff in the chat so the way i'm showing this done right now karen is i would say what call the basic of basic ways you can go and install this skill in claude but what i've done is i've made a fairly complex skill where i have a skill file and i have different reference files that all come together in one folder it's using claude c or cloud code or complexity computer it's easy to just upload and access that whole file so for the purpose of this i'm trying to do it the basic way which is like you just upload the files in that folder to a chat with whatever l you're doing and you put the tasks that you wanna do it this is not like oh i'm gonna do this a hundred times i have it perfectly installed in claude this is a i'm not sure what i'm doing and i wanna get started really quickly and really simply does that make sense right yeah okay yep so that's where we're at here that's all we're looking at and so i have the skill file i have the reference files and i have a transcript one of the things that i think's is really important karen is i've scoped this skill to understand the type of work the person's asking to do and i basically have it is this a task transformation a deeper like individual level transformation a team level transformation or a department level transformation and so you'll see here that i have it changed the output based on the type of work it is in this case this is a task so i have a much shorter output i have a two to four page output versus like if it's a team level transformation you could have like a forty page output so it's basically reviewing back to the skill in real time to understand what type of work it is and scope the support that it needs to give so what's really important the next thing here is that it's very flexible depending upon the type and scope of the work you're doing okay then i also have it checking for current tool pricing to make sure that it's giving you like reasonable accurate pricing and the recommendations it ben goes and goes through all the steps and writes the report and here is the output of the ai task transformation plan we're giving you an amazing skill that you can use to transform any work you're doing to be ai first you can just talk to it type it upload work files record what you're doing and it will tell you how to do it in a better ai first way what it's gonna cost how long it's gonna take and how to actually do it and you're gonna stick around to the end of the show we're gonna give you a bonus skill as well yeah you can find everything in the link below just give that a click and you can get those files so this is an output that you're gonna get no matter what you put through the skill you're going to get a report like this which is basically what the heck am i trying to do here's the task that i am currently doing it understands the tools that are currently used to do this task and the current pain points of doing the task then it gives you a before and after of what are you doing currently and how long it takes versus using ai how long they think it's gonna take and so this is open deck manually hold a evaluation criteria in your head while assessing each item right and then it's showing you when this case what claude could do in that same way and it basically then is showing you how it's thinking and working through the problem and how much time it thinks it saying hey if you do this i as a cmo we're gonna recover one hour to two and a half hours of time per week is its estimate and it's gonna tell you if it's just time savings or if it thinks it has improvement in quality in addition to like the time and efficiency it's gonna give you tools recommendation this is like the most basic task so it just has claude as a recommended tool in here it's not like a very complex multi step task and then it gives you the next step to go and try and build it out for the week and long term so literally you can just record yourself and it will tell you how to do what you're doing in an ai first way thoughts and so it's not building your skills it's recommended them like a workflow an ai workflow it is recommending an ai workflow either for a human to do with an assistant which is this example or if you're talking about a much more complex example it will show you actually you know what agents you should build the skills you need all of the more in depth answers like for example if you have a problem that you wanna just run through we can also put a more complex task in here and kinda go through and see what's happening and what it recommends so in the report can you bring up the ai workflow what did it recommend k this is the ai transformed workflow yeah so this is saying this is what you're currently doing and it takes twenty five to forty five minutes this is the ai i assisted takes eight to fifteen minutes and this is the steps it's telling you to follow specifically and get this a very basic task this is literally just trying to show people that you could do this with anything when does it get the current time from that twenty five to forty five minutes how does it calculate that number yeah it's estimated because i go through in my voice over of it and say what i'm doing okay it's like i'm going through each slide manually and so it's an estimated time right now obviously if you are going through and just doing something in real time right if you just record an hour long video of yourself doing something then it's gonna know it takes you an hour yeah yeah yeah yeah so here's a different example karen but i took one of richest looms like the three hour one yep where he's just working claude code and i put that in and it did a full breakdown of what he was doing in the technology so like it says it's like hey it's three hours because that's how long the video is it breaks down on all the current steps he's doing to do data analysis right yep and then