A $200 Claude subscription won't replace your marketing team

0:00The AI hesitation when it comes to buying software is going to be around for a while. Like even just before this we saw random website, right? And I it was like obvious aiwebsite.com, you know, it's like blue buttons, white background, like everything you would expect. And when I visit sites like that I have like a very innate just like automatic like discount that happens in my head. Uh and it's not because I hate AI. In fact, I I I used to really hate AI and all stuff. But as it goes on I'm I'm like, oh this is the this is amazing. Um but I still have that like is there anyone out there? Like hello, you know, like tapping on the glass where no one's behind the curtain? Ladies and gentlemen, welcome to the

0:46factory floor. I am stoked to be back with my uh my fellow homies here, Corey and Zach. We're going to be talking about a bunch of great stuff. And I'm actually going to send it to Zach to explain what we're talking about today and probably a little bit about why we're talking about it today. >> Today children, we're going to be >> [laughter] >> we're going to be talking about what founders and CMOs don't get about AI marketing, design, and development. Specifically about what goes into it and their expectations around what is actually possible regarding time and efficacy with the outputs of the entire process. And the reason this came up was because uh we had a client a while back who had mentioned something like that they

1:39they were expecting us to do a full website rewrite and redesign in 6 minutes. At least that's what they had indicated. Um and they said this because they had built a series of programmatic pages, like you know, so a templated page and they built several of them, you know, I would imagine they could probably do hundreds, you know, of these templated pages. And their assumption is completely false. Like, everybody who has tried to code a site even with cloud code and rewrite it knows that you cannot do that in 6 minutes unless you're okay with shoddy output, bad copy, bad design and all of the trickle-down negative effects of doing something in one shot even with a tool that is loaded with Corey's marketing

2:30skills, that's got front-end design skills. It doesn't matter. Those things do not stop you from creating something really bad with that expectation. So, we go and talk about it. >> [laughter] >> Was it you who ended up responding to that? >> I did, yeah. >> Yeah. >> Well, and there's so for some story, uh this this particular founder had mentioned some other websites that I know for a fact have spent millions of dollars on their visuals, on the photography that goes into them. These are like multi-million dollar Department of Defense level websites that they were aiming for as their aesthetic. And they also didn't have anything regarding copywriting. Like, there was nothing, a a zero-level foundation for this

3:25website. It's different if you have, you know, a site like ours, like Conversion Factories before we started working on it um in cloud code. We had a solid foundation. Like, the copy didn't change really. We were just polishing the visuals and that still took a couple days to get right, you know, the to the point where we felt really satisfied with it. So, what we want to talk about the reason I want to bring this up is because it sounds akin to what I would hear in some of like the early days of my design career, which was "Hey, just Photoshop it. You know, like can't you just Photoshop it? Can't you take, you know, this like take this really crappy picture of like a a beached whale and turn it into, I don't know, Jessica

4:06Simpson or something on the beach. Can't you just Photoshop it into that?" It's like "Sure, if you give me I don't know, a month to to turn this beached whale into Jessica Simpson." And I think that's what this sounds like. It's like you're trying to turn chicken into chicken soup in 5 minutes, and nothing is that magical. >> My my sense on why there's such a like misunderstanding of of what people want versus what they how they think it gets done is that a lot of people don't really have a good sense of like what the outcome should look like. And so any outcome is a good one. Like if you press a button and you don't really know what you need or want, and something happens, and you're like, "I

5:00had nothing and now I have something." That something is amazing cuz it took 10 seconds and you clicked a button and then it showed up. But it it does not relate to if it's the right thing. I think that's where the biggest like that is kind of like I guess similar to the like just Photoshop it button. If you really don't know what you want and you click this button you'll get something, and if you don't know what you want, then you can't know if it's right when you get it. Is that makes Is that Am I kind of in line with maybe where you're like where that the spirit of the the like what it What are you talking about? You don't know what you're talking about kind of came from? >> I think that that is part of it, the

5:37novelty of like I couldn't do these things before and now I can. That's how I felt when we were like, you know, just getting started with Cloud Code and felt like I had the potential to do things that as a designer I know would have taken a really long time to code by hand or even to communicate to a developer effectively. Um and the fact that I mean you can look at Conversion Factory's website, that's spark animation that's on our homepage on our hero. That would have taken weeks to try and do by by hand. And I could do it because I'm familiar with the terminology. I to communicate what I did and did not want with enough technical sophistication that the bot understood me. And with a good amount of patience cuz doing that

