Why My Marketing Career Needed AI—and Yours Does Too

Every journey into AI marketing is unique and this podcast has been me trying to document that story, but even then I forget to share a lot of the important elements that happened before and during the creation of this very podcast. And I wanted to share that story in a more robust way because I find that stories have lots of lessons. It's the reason why I like to even read biographies of famous business leaders because there's so many business lessons in it and you tend to remember it more. So I wanted to share an episode of that I just recorded on another AI marketing podcast called the AI Marketing Case Studies podcast with my friend Fernando, because he did a good job of asking questions to pull out all kinds of interesting parts of the story, as well as some tactics that I'm using on this show that are AI related, but are really helpful. And I think you'll be very fascinated in what I'm trying to do with this podcast. So take a listen.

Hey, everybody. Once again in another fun episode of the AI Marketing case studies podcast, where we interview marketers doing really innovative and creative things with AI. Today, I'm very privileged to have Dan Sanchez, who is an AI marketing consultant creator. He does a lot of AI stuff with, Social Media Examiner. He has his own podcast called the AI Driven Marketer. He has worked with Sweetfish Media with James Carberry. I wanna little learn a little bit about that in the process of the interview. Where are you where do we find you today, Dan? Where are you located? I'm near Nashville.

You're in Nashville. Okay. Perfect. And I am actually in a Starbucks at Queretaro, Mexico. Long story. But this was the only makeshift studio I could do deal with right now. And not the standard one where I have all the perfect lighting and set up, etcetera. So if you hear ambient sound of all the, you know, jazzed up coffee goers with their caffeine and their brownies, you'll know what it is.

So Dan, just for a little context for the listeners over here, run us through a little bit of, your background. So you, you have a, a pretty good experience there with marketing and with podcasting, etcetera. Tell us a little bit about how you got to where you are today from your start in this world of marketing.

I mean, like most marketers, I kinda stumbled into marketing. Like, who who goes to school and thinks I wanna be a marketer when I grow up? I literally just wrote a kid's book about that, so I needed more kids to know, at least my kids. But I stumbled into it from the art and design and kind of web design and again into digital marketing and became full blown marketer. And there was a time early in my career where I was at a a conference for a software called Infusionsoft, and they had this competition called the ultimate marketer. And they were doing these crazy automated campaigns, but they weren't graded just on, like, how much revenue they got, but the creativeness of it, how the efficiency of it, time savings. And I was I remember seeing this competition once, and I remember it was like a pivotal moment for me. So I was like, man, I I wanna be that. I wanna be the ultimate marketer.

And something inside me when I was in my early twenties was like, I that's that's my calling. This I am going to be the marketer's marketer and I'm gonna master all things marketing. So I did. And, you know, you take it a channel at a time, and I went really deep on every single channel until I had only a few left. And funny thing was is podcasting, like we're talking now on a podcast. It was one of my last channels, and that's where I ran into James Carberry, like you mentioned before, at Sweet Fish Media. It's the largest b to b podcast agency in the world. And he saw all the other marketing goodness I was doing, and he was like, dude, you gotta come join my team.

You are you are running circles around all these other marketers, and you're real you're not just good at one thing. You're good at all this stuff. Like, that's our audience. You gotta start getting up out there. And I wasn't even posting. I wasn't doing anything. I was like, oh, okay. I I thought everybody did everything.

He's like, no, bro. Like, I've interviewed a lot of marketers. You gotta come start talking about this stuff that you're doing. Because I was crushing it at a at a private university and we tripled enrollment. We automated everything. It was cool. But he pulled me in to start talking about marketing and got me started on LinkedIn and got me started on LinkedIn LinkedIn and got me started on his podcast, b two b growth and taught me the ropes. It's where I really like I think social and podcasting were like the last two big games I hadn't really mastered yet.

And he fast tracked me on that. I I picked up a lot of podcasting. I did hundreds of podcast episodes and then did really well on LinkedIn. Because I had I didn't realize I had a lot to share after hustling, after you know, when you find that core motivation, you know, that like lights a fire in you and you wake up 04:30 in the morning enough mornings to learn how to do the thing and then practice it during the day, you end up you end up picking it up. And then recently, it's just been AI. Because I I saw the writing on the wall maybe a year and a half ago, and it was just obvious. I'm like, this thing's gonna take over everything. That's it.

I'm putting all my chips in. AI is the thing. I'm starting a whole podcast about this. I'm reading every book I can find about it. I'm talking to every person who's doing anything cool with this on my show to learn everything I can. So I'm in full learning mode trying to get everything I can about this now.

Yeah. I was actually looking at, your podcast on Spotify, and your first episode was December 2023, which was yeah. That was pretty early on. Actually, I started, a meetup in Austin, in June of twenty twenty three, called the, marketing automation and AI meetup. And then Sharon DeCaro joined me as the co organizer, and now she's runs it. But that was really at the beginning of of this AI stuff. And, you know, I would say that that time from, like, about, you know, April, May, June to December, of twenty twenty three was those really early days. So I think that you were very prescient.

You were very you were you were at the right time to launch your, you know, the AI driven marketer, podcast during that time. So I'm lucky.

Yeah. After Sweet Fish, I joined a tech company, and that founder had the vision immediately once Chad GPT came out. He, like, changed the whole road map for his product and got all into AI in, like, spring of, 2023. Following along Is that an element of that company? Company long. It was called element four fifty one. It's a higher ed CRM and education company. But they were doing incredible stuff with AI way ahead of everybody else, and I think getting a vision from there. And then it it still took me a few months for it to, like, simmer after I left that to figure out, like, oh my gosh.