it's actually telling him all the things from an infrastructure standpoint he should do to make it faster and better so like run project bootstrap script create directories import docs converts to md like basically automate a lot of the upfront work to basically change his workflow right and you could literally then just take anything from this report and put it into claude per complexity chat gp would have you and have it build you a more detail plan right like you can keep building on any of this this is meant to basically give you an initial breakdown of like you're doing this thing and you could do it in a different and better way yeah we'll be right back wanna tell you about another podcast i love the d pod hosted by ramon bar and blaine bolus is brought to you by hubspot media d pod is a podcast about all things direct to consumer ramon and blaine cover everything from starting growing and optimizing e commerce stores and direct to consumer brands they talk with founders marketers platforms creators and marketing and growth agencies to cover topics like brand building social media influencer marketing website conversion paid media facebook ad and much much more if you're interested in the stories behind your favorite consumer brands this podcast is for you they did an amazing show called meta ad secrets how top d brands spend three hundred k monthly profits you can listen to the d pod wherever you get your podcast what don't you like about it what would you do differently i think it's really good i think when you look at it it's quite overwhelming the amount of information for a single workflow like i have hold report for a single workflow is a lot right you're having to read like a four to six page report on how i could do slides better well you don't have to read it you could literally just upload that report to claude and have it give you the prompts of what you need to do right you're creating like a more detailed sc of what the work is really yeah i also made it that way because there are a lot of people watching who have to deal with very conservative companies and bosses and just like they're gonna need that level of not just detail but simplicity of detail of like you have to do these things and do it this specific way and this is how we think it's going to work so that people aren't not like freaked out yeah i think i would kill myself if i was in a company where i was getting reports somehow to automate my work well i guess what i'm saying is i don't think i would ever actually read all these reports but i would use them as inputs yeah to then build the agents and build the workflows and build the skills and everything that you need right yeah i think the ability to upload our transcript and get ideas and how you can do that better is valuable for all people right now everyone's trying to work more efficiently everyone's trying to work faster and this is a good way to kick off from lumen with your teams and gets some ideas on how i can go faster there are a lot of teams in have very complex workflows right it's like oh i'm using you know canvas and hubspot and g drive and snowflake and all these things right and actually one of the things that i learned from you very early on in this ai journey is that most teams are not clear on like how the work they do actually happens right right and so part of what this is and i think why you need a more detailed output is so that everybody's actually clear like this is how the work we do right now actually happens yeah and do we agree that that's how it actually happens and if so then can we agree on a new and better way to it yeah yeah yeah it's like having another little work analyst exactly and again you can click the link below if you wanna check out the skill the other thing i think we're gonna put in there karen is in part in making this i made a separate skill that's a meta skill for making new skills and i know you you're like well all of the platforms have like skill builders right yeah the direction i it was to go and crawl all of the current ai skill directories out there so like what are the most popular skills the most used skills and how are they set up because remind everybody when we're talking about these skill files you are limited to five hundred lines in one of these files that's how long that file could be it's like the challenge of what you're trying to do is pack the most density possible into five hundred lines and so i went out and i actually did a huge bunch of research with in imp complexity computer in this case and basically learned from github h corp a bunch of other places what skills were performing took the best practices and basically consolidated that into one skill template that you can now use to build any skills which i think it's pretty cool that's pretty cool yeah so that is a skill builder yeah you can see so this is a skill architect it builds remarkable agent skills that are discover concise effective across all major platforms so i'd also like make sure that they work on everything when to use the skill create a new skill from scratch for any domain improve or ref and existing skill design a collection of related skills a skilled family i've been working on a couple skilled families decide between a single file versus multi file skill architecture so a lot of the stuff like you don't wanna take the time to think of this crap right no but it's one of those things where it's like it's that it's compliant that it has the right architecture that it's actually improving what you're already doing this is something that if you have just you can run a pass on anything you're doing and it's just gonna make it a little bit better and that little bit better is really impactful especially if it's you're working on building a skill that you're gonna use a whole bunch yeah this is really good i think this is gonna be really