6:25still took like half an hour to to nail and really get right. It was like, "No, no, no, it's not cuz it initially gave me things like flying embers all over the screen that looked like like these really weird solar flares that you would make in Illustrator or Photoshop and they were ugly. So it still took a lot of finesse and that's half an hour spent on one animated piece of our website to get that level of polish. And the funny thing is it's like you obviously trust us based on the caliber that of our website cuz that's where we get judged on our work. It's like if our site looked like garbage, no one would come work with us. And I can tell you that it's not something that's done in one shot or an hour or even several

7:19hours. It took time. And we also had an aesthetic established before we jumped into that, which is a big piece of it and getting alignment on it between the three of us of like this is where we're going. We're starting from nothing. It's ex nihilo creation that then has to be agreed upon and get consensus uh before we move forward. And that's made doubly long when it's not you guys who trust me and our team to just do stuff. Where it's somebody who's going to have other opinions. And that's fine. They should have opinions about the way that things look and the way they sound, the way that they like the way things feel. But it's not going to get done in 6 minutes. >> [laughter] >> It's We had We had design, we had, you

8:05know, creative direction, we had copy for the entire homepage, we had all of our images, all of our graphics. We had uh case studies, case study images. Like we had everything to then develop it and it still isn't like a one click I just built 11 pages in 6 minutes. Like that that cannot That isn't how it works. Or if at least if we tried to do it like that, the output would be incorrect. Yeah. >> Well, I mean, so Corey, I'm interested to hear what you think about this because you've done this as a non-designer. And like what do you think about that perspective? I mean, I I'm fully aware of like if you can get something out there that gets traffic and gets you people signing up for your website, go for it, man. You got to

8:49start somewhere. But when you're coming to somebody else, I feel like I had this thought actually. Like if you are going to pay somebody the level of amount of money that you like the amount of money that you're going to pay us even just on a monthly basis, if you pay anybody $6,000 to do something that you could do yourself in 5 minutes, you're a >> [laughter] >> Don't pay. >> That isn't a a very apt observation. >> Yeah, it's true. I mean, if [laughter] if you could just do the exact same thing in 5 minutes, then like go for it. And I feel like that's been a lot of the conversation that we've been having is cool, if you can, why don't you do it? And the answer is because they don't and they can't.

9:34Every time I hear somebody say, well, uh I'm not going to say who, but what I had ha- I had another comment today about someone who's like, oh, um, can't you just do this with your $200 a month Cloud Code subscription? And my answer is, yeah, let's see what you got. It's not going to be the top tier level of design and copy and SEO perfection that we put out. Um, and and maybe one day you can, but that day is not today. And I don't really think that you will be able to one-shot those types of things. It takes a lot of rounds of iteration, a lot of polish, a lot of um, going through like over and over and over again and experimenting with different layouts because if you do want something that breaks the mold of what

10:24the AI slap, um, generators just create from the get-go, uh, then it's going to take time like just by default, or sorry, by definition, you're going to have to spend a lot more time on it than the average person to get something that doesn't look average that they created. Um, so I just don't think that they're actually spending the time to do it or to try it. It's just like baking baked in their head now. Like you said with Photoshop, I think that's actually a really, really good analogy where they have this conception that, oh, in 5 minutes you can just use Photoshop to fix anything and turn a beast well into, uh, bikini babe. That's not how it works. It's going to take a long time to do that.

11:07Um, and so you hire us to not only spend the time to to do that level of polish and to get it that level of quality, but also to have a better starting point in the first place where you don't have to transform something ugly into something beautiful. We just make something beautiful to start with, and then we can get it from 90 to 100 much quicker. That's That's kind of what I don't know. Every time I hear it, too, I'm like, "Well, I think the bar is just getting way, way higher. Like, sure, maybe to you this is a really nice website. I don't think it's a nice website, to be honest. Yeah, like >> never I've never seen somebody who said that who has sent over a site that's like, "Hey, you know, it's nice."

11:54Never. I've never, ever come across somebody who had one-shotted it in like 6 minutes and got something gorgeous. >> Yeah. I was going to ask cuz I think the the craziest part about this to me is that, in my opinion, humble opinion, we are of the more AI-forward creative marketing agency. You know, there's There's other people who do the same type of work as us. And in From what I can see, it feels like a lot of agencies kind of die when they don't adapt. They have a way of doing things. They like doing things that way. And they know it works to some extent. And so, they don't want to change it when the new tool or the new thing shows up.

