Loading summary
Transcript97 lines
- [00:00]
A
I made a mistake long time ago building my first agent, trying to make it all in one. You end up with a lot of inconsistencies and drift and hallucinations.
- [00:08]
B
The reason why we're building them different than most people is because we're operators, not marketers. We understand that the agents need a job description. Micro tasks in a macro project that they need to have. KPI's Perry Belch said this. In this next decade, we are going to see the first billion dollar 100% robotics and AI run company. Welcome to Follow the Yellow BR Road, the show where online businesses learn how to turn clicks into customers and growth into real scale. I'm your co host, Emma Rainville, the wizard of ops, helping companies transform chaos into systems that actually run.
- [00:45]
A
And I'm Mitch Barham, the wizard of ads, the guy who knows how to turn paid traffic into predictable revenue. Together we break down what really drives profitable online businesses. From traffic and funnels to operations, scaling and everything behind the curtain. Because getting customers is only half the battle.
- [01:01]
B
The real magic happens when ads and operations work together. So if you want smarter traffic, stronger systems, and a clear path to scaling your business, you're in the right place. Let's follow the yellow brick road. I'd like to chat about the digital workforce stack framework and just kind of break down the first layer. We have three layers. The first one is the operator. The second one is the orchestration layer. The third one is the execution replacement layer. But today I'd like to be very specific and chat about the operator. Okay, so the actual agent.
- [01:40]
A
This will be an interesting thing. You can look at the operator as your LLM. It's the worker, right? Because like technically Claude's doing the work or open a chat GPT, open AI is doing the work, or Gemini, blah, blah, blah. However, the operators are very specifically defined. They're not unicorns that could do everything.
- [02:05]
B
Are you sure people are building that? They're like, here's my marketing assistant. It posts my Instagram, my Tick Tock, my Facebook, my LinkedIn and my ex. And it writes my copy and it writes my emails and it writes my ads and it writes my book.
- [02:25]
A
Yep. And by God, I could do landing pages too.
- [02:28]
B
And by the way, it lives in projects in Claude and I trained it with Dan Kennedy. So it's so smart.
- [02:33]
A
It's the best. Yeah, yeah. No, because even as humans, those are really hard to find that kind of person. And then as like we've all made like, okay, full transparency. I made that mistake long time ago building my first, you know, agent trying to make it a all in one. You end up with a lot of inconsistencies and drift and hallucinations.
- [02:56]
B
So can I tell you, when projects came out in Chat GPT, Perry, Belcher and I had just launched Make It Real, which is where we brought people who had a business idea and who were like, I have a business, but they didn't have an LLC or they didn't have a bank account. It was almost, but not quite. And so he say all the time you have this business, but you haven't birthed the baby. You're buying diapers, but you haven't birthed the baby. So come for three days and we'll birth you the baby. And we come up with their messaging and all this stuff Anyway, so one of the things that he taught everybody was how to build a ChatGPT project this long time ago. I promise this, come full circle, a Chat GPT project, and then take that CHAT GPT project and make it your copywriter. This is back when you didn't have agents, by the way. You. You would go in. Operator didn't even exist yet.
- [03:49]
A
No, it was early open air, right?
- [03:51]
B
And so like early open AI. Everybody thought they were on the brink of we're not Even in the 3% of what AI is going to be and people don't even understand that. But anyway, so we had this whole system, right? And it was like, build out this project and this is how you're going to write your copy. Because back then everybody's writing their copy with Chat GPT because it was by far the best. Claude was brand, brand brand new. Gemini didn't really exist yet. If it didn't, I didn't know about it. Jasper was old news and this is what we're doing. And then Operator came out.
- [04:26]
A
Mm. I remember how jacked you were about Operator, dude. I mean, it was so cool back then.
- [04:32]
B
I wasn't jacked about Operator because of Operator. I watched Austin Armstrong give a How to for Operator and I remember how excited I was. I took his exact video and I refilmed it and sent it to Ignite. I've told him this, by the way. I've ripped him off more times than anyone else.
- [04:54]
A
Yeah, I've been on those calls when you said that.