helpful for people which is you can just install the skill on your system of choice and then ask it to build a skill mimic this template yeah and it's just gonna follow a really good template it has the right decision trees for architecture and how it should be organized and how it should handle skill families if it needs a skilled family right what i just showed you on the ai transformation was a skilled family and folder that it had essentially made because i used this skill to iterate and improve on that last skill i showed you the ai transformation skill was made a lot better by running it through that skilled builder which is super cool yeah i think that's really cool yeah i think there's a couple of things that you need to do when you are just working on building in this ai world and some of it is just like having some meta level processes and skills and so a skill to build skills i think is something that should be anybody's toolbox hundred percent okay now when it comes to ai transformation we talked about this skill we talked about the easiest thing to do is literally just record yourself on screen what you're doing use the ai transformations skill to go through it after that's happened cured you're like hey that report is really overwhelming what would you do with the synthesized data of that report like if you were recording yourself doing something for an hour like how would you then go and use that to make it super actionable for you you get this really detailed read of what all that work is it depends what you're trying to do in terms of your ai transformation there's a multitude of different things so if you were trying to get folks to use a certain tool that i'd say you wanted to get all your team using claude mh or all your team using open ai that you would skew all that towards what are all the things that they could have done in chat or club that they're doing manually right now if you really wanna get your team using skills in whatever platform they're using then you would say we'll use this to create skills that teams can share to automate this workflow or if you wanna start to make ai part of your employee on you could create onboarding docs of how we work and that would actually really help i think it's gonna be friday you nuanced or different dependent upon what you are trying to get your team to do i think it does start we're trying to understand ways we work today and then how ai can start to help i think the big part that ai is gonna struggle to help with is behavioral change yeah talk about that i think a lot of folks can kinda maybe know what to do but changing your behaviors and habits is probably the thing that humans are struggling with most like you kinda just default to the way you do things because that's the way you do things and what have you seen working there in the the work that you're doing that's helping people change or is it really hard yeah i think it's really hard right i think that where it's changing faster is actually where people are doing things that they weren't able to do before like people actually code in where they were never able to code before people do an analysis where they were never able to do analysis for people like doing the new things like applying ai to doing new things that they were not able to do before i think it's a little less exciting to use ai to do something you were already do even though it could help there it's like shiny object syndrome yeah i think start and small as well don't try to get overwhelmed and do all the things just try to use it in a certain number of small ways and make them part of your habits and then evolve from there the framework i've been using here kieran is i think in a pre ai world the average person was more likely to be like lower agency so less dependent less just trying things less building things and higher tolerance which is like i gotta go to this meeting or the other this thing isn't that great but i don't have time to do it any better and i think all the people that we're seeing make this transformation are the exact opposite they're super high agency they're curious they're out there building they're like hey i couldn't do this before so now i wanna go and do it because like now i can and that's awesome and now they're also now very low tolerance and they're like hey this stuff stinks and i can make it way better and i can make it way better in like minutes or hours not days or weeks and like there those two things combined are really really powerful i would encourage anybody to like look around it at who you're working with and i suspect the best people you're working with fall into that like high agency low tolerance bucket yeah right people that have high agency and just wanna make things better yeah this is a time for winners transforming with ai is very hard i think karen and you and i believe it is one of the biggest business opportunities in the world is just transforming to a more ai first method of working we've tried today to package up wide that's important how to actually do it gave you a couple of skills to go and hopefully do that a little bit faster and a little bit better and we'll be with you real soon on the next step so marking it's great hey everyone you know karen and i have been doing the podcast for a while now we've been at this for a couple years we love it we could not be happier to be doing this but wanted to take things to the next level we wanna level up the impact we're having with marketing against the grain so the next step of our journey is something we're really really excited about we're gonna launch the marketing against the grain newsletter and marketing against the grain newsletter is going to be amazing if you are a marketing leader practitioner you're in the trenches do in marketing every day this is for you we're gonna deliver right to your email inbox and you're gonna get all the behind the scenes frameworks practices is tutorials from us from guest we have on the show and from people even beyond the podcast that we think are gonna be helpful and really have an impact on your day to day week week doing marketing you're going to love it it is something we've been talking about for a while we're really excited to have it out in the world we've already got a hundred thousand marketers who are on this newsletter please join it's completely free we'd love to have you as part of the marketing against the grain community and it's easy you can click the link in the description below to or you can head to marketing against the grain dot com slash subscribe