This is gonna be everything.

Well, you'll be you know, before we get into the AI, which, you know, obviously, this is the AI marketing case studies podcast. But, you know, from what you talked about, you know, how you decided, you know, all of a sudden that you wanted to be the best marketer out there. It was at this Infusionsoft conference, which now is Keap. Right?

That's right.

But their complex automation, you know, slash, CRM tool, etcetera, you and and then when James Carberry invites you to join, Sweet Fish Media, you know, that was when you said you were gonna learn, you were learning social media and podcasting. So you essentially learn the complicated stuff first. And then you went into the stuff that was like, I guess, you know, I'm sure this is at least seems pretty straightforward. Yeah.

Is very complicated. The logical stuff of automation and direct marketing, much more straightforward. That's why, like, Ogilvy, like, the OG marketer from, like, madmen era, he's like, I wish every marketer would start as a direct marketer first. I wish every marketer would be like Lester Wunderman, who was like the father of direct marketing. He's like, you need to start with that as the foundation. That's what I did. I Flint McLaughlin from Marketing Labs no. Not Marketing Labs.

Marketing Experiments was kind of like the guy who taught me how to think like a marketer. It's a very direct marketing approach, AB split testing, lots of value propositions, and all that kind of stuff. Doing paid media, all that words, SEO, inbound, drip sequences, lead magnets, and all the e all the marketing automation, even phone call and inbound call center stuff, like, all that kind of stuff was like, as complicated as that is, social media is harder. Like, at least I can set that stuff, forget it, and split test it logically and just find winners all the time. But in social media, great. You have a winner. Well, you gotta you gotta start again the next day. That's hard.

It's a the creative skill in that, I think. I can see why it went last. If podcasting's on the other side is is I think is one of the easier ones. The secret tool that I think most people don't understand about podcasting is that you think it's a content game. And I'm like, like, it is and it can be, But you don't if you haven't done podcasting, you really don't understand that it's the relational game that makes podcasting, like, boom, absolutely worth it even if you get zero content out of it. Because you'd be surprised how many people you can show up. Even authors of your favorite marketing books will say yes to being on your podcast. Even if your podcast has nobody on it, it doesn't really matter.

There's something about this, like, co creation. This is what I learned from James Carberry. He wrote the book on this topic of

I have the book that

he wrote. You have the book? See, you know. Yeah. Yeah. Content based networking.

Yes. Content based networking. It's it's so funny because when he coined that term, it was about maybe two or three years ago or something. He was talking about category design. I guess he had kind of come across the category pirates writings, you know, Chris Lockhead and Eddie Yoon and all that kind of stuff. And I thought that was brilliant. I thought, damn. That's that's really, really cool.

And, actually, he started practicing this thing where he would go to different cities and and have these, networking dinners. Right? You know, with people that had found him through the podcast. I went to one of them in Austin. That's how you

well, I'm like, you you've you've rubbed shoulders with James before when you started talking about the dinners. That was pretty before I even joined Sweet Fish. So I'm like, you you go back when you've you've known Sweet Fish for a

while. Yes. Exactly. Well, that's cool. You you've you've sort of had a very high pedigree of education in the marketing space. But let's get into your your your light bulb moment with AI. So you were like you said, you were working at element four fifty one where the where the founder really was doing some advanced stuff with ChatCPT way back, you know, as far back as, you know, you you started working there in February 2023. But what what so from that point when you were working there to your podcast when you launched it in December of twenty twenty three, maybe you launched it.

You started it before then, but your first episode hit 2023. What was what you what triggered that? What happened in your mind to to make you decide to go whole hog into AI? You know, a lot

of good things come out of really hard seasons, and that was a really hard season for me. I left element four fifty one on okay terms, but, like, they they pivoted. They hired me to grow an audience. They ended up letting me go and then just acquiring the audience rather than growing it in house. So that's fine. But then I was forced to go solo, and it was a rough transition. I ended up trying like, most people, you when you're forced to, you're like, well, I had enough people saying, like, Dan, now's your time. You've you have an audience on LinkedIn.

You should be able to make this work. And I was like, at the time, I was really trying to build a niche for myself around audience growth. And sometimes you go to market with a thing, with an idea, and find out that it just doesn't work. And it's not because the method doesn't work. It's not because you don't know how to do it. It's because people don't wanna buy it. And I was that was the thing I heard over and over again. It's like, hey.

It costs too much. It takes too long. And, the ROI just doesn't come back fast enough. And I was like, you know, it is a long game. Oh, and it it just took too much time. So I was kinda like, alright. Well, you know, after pivoting and trying out multiple things over and over again, that's that was in that season of December. I was like, man, I'm I'm making it.

I got some podcast clients. I got some marketing automation gigs. I'm making it, but dang, this audience group thing isn't working out. And that's when I was like, you know, this AI thing, I don't think I can make money on this right now. It's freaking early. But that's when you start wrestling with, like, what do I wanna be about? Every marketer knows you have to find a lane. Even if you you're you know all the channels, it doesn't matter if you're how good you are at all the channels I found out. What matters is you remembered for something.

When someone says, can you can't just be passed around as this normal marketer or normal this or normal that. You have to be known for one thing. And I'm like, this AI thing has to be it. I am I feel it. I like, my spidey senses are tingling. I remember having that feeling about marketing automation when I first saw Infusionsoft back in 2012, and I'm like, everything's gonna be like this. This drag and drop journey builder thing, this is going to be everywhere. And it did.