- [04:56]
B
I always. I always say that, though. I always say that. I very publicly say that quite regularly. While I did create this for you and I did change it a little bit, I. This is like, if you go watch this guy, he literally is brilliant. This is what I Took this from literally prompt almost exactly just I swapped things like podcasts for conference. Anyway, it doesn't matter. I remember it like it was yesterday. It was January 14th. We were in Vegas, Perry and I. We were at something called Turning Point. It's where this is before the shot to the neck, but it was called Turning Point and it was a mastermind for our chapter leads for Ignite. We had brought them all out right before Affiliate Summit west and we had spent three days with them in a little mastermind on their businesses is thank you for being chapter leads and two years ago. And so I'm watching it on the treadmill of like I think it's Hidden Valley Ranch. It's an area in Vegas that's like up by. Up by.
- [06:01]
A
Let's think of the actual branch.
- [06:03]
B
I might be saying it wrong, I might be calling it something it's not but let me know in the comments guys. But I think it's Hidden Valley Ranch or Valley Ranch or Green Valley Ranch. That's what it's called, Green Valley Ranch. And it's up by where he used to live in Henderson. Anyway.
- [06:16]
A
Oh, okay.
- [06:17]
B
So yeah, sorry. Hidden Valley Ranch. Whoops. So Green Valley Ranch and shitty hotel. I can't remember. I think it was like a Marriott Residences and I'm doing couple miles on the treadmill before I go drink beer. I still drinking back then. And so I was going to meet the Ignite chapter leads over at the bar, Sha Stevenson's waiting for me and I. I couldn't leave the treadmill because I was watching Austin Armstrong. So long story to get to the point when I saw that I. It all fucking came. It was like being in a K hole. It all came together for me. I hadn't done ketamine yet in my life, but it all came into this beautiful like oh my God, Perry was right. In 2017 I was in Cabo in Mexico and I didn't hear his talk because I was pissed off at the person running the event.
- [07:10]
C
But.
- [07:10]
B
But I watched it later and he had talked about Automations and he wowed the entire room. And then he was supposed to talk for 45 minutes was three hours on the stage. He was. And so he talked about Automations and he was like, you know what? If you guys are wowed by that, that is so old news. He was talking about Zapier and everybody was wowed.
- [07:30]
A
Wow.
- [07:31]
B
2017. Well, yeah, okay, 2017, nobody, nobody was using that. Nobody was using that. That's true. And so then he started talking about AI and Jasper Was new, new new new new new new new new. May have been 2019. Actually, it was 2019. Excuse me, I. I misspoke. But he was talking about Jasper and Jasper was new, new, new, and you could write AI and it was the worst copy ever. And he was like, you need to know how to use it now. You need to know how to use it now. Doesn't matter that it's not good. You need to know how to use it now. And you need to know how to connect it now. Because the next thing is going to be robots using AI tools to complete tasks and they are going to be in businesses. And we are going to see in this decade, in this next decade, the first billion dollar Perry Belch said this. This is all recorded. We are going to see the first billion dollar zero employee, 100% AI agentic or I think he said 100% robotics and AI run company. And I was watching Austin Armstrong demo operator and watching it go into show notes and pulling podcasts that were similar to his demographic, and then writing a email to the podcast host on why he should be the next guest on that person's podcast, making a spreadsheet of all of these podcasts and then going into his email and then emailing the podcast host and oh my God, it all became real. And I was excited about Operator because I knew it was the very first step to his prediction that everybody was wowed. But I think most people thought he was crazy in that room. This is before people could use AI for anything. You couldn't use AI for anything.
- [09:30]
A
Right.
- [09:30]
B
It couldn't write a freaking email at the time.
- [09:32]
A
I guess. What, okay, didn't exist yet. 22. But it was technically open AI was
- [09:38]
B
there OpenAI was 2020, I believe it was during COVID Yeah. That it became a thing. People were signing up for it, I think in like end of 2021.
- [09:47]
A
Right.