12:41I feel like we aren't that. We're like, "New thing? This is You know, there's a new thing that's actually like helpful and it's kind of like integrating itself into this work, and we want to be the experts in that thing." And so, of this group of people, we're some of the more willing to dive into that. Yet, we still I can't imagine what the other agencies' level of pushback is on "Well, no, you can't We can't use Claude to do that. Like, that would be That would be bad." Or like, "We can't use any AI to help us with to help with copywriting." Obviously, like, yeah, we're not just like pushing a button and then like giving it to you, like, "Here's your copy." You know, like, there's There's work that goes into it,

13:21but I can't imagine the level of pushback that people get. That being said, we are using it. Um but we all can agree that it's not a push-a-button fix problem thing. And Zach, I know you kind of mentioned like this is just kind of like reshuffling the work, where the work lives. And so maybe I >> [laughter] >> His daughter is knocking on the door. >> Yeah, she is. >> [laughter] >> Well, it's cute. It's all good. Um maybe you can give us an idea of like when you say, "Hey, this is kind of just reshuffling how the work gets done." And why can you kind of like break down that thought and maybe we can all color it with some awesome conversion factory knowledge. >> So sorry, I was distracted by a little

14:06munchkin outside my door. You were You were You were asking when I say that this is just a reshuffling of how things were done previously. Um but with an augmented base, well like >> [sighs and gasps] >> until it is actually like a like a Willy Wonka level of like I push a button and all of a sudden there's this beautiful thing in front of me and that happens 10 out of 10 times, you always need a human in the loop. You need someone who is directing it the the charge. And I mean like the the difference now is that you like I haven't touched client work or like most design files in a while cuz we have a team that executes on Figma for me and takes a lot of the ideas that I come up with and then

14:59is the hands and feet to them. The difference is now everybody on our team has hands and feet in the in the form of an agent that is going and creating things. But you are still directing and it's almost like it it's totally different of like all of our designers now are now creative directors guiding a developer to bring things to life that they had otherwise envisioned. It also doesn't take away from the fact that you still need a bunch of other assets that get created to put into your site and give life to it. Sure, there's some instances where, you know, like if you're doing an abstraction for a software, Claude's really good at doing that. But now you have to double-check it and make sure is it actually pulling away the

15:45right information? Is it adding color in appropriate places? Is Is there anything that I can do that will embellish this to make it feel more branded and bespoke? Um all of that has to be dictated. I mean, and we're dictating the smallest details that I can fathom. You know, like I'm saying like I'm saying things like move this inner shadow at like 200 milliseconds, two pixels down or the REM equivalent on hover and then have it go exactly back out the same way. It's the same same level of detail and prescription that that I would be executing in Webflow or Figma. But I'm just talking to it now. And there's even a little bit of a trade-off there as far as timing because I can move a lot faster and quicker in Webflow

16:30than I can in or in and in Figma than I can in Claude code. For details, like that's where you run into this messy trade-off of like Claude's great at doing it big swaths of of work, but it is not fast at the detail stuff where changing a bunch of things have to be real time is time-consuming. Like I can make four variations of a button, change the color on each of them, change the border stroke on on all of them in Figma in 30 seconds. But to do that in Claude, it's going to take me twice as long to try and do that. >> [gasps] >> So, can I Let me play devil's advocate?

17:19Cuz if I'm listening, I'm like, still not worth the money maybe. You know, like I'm I'm just going to like No, no, I agree with what you're saying and that there's reason to it, but some someone listening might be like, I don't care about the two pixel hover and I don't care about the little details. I just need it to go live. I just need it to, you know, it's all about speed, speed, speed, speed, speed. Um is it that they're not seeing that the quality that you put out is directly kind of related to the quality of the product that they're buying or the service that they're buying? Is that the reason to care more about those details? >> 100% Right now, the particularly when you are building a

18:01software company, your website is the equivalent of a storefront. Like, when you look at the windows inside of an Apple store or Louis Vuitton or Hermes or anything that has any kind of semblance of design and intentionality, you can very clearly see that they put thought into even things like the trim, the flooring of their shop because it matters. They put like, you know, when you get a good tag on like, you know, you go into a store and a shirt has a super cool tag on it. Someone actually paid attention to that and that's cool. That's why this has a higher price tag on it. And if you say, "Well, my my product's not a premium product." It's like, doesn't matter because you are going to Do you want to be the person who looks