I remember feeling it about 02/2010 about WordPress or maybe it wasn't even earlier. I think it was 02/2008, '2 thousand '8 before WordPress was, like, the full blown CMS. It was just a blogging platform. I remember trying to build websites. I was becoming a web designer, and I was like, Drupal, Joomla, and WordPress. I'm like, this WordPress thing, this is special. There's something about this, and now it runs, like, 45% of the Internet. I remember I remember feeling it about TikTok before it blew up.

I didn't pay attention to it though. I'm like, man, this thing's gonna blow up. Some people are gonna make a lot of money off this thing. It's not gonna be me though. So I've seen it. I'm like, I've felt this thing so many times and I'm like, AI, I'm feeling it now. This is the moment. Yeah.

I started getting into WordPress in in 02/2007. Kind of around the, around the time time you were doing it. But, no, that is, that is really cool. It's like being known for that one thing, you know, and just kind of dedicating yourself to that. And, and just a comment about, you know, getting out there into the field and figuring it out. Like you were you first started to to talk about audience building. But it but you realized that there really wasn't a market for your content. And then, you know, you kind of fit you pivoted to to AI.

You know, that first of all, kind of hearkens back to the, you know, the whole lean startup methodology that Eric Reese talks about. It's and it also goes to something that I just saw recently, you know, was, Lorne Michaels, the founder and and manager, CEO of, SNL talked about how, you know, that he didn't have any grand vision for SNL. He they just went out and started doing it. Right? He had an initial idea, and they developed they started developing all the themes and all the new, you know, things that are that they're famous for just by getting out there in the field and doing it. So I you know, my comment to you is I think that you wouldn't have figured out that AI was the thing you had to do unless until you had started going out there and trying to make a name for yourself after the kind of disappointment and low point of of, you know, of your last, employment.

I mean, some of us call that providence, but I look back on it. I'm like, man, it was really good that I hit that point. Because now now I'm glad that it forced me to really think about it, start investing in AI early. And I wouldn't even say AI hasn't matured yet. We're still on super early days, like, super early. I'm glad I get to pine now pioneer it with Michael Stelzner, social media examiner, because he's also seeing the writing on the wall saying this is gonna take over everything. And we've talked multiple times, and he's like, this is really early. It doesn't have the feel of when social media was, like, really starting to, like, peak and Social Media Examiner took off.

He's like, but it will. And which is why I decided to work with him because, obviously, he's got much larger audience and way more resources. I'm like, I'm gonna run with you on this one. He's like, I could go solo, but I think it'll be better if we did this together.

So I wanna there's so many things I wanna ask you, but, you know, there's two general directions I wanna go in. I think I wanna start off with with your podcast. Did you just did you just start experimenting with AI and then talking about it? So I wanna talk about that, about your about the the, you know, how you got the content for your podcast. And then I wanted to talk about how you're using AI personally within your whole marketing process. But let's talk about the about the podcast, because you know, the, the latest episodes you have is, you know, customize your attributes, instructions and memories now to 10 X your work later, open AI's operator three, many and the need for marketers to lead and then co writing a book with AI. I'll have to talk to you about that as well because I actually co written three books with AI. One is Nice. My own and and two for some clients that I go straight for.

But let's get into, like, the, you know, how you're crew how you're coming up with the content for your podcast.

When I was working for Sweet Fish Media, one of the top marketing topics that was new to me because b two b was kinda new to me when I entered Sweet Fish, and I had to there was a few topics that I needed to learn everything about in order to understand b two b. One of them was account based marketing. I had to learn everything about that. And then another big one was thought leadership. I'm like, oh. And there was something about it. Maybe it's because I'm like that INTJ, Myers Briggs, like, Enneagram five with the wing four. And something about that was just really appealing to me.

I'm like, this is interesting. You mean there's whole groups of marketing that just get, like, the whole idea is putting out original thinking and ideas and leading people's thinking. And, like, that's the idea behind thought leadership. I'm like so I went and read every book I could find on it looking for a process and how to essentially become a thought leader or how to it I hate calling it that, but, like, it's still the appropriate term for it. You can call it authority marketing. You can call it point of view marketing. Whatever it is, it's thought leadership. So I wanted to learn more about it.

I had a hard time finding a process of what one would do in order to do that. So I ended up inventing one in the in the process of trying to figure it out. That became a really helpful thing for me that I now apply to AI, and I call it the thirty thirty thirty plan. Okay? Because every time you if you wanna become known for something, you don't have to become an expert. You don't have to be an expert in it right away. But if you start as a student and do the thirty thirty thirty plan, you could be gained authority of over and over time. And that's what I'm trying to do with AI marketing. It is 30 books, 30 interviews, 30 blogs.

Okay. Whatever topic you're trying to dive into, go and read every single book published on the topic. If there's more than 30, then it's too big of a topic. Niche it down further. Most topics, like subtopics is like within marketing, there's probably only 10 to 20 books on it. I promise, because I've done this multiple times with multiple subjects cause I'm a nerd. You could see a bookshelf behind me. I read a lot.

I've done it with multiples topics of marketing. And thought leadership had about 15 books. So I went and read all the books on thought leadership. Like, every one, even the unknown ones, even the ones that were, like, cheap throwaways that nobody buys, I bought it. I read it. You'd be surprised how fast you can read a book when you're reading them all on a very small topic, because each additional book after the first three only has an incremental amount of new information. So the ability to skim through them and just look for the new parts very fast. So it's very and this is the people are like some people definitely listen to this being like, 30 books? I haven't read 30 books in my life or some after since college, you know? I'm like, I promise it feels more like reading six books or five or six books.