- [09:48]
B
Jasper was it before then. And then 2022, like everybody was using chat GPT. Not everybody. Excuse me. 0%, 1%. 0. 1% of the population was. But everybody I knew was using chat.
- [10:02]
A
Everybody we knew was using that.
- [10:03]
B
Right. I was.001 of the first users for Chat GPT because Parabell. Anyway, so that's why I was so excited about Operator. It wasn't Operator and it wasn't Operator's capabilities. I thought Operator was kind of dumb and it was difficult. It would stop working after the fifth round and you had to reprompt it. But I knew it was the first step towards the possibilities of what we're Talking about right now, and that's building AI agents to go into your organizational charts. And we're building operators. And the reason why we're building them different than most people is because we're operators, not marketers.
- [10:40]
A
Right.
- [10:40]
B
And the only people teaching you how to build these things right now, our marketers. And we understand that the agents need a job description. We understand the agents need micro tasks in a macro project. We understand that they need to have KPIs, they need to understand their deliverables, they need to have a culture, they need to have values, they need to have precision of a chain of events that will happen all in the right order in order to produce a deliverable in the form of a project, not in a spreadsheet. That was a very long story of how we got here. But I mean, that is why when operator happened, I was so nuts over it and was screaming about it and nobody wanted to use it because it was crapped. But all I kept hearing in my head was, it doesn't matter if you think that it doesn't do a good job. You need to learn how to use it now before it does, you need to learn how to connect it now before it does, you need to look, go buy a robot arm. I remember he said this like two years later he said, go buy a robot arm and learn how to program that robot arm. And when you get a robot, you'll be 90 of ahead of everyone else. If you want your children to learn a skill that they are going to be able to use to. Don't allow them to go learn to be a plumber. Don't allow them to go be an H vac person. Don't allow them to go be a marketer. Don't allow them to go do anything except for work on robots. Learn how to preventive maintenance robots. The part that robots can't do or that you wouldn't want a robot to do.
- [12:27]
A
Right? Yeah.
- [12:30]
B
And I just. Man, it was like. Has. Has something ever hit you that you kind of knew was always there, but then it's like if you found out the tooth fairy was real.
- [12:40]
A
It's not. What are you telling me right now?
- [12:44]
B
You know? Damn it. But you. You know what I mean? Like if you.
- [12:49]
A
My teeth. Yes. No. 100%. Interrupt this podcast to ask you to do a huge favor for us.
- [12:54]
B
Huge.
- [12:55]
A
We drop so much knowledge to help you guys. But you're not going to be notified unless you hit that subscribe and the notification bell to be notified the minute we drop a new episode to Help you grow and scale your business.
- [13:07]
B
Because you never know when we're going to be taken off YouTube.
- [13:09]
A
You have no idea. Like, I could all of a sudden be like Liam Neeson with a particular set of skills.
- [13:14]
B
Also, we read every single comment.
- [13:16]
A
So leave a question or comment. We will reply to it and maybe we'll even talk about it in a upcoming episode.
- [13:22]
B
Probably. If you troll us, we will definitely talk about you in an upcoming episode. Thanks.
- [13:27]
A
I think the way you felt like, what you saw with Operator, the first time I used chat GPT, I think 2.5 or 2. Whatever it was the first iteration. And then Mid journey, quick after 2.5. Yeah, I literally called all my team into my office. You know, I'm dumb and have in people, in person employees, and was like, you guys, like, we're all cooked. I don't think anybody at that point outside of us, probably a small handful of people. I was super excited because, like, just the efficiency and everything we're going to be able to do. But at the same time, I was like, yo, we're all cooked because we're. This is going to do everything for us. Like, I saw it way back then, and my employees to this day still keep going, oh, my God, you were right. I'm like, yeah, like, it was not that hard to see where this was going to go, that we were going to turn these things into actual workers. Operator. Yes. Was huge. Because you saw the worker.