18:47like they don't care or do you want to be the person who looks like they do? Cuz if you don't, there's nothing economically smart about being stupidly ugly. >> There's nothing economically smart about looking stupidly ugly. That should be on a t-shirt. >> That's a line right there. Bars. >> Thank you, Bill. Thank you, Jordan. >> I want that out completely out of context on a shirt. >> [laughter] >> Oh, that's so good. >> Corey. You you you take a lot of one shots at things. You take a lot of one shots at websites and stuff. So, I kind of want to hear your flavor to it. >> Mhm. Yeah, one more thought about that just a little bit earlier was just that okay, let's just say that it does take 5 minutes because at some point well, not

19:33at some point just the way that the LMS go, the models, the all the research they do, they get better over time, right? So, maybe today um yeah, it takes 5 minutes to make something pretty mediocre, but like not horrible, right? And maybe it takes let's just say it takes 3 hours to get something really, really good, right? Um and then maybe in 3 months it takes 2 hours and another 3 months it takes 1 hour. That doesn't change anything because at the end of the day someone has to be in the driver's seat doing this, working with the agents to produce this. And number two, that value's going to get passed along to the client. And that's why I think that our not to toot our own horn, but our pricing model is

20:21pretty um impervious to these kinds of changes and updates because uh we're not charging per deliverable. We're not charging for um a certain guaranteed amount of anything. We are a retainer and so with our number of work streams that you have active we can crank things out at the speed at which the tools allow us to. And that will just get passed on to the client. So, it doesn't change anything. If at the end of the day we end up producing things in 5 minutes for our clients, then guess what? The client's going to see something 5 minutes later. And then they're going to see the next thing the next 5 minutes later, right? But um yeah, going back to the one shots uh yeah, there's there's this tweet that someone had

21:07recently and they were like, there's going to be two types of people, and this is like the dynamic duo in the age of AI. There's going to be They had these really clever names for them that I'm forgetting now, but there's going to be the person to basically go from like zero to 80, and they're going to start something, they're going to flesh [snorts] it out, outline it, they're going to get a V1, they're going to kind of one-shot it, and then they're going to leave it and move on to the next thing. And then person number two is going to take it from 80 to 100. They're going to be the polishers, the connectors, the person that makes it work, that QAs it, that takes it to that next level and applies that level of

21:49craftsmanship and detail and hands-on attention. And so, yeah, I one-shot a lot of things. I was doing it yesterday, where I just one-shotted like four pages, and then I was like, "Hey guys, take a look. Let me know what you think." I'm not going to spend all the time to nitpick it and detail it and like go through all of it because it's actually better and more valuable. >> [laughter] >> Yeah, for one, I'm I have a ceiling. I have a cap on my talent and able to take it over the finish line. But number two, you guys with fresh eyes are going to be able to uncover and tell me things that a lot faster that would take me a lot longer to find the same things, probably. And so, um I think that that working relationship and that model

22:39of like yes, there's always going to be the five-minute zero to 80 one-shot. But then there's also always going to be the one-hour 80 to 100 polishing, refining, iterating. And the devil's in the details, you know, there there's This is how work has gone forever, especially here in the internet, where it's like the last 10% takes 90% of the time. Cuz there's just so much detail and there's so much polish and there's so much uh work and creativity that goes into making something go from good to great. And that's our job. Our job is to make it go from good to great. People can go DIY going from bad to good all day long. I'm happy to let them do that. Sure, go ahead. But as soon as they want something to go from good to

23:28great, that's where they're going to start to feel that ceiling. That's where I feel that ceiling personally and why you always need someone to come in and and bring it over the finish line. >> Yeah, I mean well even like the like the step that I was doing for Ministry Match Corey, like the the illustrations. >> Yeah, exactly. >> There's no way that that's going to Pod can't do that. It physically cannot create those illustrations. Believe me, I've tried. >> [laughter] >> And it can't do it. So you have to use other models. And then even using other models, it's like these are to put it in a perspective, like to have someone do those illustrations by hand the way that they look, maybe we can

24:10have, you know, Michael um show them just so people get a feel for what I'm talking about. To achieve that level of consistency and have somebody else do that for you, you're going to be paying bare minimum just for three of them $1,000 a piece. So 3,000 bucks for each one of those illustrations. It's also not easy to achieve that same kind of consistency with the AI. You know, like the it's getting better with like the node-based tools and AI's recognition of these things, but it's small details like that that actually make the entire website delightful and useful and set you apart from everybody else and help you create the the aesthetic that supports the message that makes you irreplaceable and unlike