And what's that in a course of a year? Not much. So if you know everything that's been published on it, then you have a pretty good understanding of the field, which builds into the next one. Do 30 interviews with people who really know what they're talking about on the subject. Probably call on interviews some of the authors you've read. Right? You're gonna be able to go way deeper actually having real substantive conversations with the people who actually know stuff. And after you've read all those books, you're probably gonna have some questions. You're probably gonna be able to go deeper than you would have in those interviews before because you've wrestled with the material. And then you need to publish some information.

You need to actually, like, work through it. And I I really like the process of, like, just answering questions that people have. In fact, I like ranking for SEO. SEO was one of the games that I played that I was I was pretty good at for a while. So I know the process of finding the keywords people search and writing authoritative blog posts on the topic in order to get ranked. Write those blog posts even if they don't rank. If you just try to answer those questions really well, you really master the fundamentals of reading all the information, interviewing the experts, and then answering the most frequently asked questions on topic. And you do this all out and kinda like post about it on LinkedIn while you're doing it being like, oh, I read this book.

Here's what I learned. Oh, I just talked to this person. Here's what I like. Hey. I just read a blog post put a blog post about my journey into thought leadership. Here's what I learned. Here's how to become a thought leader on LinkedIn. All of a sudden, after some time goes by, people are gonna look at you and be like, hey.

Dan's Dan knows some stuff. Yeah. Absolutely. Have some original thoughts to offer, people are going to listen and then thought leadership. So that's essentially been my process for LinkedIn and why there's unfortunately, there's no not a lot of books on this topic. As they come out, I read them. But why I use the podcast in order to learn, now I do less interviews, so I still will do some in the future. But I'm now publishing more solo episodes as I wrestle with things and experiment with the AI tools myself and publish my own findings via the podcast.

So when you when you interviewed these 30 people, was this for your podcast? Trying to think through.

It was about 30 between 20 or 30. The thirty, thirty, 30 is, like, there's nothing magic about the number, but it does start to put into perspective, like, more than five. Yeah. Like, two dozen. And and it's, like yeah.

Yeah. And it almost feels like as though, especially if you, like, read the 30 books and interview the 30 people, it's almost like an accelerator for the ten thousand hour rule that, what's his name came up with. The one who wrote Blake and all that Malcolm Gladwell. Malcolm Gladwell. You know, it's almost like an accelerator for that. I I actually love the thirty, thirty, 30 rule now that you mentioned it. Okay. Well, that's well, that's very cool.

So tell us a little bit about how you're implementing AI in your own marketing processes. If you're,

you know, like for example, I know you collaborate with, or you work with Michael Stelzer over at Social Media Marketing World. How you apply it over there? One thing that I've been working on for a long time is trying to figure out how to like speed up this whole podcast production system. So I spent a lot of time trying to figure that out. And I've only recently I figured out how to automate a lot of the preproduction in a really cool way. You could check out my showrunner.com. I built a custom GPT. It was one of the first things I did. I did it as a collaboration with Susan Diaz, who was talking GPT preproduction process.

And I I was like, And then over a weekend, I flipped it into a custom GPT, and we both worked on it and made it better. So if you ever wanna see that, I released the instructions for you so you can just totally copy it, rip it off. And I I promise, like, if you're haven't done a custom GPT yet, go to my showrunner.com. Just read the instructions. It's very clear what it's doing, and you'll learn how a custom GPT works. And this is a really robust one that I find get a lot of value. It speeds up the whole preproduction process. I use it for every interview I do now.

But now I'm, like, trying to interview trying to automate the whole postproduction process. Hence, like, why before we jumped on this call here, I was talking about how I use Zencastr. Right? It's a more automated podcast recording and then post production. Do I can publish the episode? But now I'm building a whole automated workflow so that it can take the podcast, because I I run a video podcast, the video feed. It actually plays the video on Apple. And then take that feed and then automatically grab the video from it in the title and upload it to YouTube, crop the Square podcast thumbnails, upload a unique thumbnail for every episode, crop it, upload that to the thumbnail image, take that YouTube video video or actually take it, transcribe it so I have the transcription, turn that transcription into a blog post, put that blog post in my WordPress site, embed the YouTube video into it, that becomes the header for that blog post, And then actually take it and run and scan it with chat GPT again for the transcript and be like, here's three unique ideas, then turn each of those unique ideas into a bunch of social posts. Right? So three unique posts per episode, and they just trickle out over time across most of the social networks. And that just runs on autopilot now.

Every time I publish an episode, bam, bam, bam, bam, bam, bam, done. Now I still have to before I publish the social post, I do have to look at those and tweak them a little bit. I haven't crafted a prop so good that I don't have to, like, test it a little bit. You know? So on Airtable, I just have a little button I push when it's approved, and then it takes it the rest of the way you know, that used to take a long time. The next part I'm still working on is the newsletter. I wanna integrate it so it pulls specific episodes. Only certain episodes has a little filter and then pre drafts it into the newsletter where I can then upload the images and send it out because I struggle to get the newsletter out every time even though I literally have a whole podcast series that's I literally record it so that I can repurpose it as a newsletter. But, you know, you improve these things over time.

For Social Media Examiner, I do a lot of little I do a lot of social media, of course. It's Social Media Examiner. And I've built a lot of of course, at Social Media Examiner. And I've built a lot of custom GPTs and projects just to speed things up. I have a series I like, every time I come up with a social post or a video, like, you gotta write all the captions for it. And you gotta write different captions for different platforms because they all they all want something a little different, different lengths, different types of hashtags, no hashtags. Mhmm. URLs, LinkedIn bio.