on today's show we're gonna show you the single most important chart in ai that explains why your company isn't going to be a winner just because you use the best models you heard that right the best ai models are not gonna dictate who wins and who loses in the ai we're gonna give you the actual winning in formula and how you can implement it at your company today all of that and more on this episode of marketing against green here's a quick word from hubspot hubspot helped tumblr solve a big problem they needed to move fast to produce trending content but their marketing team was stuck waiting on engineers to code every single email campaign now they use hubspot customer platform to email real time trending content to millions of users in just seconds the impact three times more engagement double the content creation wanna move faster like tumblr visit hubspot dot com alright kit we are here with yet another ai model gp t five point four is another model beats lots of benchmarks apparently one of the best models on the planet and we're kinda here to argue in this twelve to fifty minute video that it doesn't even matter it that this model does not even matter doesn't matter how good it is model capabilities are not the important thing right now in the ai industry yeah we're gonna instead tell you what the most important thing instead of the model's is getting better what that is and how you can leverage that for your business to actually change growth in this new era and kieran based on that there's a chart that's going viral from ant philanthropic that i think was the lightning rod moment for the conversation we wanted to have and a lot of people are interpreting it one way we have like a very different interpretation of it so gp five point four great code model great intelligence model beat them and t and lots of benchmarks but we've been here before every single model that comes out kind of goes a little bit more up the benchmarks and now there are ways that they kind of align themselves a benchmark to do really well but we are believers that model capabilities are already very very good and actually given people even more capable models is not gonna make much of a difference right now because of this chart so this chart is being shared pretty widely around x and it's from ant so for picture on t they're putting at these to try to show what the impact of ai could be across industries so what it's showing you is theoretical ai coverage which means how much of that industry could be theoretically automated with ai and there are no surprises here ai is really good at code and math it's very good at finance it's very good at engineering type roles it's very good at legal very good at like arts and media pretty great at all of the office and admin work so it can get you a lot of coverage in a lot of different industries and the red is the observed ai coverage just how much ai is being deployed within this industries and how much of that work it is automated and i think what you and i talked about on slack is we don't believe that for the average ai user the model is already good enough it really is not going to matter this year if they get another model another model another model the really hard thing about ai is actually integrating it into your existing workflows i think one important thing to understand is like i think everybody's is looking at this chart and they're looking at this blue area which is theoretical and they're like oh gosh ai is gonna wipe out a lot of the jobs in all these categories and like that's very theoretical might happen might happen year from now might happen a decade from now we don't know what's the most interesting i think what you and i are interpreting this differently is that there's this massive gap between the red and the blue and there's some work that's happening in coding and in business process and in sales and some of these customer service these really important markets but it's like still small relative to what the perceived opportunity is and that blue is really kind of the perceived opportunity so right if we all watching the show today agree that like oh there's some theoretical much bigger opportunity versus where we are now like what the heck does that mean is the big question we just dropped the ai adoption playbook it gives you the exact framework to actually redesign your business with ai it's a proven roadmap that helped one company book eleven thousand meetings and another to resolve issues thirty nine percent faster get it for free click the link in the description so there's a good story that this has happened you know once before in time i i was giving you a good quote from a dinner i was at and i'm with a bunch of finder and one of the founders in there is incredible and he had this incredible quote where he said you cannot walk into the future if you're looking back at the past and i think this story i'm gonna go through kind of explains what's happening which ai is being compared a lot to electricity alright it's like a fundamental thing that everyone is going to access and get a lot of value from and thomas edison built electricity generating stations in