24:58anybody else. >> Yeah. >> Yeah, I uh >> Before we continue, I have something really important, which is I did not know we could ask Michael to do things. So, can Michael, you put a cowboy hat on my head and like a toupee on Zach? >> I don't want a toupee. >> Yeah, you're going to have a toupee. And then, Corey, I want you to shave his head. Uh so, I want him to be bald. This There's no way this happens. He's just going to put like no on the screen or he won't even notice. So, okay, sorry. Corey, you were going to say something. >> It's a test. >> No, I was going to say like, on principle, right? Any one of these examples that we could give about Okay, today cloud code can't do this. It's going to be hard for anybody to do

25:37this. So, you have to have someone with the level of expertise and time to do that. That thing is going to be available for the masses in the next 6 months. Just That's like the rule of AI now, right? But, that doesn't matter because then in 6 months there's going to be another thing that AI is going to struggle with that if you want to take it to the next level and go above what the average person does, it's going to take that level of care, craft, expertise, and the know-how with the tool in to be able to make it work. So, there's always going to be this carrot to chase of Okay, now that everybody is doing these product UI illustrations and that's pretty easy to do. Now, they all start to blend together

26:22and what is going to be the thing that helps me stand out? Oh, now it's going to be these custom uh you know, painted um pictures, basically, that are like overlaying behind a I don't even know how to describe them. It's really Yeah, like lifestyle, artistic >> Like editorial art. >> pictures, editorial art. Okay, now AI is good at that. Now, the thing that's going to help me stand out is Y. Now, the thing that's going to stand out is Z. And so on and so forth. So, there's never going to be a point where where like, "Okay, AI checks the boxes on everything. My job is obsolete." Going from a good to great is always going to require the experts to come in and and get you there. >> Yeah, like there's there's problems

27:11that we don't know we have yet. Like we don't know about the problems that we need to solve until the AI starts to solve the problems that we currently have. Or like we were the ones who can solve it and then at some point the AI will solve it. And then we'll realize the pro- the next problem higher on the list. Like there there's never going to be a point where we're all just, you know, like Wall-E sitting in chairs and just like screens in our eyes. Well, I I shouldn't say never, but I we are long ways away from not having a problem that a human would then solve on top of the AI. Also, something worth noting when you were talking, Corey, about that example of, you know, there's the zero to 80 people and then there's the

27:4980 to 100 people. It makes me think of the the like, you know, what's the what's the rule? Uh about 20% of the val- 20% of the work producing 80% of the value. Yeah. It's the same idea and we hit this a lot with clients. Like the amount of times that we're meeting with someone or we're working on something and they're like, "I just want you to do this thing." And that is the 80 thing. And we have to be like, "Hold on. Just like I I need you to just be ready that we're going to ask to do something after that that seems more boring, seems less useful, seems more painstaking, and it takes longer, but we know that it will it will be the the highest producing potentially part of the process. But it's not flashy. It's not a new site

28:32launch necessarily. It's not even like a new logo or a new color. Like it's not even necessarily like the thing that's the most exciting and seems to make the biggest splash. It's like the dumbest smallest thing like AB testing a headline or rewriting your onboarding emails. It's like all These are like some of the most boring marketing tasks that can have the strongest amount of lift. And I think that's worth noting with with your example. Um Yeah. Uh okay, Zach. What else? What else you want to go through with this with this topic? Is there anything we left unsaid? >> Sure. I think one thing to touch on was So, that person who mentioned that I want to color a little bit more context and give the show where I think

29:15AI is actually extremely powerful and that is as a multiplier. Here's the issue and the difference that I want to point out. What this person had referenced was they said they created several different versions of a templated page that already existed. Meaning all they did was swap out copy and maybe some images, which I don't know if they had or if the AI was generating abstractions for it. Maybe they didn't even need to change out images cuz that's entirely possible, too. Or it could just be the same image and then they're swapping out the copy. That's going from one to 50 and adding a multiplier within there. If you multiply 50 * 0 or 0.5, you end up like you cut yourself short. Uh it's like yeah, you have 50 blog