There's all these little rules for every social platform. So I'll upload a single social post or maybe the transcript for a short form video, drop it in something like a social media post writer, and it's just like, bam, bust them all out. Face here's the here's what you need to write for Facebook x, for LinkedIn, for TikTok, and it just gives it to me. I post them into Google Drive and send it off to the editor just to double check and make sure I didn't do something dumb. So it's a bunch of little critical processes like that that I use to automate workflows. I haven't really started digging deep into Social Media Examiner and starting to build, like, my podcast production workflow. But we actually have a really technical guy on staff that's that's building some of those more sophisticated automations that's he's much better at it

for me.

So I'm like, I'll let you handle that. I do a lot of education for their AI business society though, so that's where I get utilized the most.

Well, let's I'd love to dive a little bit deeper into your podcast production, your automated podcast production process. That whole thing that you mentioned is just it's it's earth shattering. And I don't say that lightly. It's like, that's what I need for this podcast. So I write podcast music.

Yeah. I haven't published my whole workflow on it yet, though. I started teasing it. Because I'm still in the process of like perfecting and getting it up there and running it, running through it multiple times in order to make sure that it works right before actually like, Hey, say, Hey, this is it. I've been trying to work on this particular problem for almost a year now because we all like, I don't know. You have a podcast. We all know that the distribution of the podcast is everything because podcast has almost zero discoverability. If you don't rank for some kind of keyword like b to b marketing, like b to b growth does for Sweet Fish.

Right? They that's how they grew. People were searching for b to b marketing. B to b growth came up first. It was one of the first ones in there. You know? Like, that's the only discoverability you get in podcasting. So you have to repurpose it for social. You have to get it onto YouTube. You have to do all these things.

And I know it's a massive problem for most people. And I've been trying to work on it. The problem is, like, Zapier can't do it. Make.com can't do it because the the video file is too big. You can't it's really hard to automate video because video files are gigabytes big, especially like this interview. This will be, like, three gigabytes worth of video once it's even once it's compressed and finished. So I did find a program that does do it, and it's n n .io. Fantastic little software.

N. N like the letter n, eight, the number, and then n the letter again. It's stupid name. It's definitely named by some developer that didn't have any sense for marketing, but you know what? They built an awesome product. So here I am talking about it. Right. So so that's your that's your it's like your a replacement for Zapier or for make.com. Right? Yep.

Yep. I don't use Zapier or make.com because I just Zapier is like stuck in the past. And then make so many people are using Make, and I just felt stupid using it. I'm like, I can't get anything to work in here. I feel like an idiot, and I'm pretty good at this stuff. I didn't realize how bad Make was until I talked to that technical director at social media examiner, and he felt like an idiot using it. I'm like, dude, if you feel like an idiot, like, make broken. This is dumb.

And I feel bad slamming make, but I'm like, dude, like, it really is a struggle to get that thing to work. And I will say n a n is hard. Like, this stuff's hard. It takes me an hour to three hours per node, which is all the little automation things that you set up to hook this, and it goes to this, and it goes to this. It takes me a long time to set up each one, but I can actually do it. It actually as time goes by, it gets longer and it gets better. I couldn't get it

done. Wow. So so so you record the podcast, and I know off camera, you we were talking about how with with within Zencast, right, you know, I use Riverside over here with Zencastr. You can kinda do some edits. You can do some automations to get take take out the umms and the ahs and the That's right. You know, the the uncomfortable pauses or what have you. Get rid of ambient noise, and then you can upload it, and then it could go to all the different podcast platforms. But, like, at what point or or like so, like, what would you use an end for in that process? So so because you wanna are you creating video shorts on your podcast?

You don't use it until after the podcast is live on all the podcasting apps. Gotcha. Zencastr takes care of you from recording to publishing on the RSS feed. Right? And every podcast has an RSS feed. You hook that RSS feed up to Spotify and to Apple and all the other podcast apps out there. Right? And every time you update the feed, it's what happens when you publish it on your host. Who do you host with? Like Libsyn, Blueberry? I'm using

the createcreators.Spotify, the one that used to be Anchor. Yeah.

Yeah. The one that used to be Anchor. Yeah. So Spotify puts out that RSS feed, and they have their own in house system for for video or for for publishing to Spotify. But they do publish an RSS feed that then goes to Apple and all the other ones. Now Spotify kinda hides it from me, but you can find it. Because if you get that RSS feed, you can hook that feed up to something like an automation tool, and it just waits and it's checks the feed every hour. That's what Apple does.

Apple's just checking your feed every hour to see if there's something new. Once it finds something new, uploads it to Apple. So and that's how that's how I kick off that tool. RSS is a great way. RSS used to run the world of Internet. Remember that when back when people used to, like, subscribe to my RSS? That was a long

time ago. I would subscribe to RSSs of all these blogs. You know, I would do it. Just go to my RSS reader and just see all the latest blog posts.

What was it? FeedBurner? We always use we all use FeedBurner to aggregate the feeds. FeedBurner. And we all subscribe to things via Google Reader or something like that.

We're telling people how old we are here, dude.

Back in the Internet marketing days So

so I think I I I so I'm doing it kinda backwards then, Dan, because what I do is I'll I'll download the, and I I don't even use the account level within Riverside that gives you the transcript. That's another level up. I just used the first tier paid version. I'll download the video. I'll put it into Descript to get the transcript. And then I'll I'll but then I'll get the video shorts from Riverside. And then with each one of the video shorts, you get the transcript to do a social media post. And it just it seems a little labor intensive compared to what you're doing.