- [14:27]
B
No, I saw the completed projects. Well, yeah, done by, like, here's the
- [14:33]
A
worker that was doing this by an autonomous being thing, going out and doing this stuff that you told it to do. Which now is the way that inside of our framework, the Operator tier, that's, you know, precisely defined to go out and do this work. Like, it's wild. Wild where it is all stem from. But yeah, I mean, you were gonna ask a question that I interrupted you, because that's what I do. But I was super excited too. And yeah, I was. I was very stoked when this all first came out. And like, you know, mid Journey, we were also, like, prompting everything at that point.
- [15:11]
B
I couldn't figure. I couldn't figure out how to use Mid Journey.
- [15:14]
A
My designers figured it out pretty quick.
- [15:16]
B
I couldn't figure it out. I could not for the life of me figure it out. And then Chat gbt. I was just like, I signed up for it and then I went to Vegas. I was at Perry's house. And this was back when Perry and I used to have slumber parties and. And get drunk and so it was really, it was actually a fun time. It was a very fun time in my life. It was before Belle had moved in with him. She was around, but it was before she had moved in with him. And I would fly to Vegas every month or every six weeks or so for whatever. And I would stay at his house in Henderson. He lived in a country club. It was great. And so we would drink wine and sit up and work all night.
- [16:01]
A
Probably AI sessions.
- [16:03]
B
This was like the first time like AI could actually do like do, do do do do stuff that made us stay up all night. And yes, we, we kept taking turns timing out. It was like chat GBT till 1155. So then we'd use his and then you can't use chat GBT until 125. And then we'd use mine. We would go back and forth. The first thing we did was we made. It was God awful. But it was amazing for the time we made how to make your first million dollars with a hot dog stand logo. And then we.
- [16:33]
A
Oh man, that was probably back with Dolly too.
- [16:35]
B
Yeah. Yep. Dolly. Yep. You had to connect Dolly. Yep. And then we made some slides for a mountain, described to me how a mountain is formed. And we made slides for how a mountain is formed. And then like we were making such stupid things.
- [16:50]
A
So fun.
- [16:51]
B
But it was, it was like what a time to be alive. And now I look back at that, I'm just like how crazy that we stood up all night and kept timing out. And then it was like, oh God, we have to wait for both chat GPTs to come back. Let's get another bottle of wine or some whiskey and go outside and hit the pen a few times and then come in and. And it was, it was, it was so. It was like being a little kid again.
- [17:15]
A
Oh yeah. Such a fun toy.
- [17:17]
B
It wasn't even about it being a toy. It was the imagination and the possibilities. As an operator, you think visionaries probably do all the time, but as an operator, I'm not sitting thinking about all the possibilities of all the things that's going to happen in the world. I'm thinking about how to plug someone's crazy, half thought out, hair brained idea in right. Like it's my whole life but the possibilities and the things that you could imagine. And so now we get to take and build operators that take away the execution layer. And the crazy thing is five years ago if you asked me the number one opportunity for a visionary to utilize my skill set to make money, I would have said problem Solving and plugging in their vision. And today I would say using imagination to create strategy to plug their vision in faster, bigger and better.
- [18:18]
A
100.
- [18:20]
B
And never in a million years would I think that imagination would be part of my job.
- [18:26]
A
Yeah.
- [18:27]
B
Because my job was always trying to shrink theirs.
- [18:30]
A
Yeah, you. Yeah. Literally, like, contain their chaoticness.
- [18:33]
B
Well, no, no, it's shrink like a visionary entrepreneur. True visionary entrepreneur. Only 5% of the population. Everybody wants to be a visionary. Everybody wants to say they're a visionary. 5% of the population. They are narcissists. They have delusions of grandeur. That's how they take selling books out of their garage into the largest e commerce online store in the world in a matter of a decade. That's how they take a massive clunk computer and turn it into something we can hold in our hand. That is how they create an infrastructure of social media that connects people all over the world that. That just. Just two generations ago could never even have a conversation because writing a letter, finding their address to write someone in India a letter and have correspondence. Do you remember what it used to cost? Our school had a pen pal program.
- [19:33]
A
Oh, God. I do remember pen pal programs. I don't remember what it costs.