manhattan and london in eighteen eighty one and within in the year electricity was being sold as commodity so we had electricity as a commodity in eighty eighty one this incredible new invention but by nineteen hundred less than five percent of mechanical drive power in american factories came from electric motors and so companies had not integrated into how they were doing work and if you looked at that it's because those factories kept their old layouts they just swapped street steam for electricity and then ran the same processes so think about that in terms of ai the productivity explosion of electricity when in factories only happened when they redesigned their factory floor around electricity right so they started by stepping into the future with electricity by looking back at the past the same old processes i'll keep the same processes i will swap steam for electricity and i'll just do that thing and there was only five percent adoption it got wide adoption when all the factories when they redesigned their exact factories around ai that i think is going to be the fundamental shift that happens in companies not ai model capabilities but redesign the company to be an ai native company team structure skill sets how people do work and that's gonna take way more time where it takes these companies to come out with another model another model there's a cool thing you can do now go to complexity go to chat go to any one of these models and ask it to build a table of open jobs in google deep mind claude and t and open eye and tell it to show you who they're hiring for because if you ask who they're hiring for it tells you a lot about their strategy and i do this all the time they are hiring for ford deployed engineers right people who can help companies integrate it into that business but not swap steam for electricity but how you start to redesign your business around these models i think this is honestly one of the most important points we've ever made on the show is that the bottleneck is humans that human bottleneck is going to take i don't know karen and what do you think decades to fully get through i think it depends if we get agents that get deployed as an employees and can actually do a bunch of that work internally but for the average customer you and i have talked about this like there's like we go to dinners with ai finder and it's a different planet everyone's in there like cloud md file trying to like get my open claw agent to like run a bunch of my stuff then there's a category where are kinda doing some stuff with ai and they're like hey i'm using chat on a day by day basis asking it some questions maybe getting it to write an email and then there's a group of people are like what the fuck you all talking about the majority of humans but that's the majority here was a study majority yeah there was a study recently maybe from open ai or another one of these companies where like normal people it just people surveyed eighty four percent had never used ai right that's how early we are if you look at some of the data only eight point six percent of companies have even deployed an ai agent in production right there is such a small amount of usage and open ai i had a state of enterprise report actually that was pretty interesting where it just show the top five percent in terms of intensity in terms of usage where orders of magnitude way ahead of everyone else like the gap between the top percentile and everyone else has never been greater well one it shows you that the single biggest opportunity on this chart is actually the gap between the red and the blue and right if you can help any a company transform and become ai native then there's a lot of opportunity a lot of money to be made there because like we said the model's are smart they can do all these things right back but i wanna tell you about another podcast i love the d pod hosted by ramon bar and blaine bolus is brought to you by hubspot media d pod is a podcast about all things direct to consumer remote and blame cover everything from starting growing and optimizing e commerce stores and direct to consumer brands they talk with founders marketers platforms creators and marketing and growth agencies to cover topics like brand building social media influencer marketing website conversion paid media facebook ads and much much more if you're interested in the stories behind your favorite consumer brands this podcast is for you they did an amazing show called meta ad secrets how top d brand spend three hundred k monthly profit you can listen to the d pod wherever you get your podcast i think what's going on g is a few things one first of all all the cost of these models is greatly subsidized right now and like if you compared the actual cost to get from red to blue like in terms of actual cost versus a non you know subsidized cost it would be pretty expensive right like the amount of compute that you're paying for inference that you're paying for to close that gap it's gonna be a lot it's not going to be tens of dollars or hundreds of dollars can be tens of thousands of dollars right yeah i agree with that okay but we are not the episode just to give you the opinions just to give you like here's what we think we wanna give you the pro calories could you've been working on a skill that can actually help companies close that gap okay so i'm here in complexity computer which is one of many ai tools it's kind of inc to what what we're using here and i had to do a task i gave it a very complex prompt it ran for well over an hour and this is my v one i'm gonna edit it through this and and i'll give you the skill at the of the day but what i had to do is basically can it build a skill that you could use in any ai tool chat petite claude per complexity cloud code what have you any tool to actually help you take your team yourself your team your company