But what you're saying is after you get it off onto the RSS feeds, you know, and you and you get it onto Apple or to Spotify or to Google Podcasts, etcetera, then that's where n h n eight n comes in to do its magic. And so are you creating video shorts for YouTube shorts or Instagram? It doesn't create

n a n doesn't create shorts. It just comes up with the text post. And because I have a graphic that loads with every single episode, which I I I put more and more time into because having one good graphic per episode, because that YouTube thumbnail is so freaking important. Right? But I'm like, if you're gonna put that much time into the YouTube thumbnail, you might as well I design it as a square, but with, like I designed it in mind that it's going to get cropped from the top and the bottom and become a horizontal image later on. So I put a lot of time into that that image. If you go to my website, like danchez.com and scroll down, you'll see the images and you see, like, they put more time into it than normal. And I use a combination of MidJourney and Photoshop, those AI tools, in order to make those. So I cropped I do that.

Now NAN does do the resizing of it because it's way too big of an image for YouTube. It has to resize it, then it crops it, and then it sticks into YouTube. So that way I don't have to spend extra time uploading the video, writing the title, copying and pasting to the description over. I just automatically loads in the whatever the podcast description as it goes to YouTube, adds it to the podcast playlist for my podcast on YouTube, and then sets the thumbnail. So that's done. I don't have to deal with YouTube. And then the blog is another thing. It'll publish it publish the blog post.

And it's just using the transcript to make the blog post and then the transcript to make the text for the social media post. And on some of the social media posts, I'm also using that podcast cover image as, like, a graphic to post with it. Because, again, I put a lot of time into it, so it becomes a nice little social image. That way I'm getting all the use I can out of that one freaking image. Right? So that's what I'm doing.

And why why are you spending so much time on the, you know, because you may you did mention that people spend a lot of time on the on the YouTube thumbnails, so you might as well do the same thing with your podcast. What's to you, what is the importance of that? The the the episode image? The click

through rate. I mean, you want more people to let have a first listen and a lot of people like YouTube, YouTube is driven by thumbnail images. That's why mister b spends about $10,000 per image and he makes multiple images per video. The image is almost as important as the video. It's

like the hook in a LinkedIn post or in a blog post. Right? Or a title. You know? What is that? What is the thing called? I think it was maybe was it Eugene Schwartz who who talked about this or Joe Sugarman who talked about this slippery slope where, you know, 80% of the people will see your your headline. Joe Sugarman, man. To at least read the first line, and then the first line is gonna get people to read the second line. So you gotta spend inordinate amount of time on the headline, the the opening hook. If you're talking about LinkedIn post or the thumbnail or your, you know, episode image, that's that's really where

the the episode image the point that's using Joe Sugarman's language. The the point of the episode image is to get them to read the headline. The point of the headline is to

get them to click play. That's it. Exactly. This is Joe Sugarman. Okay. Perfect. Perfect. So I would I noticed that one of the episodes that, and I'm gonna I'm gonna listen to this as soon as we we hang up over here.

But is how you're co writing a book with AI. I've actually co actually did two ghostwriting gigs for two clients of mine from Mexico. There they have, they both have outsourcing companies providing, you know, developers and other things, other types of, skill sets to people to companies in The United States, and they want to expand their marketing efforts in The United States, and they thought that writing a book would be a good idea. And so what I did was I, you know, interviewed them, and I and I leveraged AI to come up with initial drafts of book. And then, of course, you have to have a human in the loop, you know, the you know, as a Russ Hunnabury said is Russ Hunnabury said that AI is kinda like you're a ghostwriter, but you gotta go in and you gotta edit it, etcetera. But I was wondering if you could tell tell us a little bit about your AI book writing or co writing process that you talked about in your episode there.

Yeah. I do find the better you get at using AI, the better it can write for you. And it works just like a ghostwriter. Like you've done some ghostwriting and working with it. I've I've worked I've had some friends as ghostwriters. I know the process. Like, you come up with the idea and they have to interview you. They have to suck it out of you.

Right? And then you give them you first, you give them the baseline story like, oh, here's the idea. Here's the top line. They're like, oh, cool. Well, that's not enough. Now I have to go and find out what the tell me more about that. That seemed like a critical point for you. Like, that's why I even started with that story about the conference and me wanting to become an ultimate marketer. I know that because AI and I have sat down, and it's interviewed me and asked me, like, hey.

Tell me more about that story. It seems like that was a critical part, and that's why I tell it now because AI has pulled it out of me being, like, you should talk more about that. I was like, dang. AI is a great coach or consultant. The funny part is you have to get it to prompt you. You have to, like, almost, like, lead like, lead it help it lead you, and it's really good at that. So what I'll do is I'll I'll give an initial pass to the outline. And I set all this up in a chat GPT project.

This is the best way to do it right now. Because in the project, all the separate chats you have remember conversations from past chats within the project. So if you have one project for your book and have one chat per chapter, it can maintain much more of the context from chapter to chapter, which you weren't able to do until that projects feature came out recently. So that's a new thing. Before that, I'm sure it was a lot harder. But now you could be like, okay. Here's the general idea from the book, and here's some of my basic content. I'm gonna load that in as instructions and material in the project.

And then from there, I just I take a first rough pass at the outline. I'm like, hey. Here's my first rough outline, chat GPT. Help me flesh this out based on what you know we're trying to accomplish here. And it'll take a pass, and it'll flesh out the I outline a little bit more. I'm like, great. I'm gonna go ahead now and give you all I'm just gonna word vomit. I'm gonna dictate all the stuff that I know that needs to go into this chapter.