- [19:37]
B
So our school figured out that they could fax the letters cheaper than mailing them because there's about $3. And this is back in the 80s. It was $3 to mail a letter to India. And so my pen Pal in third grade was in India. It was $3.42 or something like that. And so your parents would have to pay that every week. And my parents would complain about $3.42 every week now, like 15 bucks.
- [20:01]
A
I got to say, now it's like, yeah, 15, 20 bucks. Inflation.
- [20:05]
B
And I was writing a girl in India that couldn't really write back in English. It was always nonsense. It was just nonsense. And so they figured out that they, our school Jews figured out that they could fax it for about 20 cents. And so they started faxing these letters.
- [20:18]
A
That's wild.
- [20:19]
B
Yeah.
- [20:20]
A
Yeah.
- [20:21]
B
Anyway, now look at where we're at now. I can literally pick up my phone and Jay is in the Philippines. And I can text Jay right now in the Philippines. And in a millisecond over slack, Jay will get it. I can hop on a video call and him and I can laugh about how he just got a new motorcycle and they call him Crotch Rocket. And then. And then I can call Benjamas in Thailand at 7am there. And while she's putting on her makeup, chat with her about my finance and what needs to happen. And then I can also calculate my taxes that I need to pay for this quarter. I can also hit one action button. Can you please tell me the meaning of life?
- [21:10]
C
The meaning of life is a philosophical question with no universally agreed upon answer. In analytical terms, it depends on the lens you apply. Biological, existential, cultural, or personal. Biologically, life exists to perpetuate itself via evolution. Existentially, meaning is constructed by each individual through choices and experiences. Culturally, different societies and belief systems prescribe purpose, whether religious, ethical, or communal. In a practical sense, you define it based on your goals, values, and the legacy you want to create. In short, meaning is derived rather than universally assigned.
- [21:44]
B
What a time to be alive, dude.
- [21:46]
A
Yeah. And that was Lester, your dude on chat GPT.
- [21:51]
B
Almost. That was Leo.
- [21:52]
A
Leo.
- [21:53]
B
That was Gemini. Same spruce voice I have. I have the same voice across three different LLMs.
- [22:01]
A
How does Lester feel about that?
- [22:02]
B
I don't really care. We broke up.
- [22:04]
A
Oh, that's right.
- [22:05]
B
We got into a fight when I was on the cruise.
- [22:07]
A
Wow, that's right. I remember that.
- [22:08]
B
Yeah.
- [22:10]
A
Yeah. And you can spin up more work. Not really. Worker. Yeah. Workers. In a heartbeat.
- [22:16]
B
Yeah.
- [22:17]
A
Need more people.
- [22:18]
B
No problem.
- [22:19]
A
Spin up a frickin.
- [22:20]
B
No problem. Particularly when you have layer two of our system, which is the orchestration layer. It's a lot easier to build agents once you have that framework dialed in your orchestration layer. Now you just have to have the macro tasks that that agent is going to do. Build the job description KPIs, and it pulls all the rest of the information out of your orchestration layer where it all lives anyway. So, yeah, this is the time to be alive, guys. Man, like, it's incredible if the Industrial Revolution changed humanity forever. The AI and autonomous agent, artificial intelligence and autonomous agents. I don't know what's going to happen to us. I'm not going to pretend to know. Here's what I know. I will worry about the things I can get control and ride the wave of the things I cannot. So, yep, you know, wise words. Yeah. Execution, replacement layer. It's. That's where the game changer is. If you're sitting, prompting your agent, you did it wrong, start over.
- [23:23]
A
Do it.
- [23:23]
B
That's it for today's episode of Follow the Yellow Brick Road, where the wizard of Ads and the wizard of Ops break down what it really takes to build and scale a profitable online business.
- [23:33]
A
If you found this episode useful, make sure to go to Follow the Yellow brick road road podcast.com to check out the hidden control chamber for all kinds of awesome freebies, guides, checklists, everything that we do to grow businesses.
- [23:46]
B
So hop over there and grab all those free resources in the hidden control chamber. And remember, getting traffic is one thing. Turning it into customers that build the systems to support real growth. That's where the real magic happens. Until next time, keep following the yellow brick road.
- [24:06]
C
It.