from you know being that factory that isn't set up for electricity to your metaphor care and to being an ai native you know organization to actually understand what it would take okay so k what i really had to do was go and research current best practices in ai transformation look at all the current frameworks rethink them and actually make a new framework for us and a new skill that's gonna help people take through so here's what it did it researched twenty existing frameworks it did a bunch of real world case study analysis it did a failure analysis and current best practices from mckinsey and h b and wharton and all of the fancy folks here's the framework here and i want your opinion on this the new framework is called rapid five and r is reveal assess the team's actual workflows maturity and map the jagged frontier of where ai helps versus hertz their specific work a architect design the target ai native operating model with workflow by workflow before and after designs technology selection and change management pea proof implement through two weeks sprints on real world pilots not synthetic measured across three horizons efficiency capability and transformation in grain shift from tool adoption to identity shift through peer learning ai first defaults and performance integration indeed d build a ninety day reassess cycles because ai capabilities change quickly alright what do you think about the framework here yeah so let's say so that's changes quarterly and that's important because it actually understands that things are changing in a fluid way so this is really good what you're really doing and what we're showing folks here and i think you really want us to go deeper on this specific use case you're building a four deployed ai skill four play engineers so people are listeners understand they're going into companies they're working for open cloud working for these different ai products and they go in and they implement ai into your business and so when i've talked with a lot of ai founders their biggest problem is that companies are struggling to integrate ai and agents into their workflows which is what that chart is saying so they hire the specialists who go in and do that for you what you're actually kinda of building is a skill for the average business to be able to do four deploy ai engineering within their own company and so this here is like a pretty good framework what is the inputs to run this skill the skill requires essential team profile core workflows five to ten current ai state and transformation goals like that's the bare minimum important data and systems environment leadership culture context constraints risks and then helpful competitor benchmarks individual skills inventory and prior performance data okay so a lot of this is gonna be tricky to get actually i think a takeaway for folks here is if you wanna close a gap between the red and the blue for your business because you think there's value and intricate ai across a certain team that you can actually ask ai to actually start to build a skill that will allow you to kind of give it some input around that team and i think to your point one of the suggestions would be have your team just do a bunch of looms of how they work and then take those transcripts and then give it to your skill and this skill extrapolate that using some sort of framework like you have here and then creates the ways that ai can automate those things and maybe we'll show that in in practice maybe we'll do an episode where we show to would part two of this where we'll like really do the screen share we'll try to simplify the inputs and you and i will just keep iterating on the skill like offline and part of that episode until we get it like really really tight and then we'll give it to everybody so i'm sure there's some people who want the skill now we will give the skill away once we think is like really good we wanna go through a bunch of fictional examples and then we wanna do a couple real world examples and then we'll know it's like pretty good cool okay that's the episode opportunity you might be ahead of your competitors but you're probably behind the market opportunity big market opportunity i think it's funny that ant philanthropic put that chart out there's a bunch you could talk about it but i think the most interesting thing is that is a real gap for a lot of businesses from where they are to where they could be appreciate you watching and hit like hit subscribe we'll see you real soon on marketing it's a great hey everyone you know kirin and i have been doing the podcast for a while now we've been at this for a couple years we love it we could not be happier to be doing this but we wanted to take things to the next level we wanna level up the impact we're having with marketing against the grain so the next step of our journey is something we're really really excited about we're gonna launch the marking against the grain newsletter and marketing against the grain newsletter is going to be amazing if you are a marketing leader practitioner you're in the trenches do in marketing every day this is for you we're gonna deliver right to your email inbox and you're gonna get all the behind the scenes frameworks practices tutorials from us from guest we have on the show and from people even beyond the podcast that we think are gonna be helpful and really have an hacked on your day to day week to week doing marketing you're going to love it it is something we've been talking about for a while we're really excited to have it out in the world we've already got a hundred thousand marketers who are on this newsletter please join it's completely free we love to have you as part of the marketing against the grain community and it's easy you can click the link in the description below or you can head to marketing against the grain dot com slash subscribe we'd