I'm just gonna use this outline and talk. Right? Because there's nothing easier than just looking at a bullet point list than just, like, word vomiting. Everything you know about it as you work through the list. Exactly. Right? And that's what you do with the ghostwriter. That's exactly what a ghostwriter is gonna do is it's just they just needed you to talk, and they're gonna take notes while you talk and go back, and they're gonna record it and listen to the recording and painfully go through it again until they get a coherent draft. So I get it all that. And but before I have it write a draft for me, I'm like, okay, ChatGPT.

I've just given you a lot. Ask me some questions of details you need still in order to make this good. And that's where the real magic happens. So I have it ask me additional questions based on the stories and the the test the the case studies that I'm giving it. And then it asked me multiple questions, And then I go through just another round of dictating, and I answer them all exhaustively as much as I can. It's just one long freaking prompt. Bam. And then it's like, okay.

You've done a good job of filling these in. And I'm like, is there anything else? And it's like, it'll either say, like, oh, I think it might it always asks some questions, but I could tell when it's got it. I'm like, okay. Write a draft. And each draft is only about a thousand words at a time. So literally, it's painfully writing a thousand words at a time of which each chapter for me, I'm walking through that whole process two to three times, sometimes four or five, depending how long the chapter is in order to get the content out. Right. So If the chapter's about four or five, those words not the whole outline for the chapter.

I'm working on just a section of the chapter at a time. Bam. Bam. And I have to go through word, like, word dump and then question and then answering the questions. And then it come up with the draft, and I might edit some things. But it honestly, usually, when you walk through a process like that, it's pretty freaking close. Like, you don't have to edit it a lot if you've prompted it, Raul, and you've given it all the context to work from. And that's just with four o, man.

When four four point five comes out sitting here, like, this can be flying.

That's amazing. So getting back to so with a project is a project similar to a a custom GPT, but it just has that bigger context window so that it could remember previous chats or It is remarkably close to

a custom GPT because you get instructions. You get to load an instruction with it and you get to upload documents for it to reference. So the one difference is you get to contain imagine with each custom GPT conversation you have, every new conversation you have, it doesn't remember any of it. But in a project, it groups all altogether, so it has the context of all those. Now ChatGPT is getting sophisticated to where even if you're having, like, a plain Jane vanilla conversation with ChatGPT, it can still reference past conversations you've had or at least recent ones even if you don't have them in a project. But by containing them in a project, it knows, like, hey. These are related, and you can call you can force it to call back and be like, hey. Remember when we talked about in chapter two? Can you pull some of that material onto this and reference back to it right here? It'll be like, okay.

And it does because it remembers. And sometimes it'll do it without even being asked. It'll just be like it'll just start bridging it together. That's really fascinating. I'm hoping the context window gets larger because that becomes a limiting factor for something as long as the book, but it's still pretty good.

So so in in those instructions and the documents that you set up in the document, I think you mentioned this before, but I wanted to just kinda, like, clarify this for my mind. Do you upload samples of your writing, or did you maybe try and write an initial chapter or section of a chapter by, you know, by yourself as a say, hey. Listen. Can you model all your voice and style and etcetera, speak at an eighth grader to fifth grade level based on this or whatever? Or how did you set the instructions for the voice?

I had already done that long ago and made my own custom Danchez GPT, like, writer. So I just took that and took some of the instructions from that and stuck it into the instructions for this book project. But I gave it more context. Like, hey. This is what we're writing for. Here's where I want the style to go. Because how you write for a blog is just different than a book. Right? So, like, I I modified it for that, but I took a lot of my key writing style tendencies and just transferred it over to the the instructions for the project.

So are you were you doing did you do any research, like, any third party research, you know, or or are you using any references to studies or to other books, etcetera, for your book? Or or how are you handling that?

How are

you getting cheap yeah.

So this book in general is about the five types of AI marketing. When we're pioneering new stuff, the new thinking needs to come in and help, create categories. I find the the process of creating categories or taxonomies helps clarify what the land looks like. If you ever played Settlers Catan, you know it's organized into, like, different hexes. You know? You got you got your forest over here, your mountains, your rolling hills, your wheat fields, like, every new land, and that's what we're dealing with. Like, we're dealing with a lot of unknowns. So I'm trying to come in here and bring some clarity, shed some light in it by grouping it so that when we can deal with it with groups. I mean, that's why you have the whole animal kingdom are broken up into groups even though there's some oddballs that, you know, cross lines sometimes as as with all taxonomies.

So the five types of AR marketing is what the book's about. And it's essentially me trying to bring some clarity. I've already published LinkedIn posts about this. I've already published blog posts. I've already done episodes about this. But I'm trying to bring it and bring some more credibility to it and some more case studies to give more examples and to package it all into a book so that people don't have to go and find all this stuff everywhere. They get it all in one nice little tidy package, and they can listen to it or read it. So that's the the purpose of the book.

I'm trying to remember. I like somewhere in there, I'm like, wait. What was this question again?

Yeah. It was the external or sources, like third party sources

Oh, third party sources. That's right. Yeah. I of course, I'm leaning on a lot of personal experience. The early chapters are, like, how I got into it. A lot of the story that I've shared here Exactly. Is part of those setting it up. I do a lot of time defining what AI marketing is, what it's not, other people's definitions.

But then I have to get into the five types, which are going to be like a more traditional business book where I need case studies. I have lots of case studies because I did a lot of interviews. So those I'm using first. Those are my primary sources. But it's not enough, and I need more. One of the great tools that just came out that I'm using right now is DeepResearch in order to go find them. So I've been using DeepResearch from Google when it first came out. But now that OpenEye has got one, I might start using that one because it's deeper and better.