30:02posts, but they're only worth 25 because you multiplied it by with a half-baked thing, you know? That kind of force multiplier doesn't work until you have a solid integer, a whole integer that you can then work off of. And if you're starting with nothing, AI is not going to multiply anything. It can't do anything with nothing. >> [laughter] [sighs and gasps] >> Yeah. It's like almost like a template, like an email template. You're like, "If I just send a million emails, I will make money." And you're just your email template is like just crap. It's like you're actually doing it worse It's actually making it worse by sending more emails. You need like an at least okay one to send to a million people. Better yet,

30:45have a perfect one to send to a hundred. Like you niche harder, target harder, better craft equals like a better result per se. Sorry Corey, I could jump >> People have actually always thought about this um for a lot of other parts of marketing like blog posts for example. Because AI has solved text a long time ago, right? People started cranking out mass amounts of blog posts. They were like, I'm going to create thousands of blog posts and I'm going to get millions of visitors to my site. And then the opposite happened. Google penalized them and then they got no traffic whatsoever and they were banned and their domain was completely shot. And um and people are still tempted to do this. They're like, "Oh, well, if I can write

31:30about anything, I'll just write about everything." So they write all these blog posts and then Google doesn't know what to rank them for, why they should be ranking. Um the blog posts themselves are bad if they're not ranking on the first page of Google. And that's that's because again, in order to produce something that is first page worthy and that is actually good and will actually get you customers as well, you can't just rely on the first thing that AI spits out for a blog post. You have to take the time to guess what? Take it from good to great. To go from 80 to 100. To polish it, to refine it, to add to it, to beef it up. And you have to keep doing that for every one of your blog posts, not just

32:12the first one. Um because yeah, they can like drag everything else down. Um so I like I like that example of like yeah, if you scale crap, you still have crap. >> Yeah. Well, I mean so they they did multiply by a whole number. It was just the wrong side of zero. It was like they they multiplied a three. It was three times or negative three times 50. So it brought them way, way down further than they could have imagined. Yeah, it's a multiplier. That's what it comes down to. It's not a You have to give it something in order to go up go off of in order for it to be really, really effective. It's got to have a a good element to then work with. Otherwise, it's going to flop.

32:59>> [laughter] >> And if it It's worse if you give it something bad for it to then replicate and work with. Yeah. >> Okay, what should people do? Give me Give me a I'm SaaS founder and I just hit 10K MRR and I can't I can't I don't want to pay for a service or another company to do it and I'm on my own. Give me at least like Give me like two things, okay? Like, don't do this and do this or something along those lines. I I feel like I I want to try to make sure we leave with like You can do this. Yeah, like it You're not doom and gloom and you we You'll die without us, you know. >> Well, uh I don't think that you'll die without us. If you have 10,000 MK MRR, you're doing something right.

33:47What's going to happen is you are going to realize the extent of your capabilities as a founder are limited, even with AI. Especially if you are if you care about the design, if you care about the copywriting, if you care about all of those polished details and you should because other people will and if you don't, then it's going to come across. Your time is going to be better spent elsewhere. That's what it's going to come down to. And until the AI is able to completely take this off your plate and produce something to your standard in a timeline that is of use to you because right now it can't. It might too might be able to at some point. You have to decide. What is more valuable? Your time spent

34:34working on this or is it spent better somewhere else? I'm going to wager that it's better spent somewhere else. Unless you want to go to design school and put in the thousands of hours that it takes to develop your eye and sharpen it and then be able to use that on your project as it stands right now. Cuz even with front-end design skills it messes up. It's not perfect. And even our site, it's not perfect. It's really, really good. It's not perfect. And you don't have the time to dedicate all the hours to this. You're not a copywriter. You're not a designer. And that's okay. Part of becoming a good founder is relinquishing some of that control to other people that are better suited to do it than you are.

35:30You want to get better at it. You as a founder got to get out of your own way. >> Mhm. >> I'm not going to tell you to go learn design courses cuz that is a giant waste of your time. >> [laughter] >> As it specially as a founder, don't do it. Don't go do that. >> Yeah. Yeah, of course there's always this question of taste too. So it's like, well if you just want to be able able to avoid making generic slop or just kind of just things that are average compared to the the average person study what's really, really good and I can get some taste get out there and dissect and analyze and really think about what makes something great versus just being good.