So I'm using deep research. So I'm like, hey, I AI. These are the types of case studies I'm looking for in this these kinds of markets using AI in these ways. Go and crawl the Internet and find them for me and link back to the sources so I can find them. Google is really good at pulling a bunch of them for me. Also, because I know where a lot of them are, they're on OpenAI's website. They're on Google's website because they wanna showcase all the best webs best possible case for these things. And, of course, some of the types of AI marketing that I have are a little thin because it's, like, one of the types is AI analysis and forecasting.

But that's still a very premature field right now. We're not not a lot of marketers are using it for analysis and forecasting. There's just little bits and pieces of it, so that chapter is gonna be a little light. I'll have to come back and update it sometime. But from the most part, a lot of them are pretty easy to find case studies. So I'm using Google deep research to go and find ones that I can't find. I'm also just finding ones from my from my own podcast and then supplementing some with, like, from the big AI companies that have already have pretty solid published case studies in order

to both. Well, that's very cool. So when is the book coming out, Dan? It's a

good question. I was gonna try to sprint it and get it out by the February. And then I was like, part of me wanted to spend a little bit more time getting a little bit more case studies first hand in order to do it well. So I slowed down a little bit, and it'll probably end up being somewhere around this summer. But I'm really looking forward to getting it done. And it's not like it takes that much time in order to put together with the process I just outlined with ChatGPT. But I wanted more first hand case studies in order to pull from.

Yeah. Yeah. Yeah. And I'm sure you're gonna do the, like, the prelaunch, you know, get a prelaunch list together so that you can have people who can who will already have read the book and will you ready for the reviews by the time the book is launched, etcetera? Maybe. Yeah.

I hate that process so much that I'm probably won't do it. I should do it, but I don't wanna do it. I'm just like so much work. I just wanna put it out. I'll even give it away. Like, I really I just wanna put the information out there. But part of me is like like, if you don't, I know, like, if you don't charge money for it, people won't read it. You know what I'm saying? It's like, if you gave away a course, they they wouldn't they would never just watch it.

So you have to charge something for it. That that way, you get some buy in to actually, like, consume the material. But I'd rather just publish it on my podcast for free. That's what I'd really like to do.

And the time we have left and look at this. It's forty seven minutes, and I can't I can't believe it's been forty seven minutes. It feels like it's been fifteen to twenty minutes. Right? Tell us about what you're excited about right now in AI driven marketing.

I if there were no more innovation coming to AI, we would still have four to five years of like breakthroughs of how this stuff's gonna impact marketing. And I mean, like, no more models come out. No new stuff. We just have what we have right now. Four o is as good as it gets or, like, the o three minutei thinking model. It doesn't get any better than that. We would still have years of of of innovation of what we have to come because we're barely keeping up with what we currently have right now. Yet new stuff's gonna be coming out next week or in a month.

It's coming out so fast. So I'm just excited because I'm like, man, now's a good time to be a marketer. There is just so much new stuff coming out. It's going to lead to so much innovation. Some people are crying about it being limiting creativity or even making us lazy. I'm like, man, I feel like I'm more creative than I've ever been before. Being able to write books, I'm like, dude, I with the way video's going, I'm like, I'm gonna make my own documentary this year. This is gonna be a thing.

I've already put out a whole music album, and I'm not a musician. And I actually like listening to it on a regular basis. I did that just this last November through Suno, that AI app. It was fantastic, And I listen to it all the time. So, like, the creative ability now to mix and match and all these things and put out really cool things is gonna get better. And the people who are not afraid to wrestle with it and come up with good stuff, and as the models get better, the projects will get better. Like, someone's gonna make a blockbuster movie in the next couple of years just with AI, and it's gonna be fantastic.

Yeah. Actually, I I said this to somebody, a friend of mine, that, 2023 was kinda like the, the Pearl Harbor moment for marketers with AI because at first, we were shocked by it. So many of us were laid off because of it. Agencies either had to pivot or close their doors. You know? Like, content marketing agencies literally had to reinvent themselves. But then after we got over the shock, after marketers got over the shock, I think, and this is and I've I can witness this by all the people that I've interviewed in the short lifespan of this podcast that marketers have been on the cutting edge in terms of finding innovative uses for AI and marketing, just like you mentioned right now. Yeah. So great.

Well, listen, Dan. You know, there's the obligatory thing about where can people find you. I know where people can find you. Dan, Danchez Sanchez. So there's on on LinkedIn, you have Dan and then brackets Sanchez. Sanchez.

My name is Dan Sanchez. My friends call me Sanchez. And, dansanchez.com was taken, so I had to figure out something. And that was a nickname my friends have called me for a while. So that's what I go by.

Well, it's fantastic. And your and your podcast is the a driven AI driven marketer?

That's right.

Master AI marketing to stand out in 2025 when you can find it on all the major podcast platforms. And YouTube. Thank you. And YouTube. Okay. YouTube. Perfect. Perfect.

And you'll be speaking at Social Media Marketing World, coming up here, March 31 in beautiful San Diego. I remember going to a traffic and conversion conference in 02/2018 in San Diego. And as we were leaving, there was all these other Internet digital marketers flying in. I'm thinking, wait a minute. You're a little late for traffic conversion. No. We're coming in for social media marketing world, which is even bigger than TNC. Yeah.

It was even bigger than TNC. I was like, oh my gosh. I should have stayed another week and sat together for that. But you're speaking on at the conference, which is fantastic. And it's just been a privilege to have you here, Sanchez. We both have Hispanic last name, Sanchez and LaBastida. So, that's good. And we will see everybody in the next episode of the AI marketing case studies podcast.

See you then.