36:19And uh and have some opinions. Have some style. Have some um be able to notice and appreciate things that are outside of your own opinions and style. Uh and then practice going from 0 to 80 and then going from 80 to 100, not being content with just what's good and what AI produces in the one shot, but pushing it and going further and being able to understand and nitpick, okay, these are the things that it usually struggles with and uh these are the things that I need to fix every time and um that'll I think that'll help people avoid producing something that will just make them blend in with everyone else. And then also be able to appreciate the value of getting someone else to take it over the

37:05finish line for them. Um and when necessary and when you have the resources to be able to hire that out and delegate that. >> I My sense is that the AI there's good like the AI hesitation when it comes to buying software is not is going to be around for a while. Like even just before this week saw a random website, right? And I it was like obvious AI website.com, you know, it's like blue buttons, white background, like everything you would expect. Um and when I visit sites like that, my I have like a very innate just like automatic like discount that happens in my head. Uh and it's not because I hate AI. In fact, I I I used to really hate AI and all that stuff, but as it goes on, I I'm like, oh, this is this is amazing. Um

37:54but I still have that like is there anyone out there like hello, you know, like tapping on the glass where no one's behind the curtain and um the mess and the like little things in a website that make it obvious that people worked on it are actually still valuable and they do actually provide a lot of relationship to customers. >> We it's going to like you think of beauty as just like a superficial thing, but it is probably one of the biggest biological indicators that we have as a species. You know, if you look at somebody and you're say you're looking for a prospective mate, if you see someone that is like that's you know, overweight or um looks like they might have something wrong with them on the surface,

38:45it's an indicator they're not healthy. And so you are not going to waste your time with it. It's the same thing The same thing happens with food, right? Like if you look at a piece of fruit and it doesn't look appetizing or appealing, perhaps it's you know, the nicely like you look at a tree of apples and let's say all the apples are misshapen, even though they all look very ripe, but some of them are off. You know, like you have some that are kind of wonky. Maybe one started growing it over a twig and it's got this giant twig sticking out of it. And you see the one apple that is perfectly shaped or it's like it's the most well-constructed out of all of them. It's beautiful. It's beautiful. Which one are you going to pick?

39:31Which one are you going to trust with your with your calories and your life cuz it could be poisoned. It's you know, there's like I'm going to bet that one right there that looks the best is the safest bet for me. And we do the same thing with everything else. You know, if you it's a trust signal. It tells you I'm worth the calories. If you spend time with me, I'm not going to waste your energy and make you die quicker. >> Yeah. >> Maybe maybe the only exception to that rule is sumo oranges cuz those things are delicious and they look atrocious. >> What oranges? >> Sumo oranges. >> Sumo oranges. >> They look like they are literally diseased. >> Yeah, but that's like it's in the context of the signal.

40:18>> I'm kidding. I'm kidding. I'm kidding. >> [laughter] >> It's different if like if it's rotting or has like mold on it or fuzzies, you know? That's that's what the That's what the AI imperfections would be. >> It still does look very plump and ripe. Oh. >> Yeah, you could tell. It's It sends signals to your brain. It says, "That's good. I can eat that. That's okay." >> Those plump signals, you got to take those, Corey. Um okay, I know you guys gave some of those you're kind of like more practical thoughts. The I guess I I suppose my parting thought is there's a little bit there's quite a bit of fool's gold still with the AI marketing trends. And that is that you can do everything for everyone if

41:02you use AI in the right way. Your product can serve all these different ICPs. Your website can speak to all these different people. But that's really not what you should be doing. You should be making You should do be doing more for one ICP. You should be doing more for one, you know, group. And And basically taking what you already have and making it better rather than taking what you have and being like, "I could just repurpose this for these thousands of groups of people and be And you could just thousand X my my, you know, MRR or whatever." That's not really how it works yet, but it can help you do things pretty good. Like you can have something now and actually make it better just by asking, "Claude, how can

41:43I make this better?" It might actually help you. Um So, I think Yeah, that that's I guess my my main takeaway from the conversation is don't try to do more than you should just because it could go fast uh because it probably will hurt you more than help you. Any final thoughts? I know we're kind of over already over time. Anything else you guys want to talk about before we peace? >> I think I've said enough. >> You've said enough, yeah. >> Yeah, or no silver bullets, only a thousand gold BBs, and AI's not a silver bullet for copy, for marketing, for design, for development. It's pretty freaking cool. It's definitely there's a lot of golden BBs, but it's not going to be something that just completely automates everything

42:30once and for all. Um, at least not today. >> Yeah. Awesome. Great work, gentlemen. Okay, we will talk to you guys next time. Peace out.